nN 1 poe oe serel ere oe ee all K Vehicle Pes de ide} Ae Ae : i ey Vv ry Ma we! das “\ ~~ LEAFLETS OF WESTERN BOTANY VoLuME V SAN FRANCISCO, CALIFORNIA 1947-1949 ‘Owned and Published by ALICE EASTWOOD AND JOHN THOMAS HOWELL Printed by The James H. Barry Company San Francisco (Pages 1-72) James J. Gillick & Company Berkeley (Pages 73-208) VoL. V No. 1 LEAFLETS of WESTERN BOTANY CONTENTS PAGE Pugillus Astrogalorum VI: Noteson Section Drabelle . . 1 R. C. BARNEBY A Collection of Plants from the Aleutian Islands . . . 9 Aice Eastwoop Primers on). titlochin concinna:) © ee as | eS JoHN THomas Howe. A New Monocephalous Parthenium . . . . . . 19 R. C. BARNEBY mene of the Genus Malvastrum: 9. os ee 28 Tuomas H. KEARNEY SAN FRANCISCO, CALIFORNIA JANuArRY 31, 1947 LEAFLETS of WESTERN BOTANY A publication on the exotic flora of California and on the native flora of western North America, appearing about four times each year. Subscription price, $1.00 annually; single numbers, 40c. Address: John Thomas Howell, California Academy of Sciences, Golden Gate Park, San Francisco, California. Cited as LEAFL. WEsT. Bort. RU A pHi} INCHES eT ame ae pau aa a 4 GG ed eH ke Owned and published by Auice Eastwoop and Joun THomas HowkELy PUGILLUS ASTRAGALORUM V: NOTES ON SECTION DRABELLA& BY R. C. BARNEBY Wappingers Falls, New York The following paper, like its predecessors in a series devoted to Astragalus, is the result of an attempt on my part to under- stand, and to place in their appropriate niche in the genus, plants collected and studied in the field. As Mr. Ripley and I continue to amass material, new problems rise to the surface, demanding solutions which can often be arrived at only after considerable research. In the preparation of these notes, comparisons have been made with types and other pertinent exsiccata at the Gray Herbarium (G), the University of Minnesota (M), the New York Botanical Garden (NY), the Philadelphia Academy of Natural Sciences (P), Pomona College (PO), the University of Idaho, Pocatello (Poc), and the National Herbarium (US) ; the symbols in parentheses used throughout the paper will be an index of my indebtedness to those who have cheerfully ad- vanced my studies. All collections by Ripley and Barneby re- ferred to below are represented in the herbarium of the California Academy of Sciences. > f 7 7 A The four leading American astragalogues of the past hundred years, in preparing their varied accounts of the genus, have each devised, both as an expression of natural relationships and as a means of subsequent reference, a framework of groups or sections composed of species fundamentally alike. Each built to some extent upon his predecessors or contemporaries, Sheldon and Jones on Gray, Rydberg on Gray and Jones. It is curious, how- ever, that in no case was agreement reached in the nomenclature (though very frequently in the substance) of their infrageneric groups. In this respect each ignored or paraphrased the others, so that today, if one wishes to refer a species to such and such an established group (a most useful device in so vast a genus as this), one is in many cases at a loss for the appropriate name. For example, a species of Torrey and Gray’s Homalobus § Dra- belle and another of their Phaca § Condensati both belong to eo Homalobus § Cespitosi Rydb., to Homalobus § Simplicifolii Leaflets of Western Botany, Vol. V, pp. 1-24, January 31, 1947. 2 LEAFLETS OF WESTERN BOTANY [VOoL. Vi NOs Rydb., to Astragalus § Homalobi Jones and to Astragalus ser. Phaca § Homalobus Sheld. Some clarification is needed, a general revision of sectional groupings which will establish the typical species to which each name properly belongs, which will define the scope of that section, and bring to Astragalus an order which has long reigned in, for example, Carex, the only genus of comparable size and diversity in temperate North America. — The meaning of Carex § Inflate is fixed and comprehensible ; Astragalus subgenus Hamosa means, at present, next to nothing at all. In an earlier paper (Proc. Calif. Acad. Sci. IV, 25: 166— 1944) I made some attempt to define and name three small western species-groups. At that time it seemed desirable to bring the nomenclature of the American species into line with that of Bunge’s classification of Old World Astragalus. Sooner or later a compromise must be made, particularly since several groups are common to Asia and North America. It has become clear, however, that Bunge’s elaborate hierarchy of subgenera and sec- tions is not only unsuited to our species, but results in a highly artificial classification for the Eurasian. The sectional names proposed by Bunge are formed as substantives, e. g. Chetodon (quite different, by the way, from § Chetodontes Gray), Cystium (but not Cystium of Rydberg) or Euodmus (precisely the Ameri- can § Uliginosi) ; whereas those of Gray, Jones, and Rydberg are formed as adjectival plurals, e. g. Eriocarpi, Campestres, etc. It seems ridiculous that the same group of plants should be known in Europe as Astragalus § Euodmus, and as Astragalus § Uligi- nosi across the Atlantic. Somewhere a reorganization must be " undertaken. If a homogeneous classification is to be applied it will be necessary to discard one set of sectional epithets. Since about an equal number of names has been proposed in each pattern, neither has an advantage on this score; either way the same dam- age will be done to the accepted nomenclature. Historically, however, the use of the species-group (§ Ciceroidei DC., 1802; § Eriocarpi Gray, 1864) is older than that of Bunge’s section (Caraganella, etc., 1869). Moreover, Asa Gray was the first to unite Phaca with Astragalus, the first to understand the limits of the genus, and the first to make out natural lines of relationship JANUARY, 1947] | PUGILLUS ASTRAGALORUM a within it. For further reasons which cannot be fully discussed here, I am convinced that Gray’s method of classification is the most realistic and convenient yet devised for Astragalus, and should be adhered to. Until a comprehensive review of the genus can be presented, it seems best, therefore, to adopt the system of Gray’s species-groups, to take up his sectional names wherever possible, and elsewhere the oldest (following accepted rules of priority) name originally proposed in Gray’s form. Following this principle I suggest the adoption of ASTRAGALUS § DRABELLZ Astragalus § Drabelle (T. & G.) Barneby, stat. nov. Homalobus § Drabelle T. & G., Fl. N. Amer. 1: 352 (1838), based on Homalobus cespi- tosus, H. brachycarpus, and H. canescens Nutt., the first (Astragalus spatu- latus Sheld.) being typical. Astragalus ser. Phaca § Condensati Gray, Proc. Amer. Acad. 6: 231 (1864), based on Phaca simplicifolia Nutt. Homalobus § Cespitosi Rydb., Bull. Torr. Club 50: 178 (1923), based on a combination of the two preceding, H. cespitosus being typical. Homalobus § Simplicifolii Rydb., N. Amer. Fl. 24: 259 (1929), the equivalent of the preceding. Cespitose or pulvinate perennials, acaulescent or nearly so; pubescence dolabriform; stipules scarious, connate, ocreiform, embracing the petiole ; leaves dimorphic, the lower reduced to short phyllodia, the upper similar but longer and narrower or with 1 or 2 pairs of decurrent or obscurely articulate leaflets ; calyx campanulate, ebracteolate, not ruptured; pod ses- sile, erect or nearly so, more or less compressed, 1-locular. Apart from the rare and distinctive A. detritalis, the species of § Drabelle have been widely misunderstood and, within the space of a few years, variously interpreted. Macbride (Cont. Gray Herb. 65: 29,—1922) and Jones (Rev. Astrag. 78,—1923) agreed in maintaining from among the several species previously described only the congested A. simplicifolius and a looser variety cespitosus. Rydberg almost simultaneously (Bull. Torr. Club 50: 178, sequ..—1923) was recognizing six species, these being reduced to five in the North American Flora (1929). The dis- covery of a striking novelty in the group has led me to examine once again the conflicting treatments and complex synonymy of the Drabelle, a revised view of which is presented in the follow- ing key and remarks. 4 LEAFLETS OF WESTERN BOTANY [ VOL. V, NO. I Key to the species of § Drabelle 1. Corolla small, the banner 5—9 mm. long, the keel 4—5.5 mm. long or, if up to 7 mm., the racemes many-flowered, 5—18 cm. long in fruit (2). 2. Raceme 1—10-flowered, 3 cm. long or less in fruit; upper leaves (or phyllodia) not exceeding 5 cm. in length; calyx-teeth 1—2.5 mm. long, erect; pod strigose (3). 3. Plant loosely cespitose; raceme 3—10-flowered, commonly exceeding the leaves..................-- la. A. spatulatus var. typicus 3. Plant pulvinate ; raceme 1—2-flowered, shorter than the leaves DANGER AIG SE AT AORN RI ire of lb. A. spatulatus var. uniflorus 2. Raceme 8—17-flowered, 5—18 cm. long in fruit; upper leaves at least 10 cm. long; calyx-teeth 3—4 mm. long, stellately spreading > pod labrous.-2-. e 2. A. chloddes 1. Corolla larger, the banner 10—15 mm. long, the keel 7.5—11 mm. long; raceme 1—5-flowered, not exceeding 1.5 cm. in fruit (4). 4. Keel 7.5—8 mm. long; all leaves simple and reduced to phyllodia ; pod rigid, glabrous, obliquely ovoid-oblong, turgid, 9—15 mm. long and 3.5—5 mm. in greatest diameter........ 3. A. simplicifolius 4. Keel 10—11 mm. long; upper leaves 3—5-foliolate ; pod leathery but scarcely rigid, strigose, strongly compressed, linear in outline, 15—31 mm. long, 2—3.5 mm. in greatest diameter a Ni AC UE ea Ramat Bo ARC oP 2 Tah td leh: Gn eRe 4. A. detritalis la. ASTRAGALUS SPATULATUS Sheld. var. TypIcus Barneby Homalobus cespitosus Nutt. ex T. & G., Fl. N. Amer. 1: 352 (1838). Astragalus cespitosus (Nutt.) Gray, Proc. Amer. Acad. 6: 230 (1864), non Pallas, 1800. A. simplicifolius var. cespitosus (Nutt.) Jones, Proc. Calif. Acad. Sci. II, 5: 647 (1895). A. spatulatus Sheld., Minn. Bot. Stud. 1: 22, 119 (1894). A. simplicifolius var. spatulatus (Sheld.) Jones, Cont. W. Bot. 10:65 (1902). Homalobus canescens Nutt. ex T. & G., l. c., non A. canescens DC., 1802. Homalobus brachycarpus Nutt. ex T. & G., |. c., non A. brachycarpus MB., 1809. Astragalus simplex Tidestr., Cont. U. S. Nat. Herb: 25: 330 (1925). A. spatulatus var. simplex Tidestr., Proc. Biol. Soc. Wash. 50: 20 (1937). Sheldon’s name was based on Homalobus cespitosus Nutt. and both H. canescens and H. brachycarpus were reduced to it. Gray had earlier stated (Proc. Amer. Acad. 6: 230,—1864) that Nuttall’s three Homalobi were “evidently reducible to one’”’—an opinion accepted by all but Rydberg and Tidestrom—and, since all of Nuttall’s epithets were long preoccupied in Astragalus, Sheldon’s proposition was clearly justified. Both Jones and Mac- bride subordinated the entity so formed to the related but very different A. simplicifolius, having overlooked the limits and true characters of that species. On the other hand, Rydberg revived JANUARY, 1947] | PUGILLUS ASTRAGALORUM 5 H. brachycarpus, maintaining it as distinct from H. cespitosus on account of its shorter pod and supposedly broader phyl- lodia. Following this opinion Tidestrom has lately transferred H. brachycarpus to Astragalus as A. simplex, and subsequently A. spatulatus var. simplex. In its wide range, extending from Saskatchewan to northern Colorado, northern Utah and southeastern Idaho, A. spatulatus encounters diverse climates and soils, and is accordingly variable. Variation affects especially the length of the leaf (which may be simple or trifoliolate), peduncle, vexillum, and pod. There is a definite tendency for the latter to become shorter and broader, and hence more abruptly acute, in the dry climate of the south and west; but this finds no correlation with other characters, and but little with distribution, so that H. brachycarpus, based on a plant with short pod (Hills of the Platte, Nuttall, P!) is to be considered a very minor entity. For those who wish to maintain it, the combination A. spatulatus var. simplex is available. The form with relatively elongate banner is occasional throughout the range of the species, and corresponds with the negligible H. canescens (type, Rocky Mts., Nuttall, P!). Asarule A. spatu- latus is loosely cespitose, but on the high deserts of southern Wyoming it assumes a reduced and truly pulvinate aspect, with very short 1—3-flowered peduncles scarcely emerging from a dense hemisphere of tiny phyllodia. While this intergrades with the var. typicus, and probably represents no more than an extreme response to conditions of wind and drought, it is a striking form, which may be maintained as lb. ASTRAGALUS SPATULATUS Sheld. var. UNIFLORUS (Rydb.) Barneby Astragalus spatulatus var. uniflorus (Rydb.) Barneby, stat. nov. Homalobus uniflorus Rydb., Bull. Torr. Club 34:49 (1917). Astragalus simplicifolius auct. plur., pro parte, non Gray. The plant treated here has caused much confusion. In Wyo- ming there exist two species of Drabelle superficially similar in their pulvinate life-form and few-flowered, shortly pedunculate racemes: the present var. uniflorus, with small flower, short, somewhat turbinately campanulate calyx which is unchanged in fruit, and strigose, relatively thin-walled pod which remains later- ally flattened to maturity ; and another, with much larger corolla, broadly campanulate calyx which becomes glabrate, stramineous 6 LEAFLETS OF WESTERN BOTANY [ VOL. Vi NOod and conspicuously nerved in fruit, and a glabrous, at length turgid pod of much thicker texture—the true A. simplicifolius. These two were confounded by Jones and Macbride who, thinking there was but a single plant involved, and finding themselves unable to separate the uniflorus element from spatulatus proper, con- cluded that A. spatulatus and A. simplictfolius were simply two aspects of single species. The subsidiary character of long bracts present in the type of H. uniflorus (Evanston, A. Nelson No. 2971, NY!) does not hold in most recent collections; in fact, we have only one col- lection (Point of Rocks, Sweetwater Co., Wyo., Ripley & Barneby No. 7916) which precisely matches the type in this re- spect. For the most part the bracts do not differ from those in A, spatulatus var. typicus. The variety is known to me only from the southern tier of counties in Wyoming, from the Laramie Plains westward to Evanston. 2. ASTRAGALUS CHLOODES Barneby Astragalus chloddes Barneby, sp. nov., inter Drabellas, a quibus foliis superioribus pedunculisque valde elongatis, racemis multifloris laxis, denti- bus calycinis stellatim patulis, aliisque notulis insigniter differt, collocanda. Herba czspitosa subacaulescens, preter legumen pilis filiformibus medio- fixis arcte appressis undique strigoso-canescens; caulibus numerosis in summa radice verticali lignea multicipiti congestis, annotinis brevissimis 0.5—1.5 cm. longis, demum ut caudicis ramuli stipulis petiolisque marcidis crebre induti persistentibus; stipulis imbricatis ovatis acutis 2—3-nerviis, petiolum adversus in ocream scariosam apice 2-dentatam concretis, iis foli- orum inferiorum etiam connatis, cupuliformibus, circa 2 mm. longis, in superiores hinc inter se liberas 7—8 mm. longas abeuntibus ; foliis omnibus simplicibus in phyllodio reductis dimorphis, inferioribus (ad nodos steriles) patulis lineari-oblanceolatis acutis 1.5—7 cm. longis, superne in laminam plus minusve involutam (1) 1.5—2 mm. latam expansis, superioribus (pedunculum suffulcrantibus) erectis angustissime linearibus 10—15 (17) cm. longis, 0.4—0.9 mm. latis subteretibus sed ventraliter sulcatis, rigidis, spinula pallida subvulneranti mucronatis; pedunculis strictis rigidis 4—9 cm. longis, in racemum haud flexuosum (8) 10—17-florum jam ad anthesin valde laxum, fructiferum (5) 8—18 cm. longum abeuntibus ; bracteis scario- sis ovatis acutis circa 3 mm. longis basi pedicellum suberectum dimido breviorem amplectentibus ; calycis tubo campanulato 5-nervio ebracteolato haud rupto 2.5 mm. longo, dentibus stellatim patulis subulato-setaceis acutis- simis 3—4 mm. longis, sinu obtuso inter se separatis; corolla saturate vio- lacea multistriata; vexilli 8 mm. longi supra unguiculum per angulum rectum retroarcuati lamina orbiculari 7 mm. lata apice emarginata; alis subequilongis, unguiculo 3 mm. longo, lamina lunato-oblonga obtusa 2.5 mm. lata auriculo incluso 6 mm. longa; carine 7 mm. longe laminis triangulari-lunatis vix 3 mm. latis, marginibus superioribus fere rectis, JANUARY, 1947] | PUGILLUS ASTRAGALORUM 7 inferioribus (concretis) per 100° in apicem acutiusculum incurvis; ovario glabfo, 5—8-ovulato; legumine strictim erecto sessili oblique oblongo vel lanceolato-oblongo, basin versus leviter arcuato vel subrecto, basi obtuso apice in stylum persistentem hamatum vel recurvum acuminato, 8—11 mm. longo, 2—3 mm. lato, de latere valde compresso, 1-loculari, suturis (ventrali crassiori) acutis, valvulis coriaceis glabris nitidis obscure reticulatis viridi- bus vel interdum minute purpureo-guttulatis; seminibus (immaturis) ob- longis, circa 2 mm. longis. Uran: in dense tufts on ledges or in sandy pockets of white sandstone cliffs, 6 miles southeast of Jensen, Uintah Co., alt. 5100 ft., 7 June 1946, fl. & fr., Ripley & Barneby No. 7797. Type in Herb. Calif. Acad. Sci. No. 332305. North of Dinosaur National Monument, Uintah Co., alt. 5500 ft., E. H. Graham No. 7718 (G, US). Dimorphism of the leaves (or phyllodia) is characteristic of § Drabelle, but is nowhere as marked as in A. chloddes. The lower leaves in the year’s cycle, those which carry on photo- synthesis during the resting period, are no longer or narrower than in forms of A. spatulatus. But the later foliage (particu- larly the two or three leaves which accompany and subtend the peduncles) is extraordinarily drawn out into acicular, rigid and pungent petioles between 1 and 2 dm. high. With its similarly elongated, lax and many-flowered racemes, glabrous pod and relatively long, stellately spreading calyx-teeth, A. chloddes can scarcely be confused even with its close allies, certainly with no other described species. Nevertheless, Graham’s flowering ma- terial was identified by Tidestrom and cited (Ann. Carn. Mus. 26 : 249,—1937) as A. moencoppensis Jones, a totally unrelated species with developed nodes, regularly pinnate leaves, and pubes- cent, dorsiventrally compressed legume. 3. ASTRAGALUS SIMPLICIFOLIUS (Nutt.) Gray Astragalus simplicifolius (Nutt.) Gray, Proc. Amer. Acad. 6: 231 (1864). Phaca simplicifolia Nutt. ex T. & G., Fl. N. Amer. 1: 350 (1838). ’ Homalobus simplicifolius (Nutt.) Rydb., Bull. Torr. Club 40:52 (1913). Astragalus lingulatus Sheld., Minn. Bot. Stud. 1: 118 (1894). Homa- lobus lingulatus (Sheld.) Rydb., Bull. Torr. Club 40:52 (1913). Astragalus exilifoliuns A. Nels., Bull. Torr. Club 26: 10 (1899). Homa- lobus exilifolius (A. Nels.) Rydb., op. cit. 40:52 (1913). Astragalus exili- formis (sphalm.) Ind. Kew., Suppl. 2: 20 (1904). This species was first referred by Nuttall to Phaca, for the good reason that the pod did not conform with the pattern of Homalobus as he had defined that genus. Torrey and Gray re- 8 LEAFLETS OF WESTERN BOTANY __[VOL. V, NO. I marked on its habital resemblance to “several species of Mr. Nuttall’s genus Homalobus,’ and Gray, when he came to sub- merge both the latter and Phaca in Astragalus proper, placed — A. (Phaca) simplicifolius and A. (Homalobus) cespitosus side by side. Much later Rydberg revived Homalobus, and moved the species into his section Cespitosi (where it doubtless be- longs), and the peculiar fruit has since been largely overlooked. Nuttall’s collection from the sources of the Platte (G, NY, P) is very mature and probably depauperate as well; no complete flowers remain and only a single pod (P) is extant. Rydberg (N. Amer. FI. 24:260,—1929) described the keel petals of Homalobus simplicifolius as “5 mm. long,” but a keel lodged among the leaves of the specimen at the Philadelphia Academy measures 7.5 mm. in length, and the corolla must have been con- siderably larger than generally supposed—in fact, altogether similar to that of A. exilifolius and A. lingulatus. It may be claimed with some assurance that the type-collection of A. exili- folius (Freezeout Hills, Wyoming, E. Nelson No. 4493, US!) represents no more than well-developed A. simplicifolius. It differs slightly in the greener, somewhat narrower phyllodia and larger, mottled pod; but in other respects, and especially in the glabrate, prominently nerved calyx and in the size of the corolla, the two are identical. Of A. lingulatus, based on two collections, in flower only, obtained by Hayden at the foot of the Big Horn Mountains and on the North Platte a little above Casper, I have seen no authentic material, but the description agrees very closely with A. simplicifolius, and Rydberg, who probably saw the type (in the Missouri Botanical Garden) eventually considered it con- specific with A. exilifolius. In amalgamating A. simplictfolius, exilifolius, and lingulatus I follow Macbride; but his reduction of Homalobus uniflorus to the same and his citation as A. simplici- folius of material properly referred to it, have left only confusion behind them. The corolla of A. simplicifolius is consistently described as ochroleucous, and the color was used by Rydberg as a key- character in separating the species, along with A. lingulatus, from its relatives. Proof that the petals are ochroleucous is, however, wholly lacking. Nuttall, as inferred above, probably saw no fresh flowers. Sheldon stated of his A. lingulatus: “corolla probably ochroleucous or purplish, the color not preserved in the type JANUARY, 1947] PLANTS FROM ALEUTIAN ISLANDS 9 specimen.” Whereas in the type of A. evilifolius, even after a lapse of fifty years, there are evident traces of purple remaining. Moreover, in specimens lately gathered in Carbon County, Wyoming (10 miles northwest of Rawlins, Ripley & Barneby No. 7744) the fresh color was of a lively purple. Until evidence to the contrary is forthcoming it may be assumed that the flowers are purple as in the rest of the section. Astragalus simplicifolius is an uncommon species, known to me with certainty from a limited area of Wyoming and extreme northern Colorado only. Rydberg and Macbride both cited a collection (Brandegee in 1877) from Cafion City, Colorado, far south of the established limit for the species. The only part of this collection seen (NY) is sterile, and I hesitate to confirm the determination. The apparently isolated station is, however, not improbable, since the shale bluffs along the Arkansas River pro- vide a suitable habitat, and are known to support a peculiar flora of, at least in part, Wyoming affinities. Material from Utah re- ferred here by Macbride belongs to the form of A. spatulatus var. typicus with short pod prevalent along the foothills of the Uintahs. 4. ASTRAGALUS DETRITALIS Jones Astragalus detritalis Jones, Cont. W. Bot. 13:9 (1910). Homalobus detritalis (Jones) Rydb., Fl. Rocky Mts. 1063 (1918). This very local species has been known only from northeastern Utah in the counties of Duchesne (above Theodore, Jones, PO, type!; west of Duchesne, Ripley & Barneby No. 4669) and Uintah (cf. Graham, Ann. Carn. Mus. 26:247,—1937). An easterly extension of range into the valley of the White River, Colorado, phytogeographically part of the Navajo Basin, may now be reported: 14 miles east of Rangely, Rio Blanco Co., Ripley & Barneby No. 7786. A COLLECTION OF PLANTS FROM THE ALEUTIAN ISLANDS BY ALICE EASTWOOD This collection was made by Harry Chandler in 1945 and pre- sented to the California Academy of Sciences. He is one of the fortunate men who enlisted in the late war with an interest in na- tural scienceand while on duty in the navy off the Aleutian Islands eagerly seized every opportunity when he had shore leave to col- 10 LEAFLETS OF WESTERN BOTANY [ VOL. V, NO. I lect the insects and plants. His chief interest was entomology but he couldn’t resist the fascinating arctic plants. His collections, made on Attu, Adak and Great Sitkin islands, are represented by 135 species that are beautifully pressed so that the colors are preserved. Instead of a separate list from each island I have con- solidated them in a table. None was collected not known before from the Aleutians but, according to the lists in Hultén’s Flora of the Aleutian Islands, he added several species to Attu and Adak indicated in the list by an asterisk in each instance. This seems to be the first recorded collection from Great Sitkin, a small rocky islet the summit of which is the loftiest elevation in the Aleutians. Besides the invaluable aid from Hultén’s Flora, I am indebted to Agnes Chase for identifying the grasses, to John Thomas Howell for the carices, to Carlotta C. Hall for verifying the ferns, and to Donovan S. Correll for verifying the orchids. a at > si = 5 oR ate Athy nitim Edlix-femiinad (is) x Ob lao orss ee ke eee —- * — Cystopteristiragilis: (Gis. Berth ei ia.). ene pe Sa eee ee). Dryopteris oreopteris (Ehrh.) Maxon ......0.......-....-1s-cceeceee--- —- * — ‘Dryopteris Phegopteris (L.) C. Christ. -.22...2...0....c.-ce-eeesenes —- xX = Eiqtunsetumbearvensey Wiis 5 eee le gage ey Ree ee —- xX =— Eyeopodnimialpinumel:.: 2b. SAS sup ind Uae Ay seetet oA Ore ere x x — HEVEGPOC tas ANIMOLINUIT Hod ees eee eet ee ee xX —- = iy copodituna clavatuin [js ae eee ela eee 20 toe ee ee a x — =— leycopodium’ Selago pg 2k ae ahs Beer Pe UPS x Pd See ep x x Spateanium hyperboreum Maes. ee Xx — = Avsrostissalaskaina: Ellie 20 wees oe We yee ene Ja —- — ANETOSUS DOKeASelantm. sce eet we ela Ne ee i INS TOSUS CXATeata’ Lrinics sy comes PCN. Na a tien pees Ww Ce aot es PMOPEGUnUS? equals -SODO!, cna meee ta Ion) ih, ENE — — Bromusialeutensis ck tiny sen acento eeu SE etene soe — ae Calamagrostis nutkaensis (Presl) Steud... —- — xX Deschampsia atropurpurea (Wahl.) Scheele _ Vahlodea atropurpurea (Wahl.) Fries..........-.....-.-------- xX —- =— Elymus mollis Trin. E, arenarius L. subsp. mollis (Trin.) Hult. ................. xX — = Besties wibreat bse tess A La ae ee Nica ena ye Hordeum ‘brachyanthemmim Nash 2.250 scccteta cc cceneefoeceeee ee cntare ct —- * — RE Fa desustnas eh uence ssa oo, Crees tM eae LE cafe Ki. Poe arcticat RisBre, jth b.25.0 sik Soto k aera artes Rae Meares eee xX = =x JANUARY, 1947] PLANTS FROM ALEUTIAN ISLANDS = Peaatcties i ior, a yiviparous: form sii ke — Dirisetum, spicata (L;) Richt, ss scae Sno tatoos ee x Sem ateaOsant hea EPCS) oe one sa ten eet gteancelec uk ee pd aves circinnata C.. A. Meyer 00 ene a ee x Maree vue yey trOTmend, 2... se ole es | er — mares macroctisets CoA. Meyer 230 .ili iced beth | abet « SOVeeeSOMallay ETO is). tne ete Wey ee, Oe x Dpmerereircna Mirteg, se ee Ps ae a ee — EE) CTE SUR G81 Fc inane Re se Se ne PY 8 4 Luzula multiflora (Retz.) Lejeune ......20.0....22c.c. cece —— Pigula parviflora (Ehrh:), Desy) 20.00.02 — Diarra WV aislenber eit Rp soca ee hicbee cust tae — Fritillaria camschatcensis (L.) Ker-Gawl. ............2.2.2.0:00- x Srrepopus amplextolus CL,) DG, oc x Momeldia‘coceinea, Richards,.-.4. ot ee x Maianthemum dilatatum (Wood) Nels..éMacbr. "42.23 — (TET: gl 2.2) | SESS PR EERE as le RR x Popped puttatum :Swattz: «(04 .0i0 000 eee x Habenaria behringiana (Rydb.) Ames .00.0...02020-0.0.0.0.sseceeeeeeeeee x Ppenariat borisiana, Chain, 2. :.20..-b- ssn. eh de ee x Pabenama tilatata (Pursh)’ Hook. 3.8.-...000 oak ae x Hlabenaria hyperborea (L.). R.Br. i. .ocseceeoestellscccqeteeeee x Listera convallarioides (Sw,) Tort, .......-::---.-cctes--sncocssuessssenen * eet errata. (1 ,.)>. 45. BE; ls ro cce e me (UIE vie Os a OD a cr ce ae cc a Baar SONI x emmmeurerea Chiaiay | ine Se ed eS nn. AK os SECTS OOF ip: 1 eB EE eee — CEU TT US By As | ee a be cw — Reamemdienestratis Greene 2.2004. ks ee x meetin sini CE.) FLowell 2 oc. sn. accu eed ed x Cerastium beeringianuam, C, & Sy. os o.-ci::2.e.csoncsccestenspoesenstecchbens — Prionckeya peploides (1...) bar bn o.oo... saceoccnatmatnanencetcen 4 Stellatia lommipes. Goldie i naas.cecdosccesnsela2-sseesnvsaerpebathessigscsatlseseien * Pear aee AayitTaar Wists BOLE. 3. cxicencsk os esndecew-ceccce ee ee cS umenraecitiy atin settrimtn ald, gio 2 danas, suaapcen eo deemeorieaae ee — Semernione. Natcisstiota Tie. Nik ee ee ee x meena Arita Salish: sok kee ee ae x BeTIMCIINAS ACTISHNS. cot he Ss Ee x Ranunculus Eschscholtzii Schlecht. ...0.......0..-..c.ecccceseeeeeteeeeeeees 24 Ranunculus Flammula L. war.eealts: (Bagel) L.. Benson. :.:5..-s<--.m sstgaetetep lata “= Ranunculus occidentalis Nutt. war, Delsonis (DC: ):. 1 Benson’: ... ka ee — Wrabis lyrata LLi........2. CO LAR ie Cra Pe Te ON ee Desyie eee * Phytiodoce alentica GSprene.)yAcwbellerd 2. 2 ak ee x Rhododendron kamtschaticum Pall. -..W.2-2..22.2..2.-2222::10e0-ceee--=- x Weacciiitin; VatiS ileal ba: = a. ener, eee See gee me Oe x Peimula cunetolianiwedeD, 2.3.5 ets bee ae en ee eS Ae x Primula cuneifolia var. saxifragifolia (Lehm.) Hult. -........... x MEAcHEAlis europea) dod ccs. 7 ee ee x Mértensia mantimaa(les)9. BaGtay: ese eee x Euphrasia mollis (Ledeb.)’ Metts. -..2...0020:2-02t tee a Lagotis glauca 1G g 20 epee eB nd a Re OI Ds en Bs Oe Dd x V, NO. I HePY le seoal sl ogewge de cle taeale. lA oe P| excimer | oe | x bel eX beh | xx UpHTS FeRoIH | | fedex! JANUARY, 1947] TRIGLOCHIN CONCINNA 13 = > Z Peet arts verticilidatdy Bas i~ o-oo et ee i * — — Rhinanthus groenlandica Chabert -.............2..22.--2--0.-c00eceeeeeeeeeoee Bt a im) oe Weronica americana Schwein, -..222-022.2..cccco Ss cecesecveceeceee eneeee x —- — Veronica Stelleri Pall. var. glabrescens Hult. -..0020202000000.... x —- — Beeronice mitninisa Dickson 2... ae x —- — Peers ercanciiorgs Gaertn, 2... 2k 8) ee xXx — — aT eTETWIN ALIS Diyos a acne ee ee ge at Kae iE MERICARI ORT C TEAC AC IAs ye | ae a ey ee 2S ME Fe Pein, kamitschaticiing Steller <2. a * — — Peels eaavanitia Btep. oe x —- — ME -BAnIOCAS Ta, Cttisn, <2 oe ee x — =— Campanula lasiocarpa var. latisepala Hult. 0222 .. —- xX — RLeARDOLCANS BOND: - 22 oe A x —- — . Anaphalis margaritacea (L.) B. & H. mrareOcineiitdis GrlCChe sg oe — — xX wuemarca dunes (1-)° Gaertn. *.- 2 x —- — mmnnngtasrtuida” 1, Nels) 05 a a Buereen en araSCukersts Iess. 2.0 tpt Se ee a ADM se 0 ee Se ee ee Cc — Artemisia Tilesii Ledeb. subsp. SPAISENICHBIS' (EVESS 3) EUG, 2 sy cect ee x — — relicanricwinta, iG: 28 aha tes. xXx — — Peewnantnemmim. arcticum Lo. 22. cts cco sseee cence a fect kamtschaticum JLedeb. °.........5 a Beetriner fciste: Witla) s¢520 0 ee oes Se x xX =— Picris hieracioides L. var. kamtschatica (Ledeb.) Hult... xXx — — Saussurea subsinuata Ledeb. ? EE ga OEE SS cee Serer he —_- * — ecto palriatas .( Pall.): Ledeb. 2... KS — — femme maencont tied (pas. oo xX — — manaxacim triconolobum Dahlist. =.:..0.--- 2 —- xX — REMARKS ON TRIGLOCHIN CONCINNA BY JOHN THOMAS HOWELL On April 23, 1944, while botanizing in the salt marshes bordering San Francisco Bay below Escalle, I found for the first time the large and small 6-carpellate triglochins of central Cali- fornia growing next to each other. I had known them for a long time but always before, each had occupied a different place in the marshes, the small one, Triglochin concinna Burtt Davy, more 14 LEAFLETS OF WESTERN BOTANY __[VOL. V, NO. I closely associated with Salicornia, the large one, T. maritima L., generally higher and near the upper limit of the marshland. When I had seen the two separately, I had presumed that the observed variation in habit resulted from the difference in habitat, but the two very different plants growing together called for an investigation. The problem presented in the Escalle marsh was to determine whether the small plant was merely an inconse- quential form of the large plant with which it grew, a variety worthy of nominal recognition, or a distinct species. Davy’s _ species, which was originally described from the salt marshes of San Francisco Bay had been variously treated by botanists, so a comparison of the two plants as they grew together was promising. A. detailed comparison of the flowers and fruits of the two plants disclosed no essential difference, only very minor vari- ations in size being apparent. In characters of habit, stems, and leaves, however, such fundamental differences were observed that I concluded that two distinct species were involved. As the two plants grew together, they were marked by distinctive habits, T. concinna being lower, more slender and essentially elegant in all its parts, T. maritima being much coarser and several times taller. The two plants grew differently, too, T. concinna spread- ing widely in a relatively thin open way over the marsh flats, T. maritima producing relatively close dense tufts of leaves and scapes. This variation is due to a real divergence in the character of the rootstocks of the two, those of T. concinna developing as short slender bracteate stolons with discernible internodes, those of T. maritima being much contracted and thicker. In T. mari- tima the bases of the leaves become indurate-corky and persist for several seasons on the rootstock, but in T. concinna the bases . of old leaves are evanescent, although sometimes some of the fibro-vascular tissue may remain for a time. It was observed that the leaf-blades in T. maritima were strongly obcompressed, but in T. concinna were nearly terete. The ligules of the two were essentially different, those of T. maritima being simple and either entire, denticulate, or emarginate, but those of T. concinna being 2-parted to the base. Other minor differences were noted, but from the principal ones that have been described it seemed amply apparent that two species were involved and further obser- JANUARY, 1947] TRIGLOCHIN CONCINNA 15 vations in many marshes in the San Francisco Bay area have confirmed the studies made at Escalle. Next, the varietal name that has been considered a synonym of T. concinna in western United States was investigated. This name is T. maritima var. debilis M. E. Jones and was originally given to a plant of the desert interior collected by Jones at John- son, Kane County, Utah. A large suite of specimens of var. debilis from the Great Basin was found to agree with coastal forms of 7. concinna in the elongate rootstock and the 2-parted ligule, so undoubtedly the variety is referable to that species rather than to T. maritima. Since the desert plant differs, how- ever, in being more robust with taller scapes and in having the rootstock usually fibrous-coated, it may be properly regarded a distinct variety. Aside from the very definite and fundamental characters that distinguish T. concinna from T. maritima, the geographic distri- bution of the two is distinctive. Triglochin maritima, in various forms, ranges around the Northern Hemisphere in higher lati- tudes and in North America reaches a southern limit in New Jersey, Indiana, and New Mexico, while in California it extends to the San Bernardino Mts. in the interior, but along the coast it has not been found south of San Francisco Bay. Typical T. con- cinna, confined to coastal marshes and beaches, is known to range from middle Lower California northward to British Columbia. The completely disjunct var. debilis, which is particularly charac- teristic of the saline desert of the Basin region, ranges widely over a vast subcontinental area from the Dakotas westward to Oregon and southward to Colorado, Arizona, and eastern California. The names to be assigned to the triglochins were definite enough until it was found that 6-carpellate plants in South America also have 2-parted ligules. This discovery complicated the choice of name by introducing seven specific names by Philippi which antedate Davy’s T. concinna. To determine the relation- ship of the South American plants, I have had to depend chiefly on original descriptions and on photographs of the Philippi types which were prepared for me by Dr. Carlos Mufioz Pizzaro in Santiago de Chile at the request of Dr. T. H. Goodspeed of the University of California. From all the descriptions, it is clear 16 LEAFLETS OF WESTERN BOTANY [ VOL. V, NO. I enough that the South American plants have fruits with 3 fertile carpels like those of T. striata R. & P.2 or T. palustris L., and by Buchenau (Pflanzenreich IV. 14: 7—14,—1903) the Philippi names are referred to either one of those species except T. ataca- mensis which is given as a synonym of T. maritima “pr. pte.” As Buchenau (op. cit., p. 9) remarks, the original description of T. atacamensis indicates T. striata, so I do not believe that we may seriously regard the fact that specimens labeled and dis- tributed as T. atacamensis by Philippi have proved to be T. mari- tima (sensu Buchenau.) as well as T. striata. In Phacelia I have come across Philippi names similarly misapplied by himself. The only other species about which there may be a doubt is T. litorea Phil. (Anales Univ. Chile 43 :539,—1873) which was described as follows: 255. Triglochin litoreum, Ph—T. radice fibrosa; scapo lzevi, folia linea- ria equante ; vaginis apice biauriculatis, auriculis acutis ; fructibus globosis ; capsulis tribus dorso tricarinatis, fertilibus; alternantibus sterilibus, fili- formibus ; pedicellis capsulam zequantibus ; stigmatibus persistentibus. The description of the fruit is exactly that of T. striata, but the critical phrase vaginis apice biauriculatis cannot apply to that species if, by biauriculate, Philippi meant 2-parted. There are only two species of Triglochin in the Western Hemisphere with which I am acquainted with 2-parted ligules, the 3-carpellate T. palustris L. which has nearly linear fruits and the 6-carpellate plant with which we are concerned and in which the carpels are never tricarinate. In response to a further communication con- cerning this problem, Dr. Pizarro sent as a gift to the Herbarium of the California Academy of Sciences from the Museo Nacional de Historia Natural a fragment of the type of T. litorea sufficient to determine its relationship. Philippi’s plant is definitely related to T. striata and the ligule is simple as it is in that species. Un- doubtedly by vaginis apice biauriculatis Philippi was describing the conspicuously expanded scarious margins at the top of the leaf-sheaths and was not referring to the ligules at all. From this brief excursion into South American botany, it appears that T. concinna Burtt Davy is the specific name to be applied to the North and South American plants with 6-carpellate fruits and 2-parted ligules. The typical form of the species is the 1 In the fruit of 7. striata there are actually 6 carpels but 3 are sterile and abortive. JANUARY, 1947] TRIGLOCHIN CONCINNA 17 plant found along the Pacific coast of North America while the desert plants of North and South America may be regarded as two varieties. These varieties are similar in the character of the fibrous-coated rootstock, but the southern plant may be dis- tinguished by the more depressed and usually condensed habit and by the more membranous leaf-sheaths. The southern plant has been called T. maritima var. deserticola by both Philippi (1891) and by Buchenau (1903) ; and the plant of the North American desert, as has already been indicated, has been called T. maritima var. debilis M. E. Jones (1895). Bibliographic references, distributional data, and nomenclatural transfers for T. concinna may be summarized as follows : TRIGLOCHIN CONCINNA Burtt Davy, Erythea 3: 117 (1895) ; Jepson FI. W. Mid. Calif. 103 (1901). 7. maritima of authors in part: Piper, Contrib. U. S. Nat. Herb. (Fl. Wash.) 11: 100 (1906) ; Britton, N. Amer. FI. 17 : 42 (1909) ; Abrams Ill. Fl. Pac. States 1:97 (1923); Munz Man. S. Calif. Bot. 22 (1935). T. maritima L. var. debilis of authors in part: Jepson FI. Calif. 1:77 (1912); Jepson Man. Fl. Pl. Calif. 68 (1923); Peck Man. Higher Pl. Ore. 74 (1941). Type-collection: Burtt Davy No. 1116, collected “. . . in the salt marshes around San Francisco Bay, Calif... .” A Davy collection with this number has not been seen; it may be in herb. Greene., University of Notre Dame. Representative collections. Lower CatrrorniA: Lagoon Head, Palmer in 1889 (US).2 CaLiForNIA: False Bay, San Diego Co., Orcutt in 1883 (UC, US); Balboa, Orange Co., Abrams No. 6564 (DS); Terminal Island, Los An- geles Co., G. B. Grant in 1901 (DS) ; Goleta, Santa Barbara Co., Parish No. 11065 (DS); Pismo Beach, San Luis Obispo Co., Condit in 1910 (UC) ; laguna at Del Monte, Monterey Co., Clemens in 1910 (CAS) ; near Palo Alto, Santa Clara Co., C. F. Baker No. 883 (CAS, RM, UC, US); Cooleys Landing, San Mateo Co., Abrams No. 1145 (CAS, DS); West Berkeley, Alameda Co., Davy in 1893 (UC), Walker No. 433 (CAS, DS, G, P, UC, US) ; Stinson Beach, Marin Co., J. T. Howell No. 21256 (CAS, DS, G, P, RM, UC, US, UW, W, WS) ; Samoa, Humboldt Co., Bohman- son in 1906 (CAS). Orecon : Waldport, Lincoln Co., Peck No. 13495 (W). 2 The following symbols indicate herbaria in which cited specimens may be found: CAS, California Academy of Sciences; DS Dudley Merharivan: " Gray Herbarium ; P, Pomona College; RM, Rocky Mountain Herbarium: UC, University of California; US, United States National Herbarium; UW, University of Washington; W, Willamette University; WS. Washineton State College. I am grateful for the privilege of studying so many fine col- lections. Also I wish to express my special thanks to Dr. T. H. Goodspeed for his help and interest in my problem and to Dr. Carlos Munoz Pizarro for a yA i of the Philippi types and the valued fragment of Triglochin itorea . 18 LEAFLETS OF WESTERN BOTANY _ [VOL. V, NO. I WasuHincTon : Double Neck Point, Shaw Island, San Juan Co., Martin No. 304 (P, UW, WS) ; Port Angeles, Clallam Co., Flett No. 3307 (US), Thompson No. 7851 (UW). BritisH Cotumsra: Goose Island, McCabe No. 7158 (UC, UW). Plants intermediate in character between T. maritima and T. concinna have been collected in the salt marsh below Larkspur, Marin County, Cali- fornia, J. T. Howell No. 19449 (CAS). Since they grew with plants typical of the two species, it is believed that the intermediate plants repre- sent natural hybrids. Triglochin concinna Burtt Davy var. debilis (M. E. Jones) J. T. Howell, comb. nov. T. maritima L. var. debilis M. E. Jones, Proc. Calif. Acad. Sci., ser. 2,5: 722 (1895) ; Kearney & Peebles Fl. Pl. & Ferns Ariz. 75 (1942). T. maritima var. debilis of Jepson and of Peck, op. cit., in part. T. maritima of Piper, Britton, Abrams, and Munz, op. cit., in part. Type-collection: M. E. Jones No. 5289, collected at Johnson, Kane Co., Utah, May 23, 1894 (P, type; US). Representative collections. Ca.irornIA: Rosamond, Kern Co., Davy No. 2234 (UC) ; Long Valley, Mono Co., J. T. Howell No. 14384 (CAS, UW); Amedee, Lassen Co., M. E. Jones in 1897 (P, US); Likely, Modoc Co., Eastwood & Howell No. 8036 (CAS, DS), Ripley & Barneby No. 5985 (CAS). Nevapa: Eagle Valley, Ormsby Co., C. F. Baker No. 1050 (P, RM, UC, US) ; Steamboat Springs, Washoe Co., Eastwood No. 14861 (CAS), Ripley & Barneby No. 5946 (CAS); Quinn River Crossing, Humboldt Co., Griffiths & Morris No. 154 (CAS, US); 35 miles west of Austin, Lander Co., Hitchcock et al. No. 4641 (CAS, DS, UW); 5 miles east of Elko, Elko Co., Eastwood & Howell No. 285 (CAS); Steptoe Valley, White Pine Co., Ripley & Barneby No. 6441 (CAS) ; Pahranagat Valley, Lincoln Co., Ripley & Barneby No. 6399 (CAS); Charleston Resort, Charleston Mts., Clark Co., M. E. Jones in 1927 (P). Orecon: Modoc Point, east side of Klamath Lake, Klamath Co., Peck No. 15150 (UW, W) ; Goose Lake Valley near Lakeview, Lake Co., Peck No. 15309 (DS, W); near Alvord Ranch, Harney Co., Henderson No. 8735 (CAS); between McDermitt and Rome, Malheur Co., Henderson No. 8733 (CAS); South Fork of Crooked River, Crook Co., Cusick No. 2620 (P, RM, UC, US). Arizona: west of Holbrook, Navajo Co., Zuck in 1896 (US); Tuba, Coconino Co., Fulton & Peebles No. 11843 (CAS, US). Urau: St. George, Washington Co., Palmer No. 407 (US); Johnson, Kane Co., M. E. Jones No. 5289 (P, type; US); near Moroni, Sanpete Co., J. A. Harris No. C23861 (WS) ; Milford, Beaver Co., M. E. Jones in 1880 (P) ; Jordan Valley, Salt Lake Co., S. Watson No. 1147 (US) ; shore of Great Salt Lake, Tooele Co., Garrett No. 2855 (CAS); valley west of Logan, Cache Co., Maguire No. 13158 (RM). Cotorapo: Blanca, Costilla Co., Bethel, Willey & Clokey No. 3996 (CAS, DS, P, RM, UC, US, WS) ; Steamboat Springs, Routt Co., Bethel in 1919 (RM). JANUARY, 1947] MONOCEPHALOUS PARTHENIUM 19 Wyomrnc: 19 miles east of Evanston, Uinta Co., Maguire et al. No. 12641 (RM) ; Lower Geyser Basin, Yellowstone Park, Rydberg & Bessey No. 3736 (RM). Soutu Daxora: Cone Hills, Harding Co., Over No. 11331 (US). Nortu Daxora: lake near Palermo, Kenmore Co., Ff. P. Metcalf No. 593 (US). Triglochin concinna var. debilis may be expected in New Mexico, Idaho, Montana, and Washington, but no specimens have been seen from those states. Triglochin concinna Burtt Davy var. deserticola ( Phil.) J. T. Howell, comb. nov. T. maritima L. var. deserticola Phil., Anal. Mus. Nac. Chile (Bot.) 1891:77 (1891). T. maritima L. var. deserticola Buchenau in Engler Das Pflanzenr. IV. 14:9 (1903). T. maritima of S. Amer. references. Type-collection: Philippi, from Tambillo, Tarapaca. This collection has not been seen. Representative collections. J Bo.iviA: Comanche, 4000 m., Asplund No. 6450 (US); Ulloma, Pacajes, Asplund No. 2576 (G). CuiLeE: Volcan Llullaillaco, 3500 m., Antofagasta, Werdermann No. 1011 (CAS, G, US); Puerto Natales at sea level, Ultima Esperanza, Eyerdam, Beetle & Grondona No. 24225 (a robust form, G, UC). ARGENTINA: Cordillera de la Rioja, Hieronymus No. 253 (US;; locality cited by Buchenau under var. deserticola Buchenau) ; Laguna Colorada, 3800 m., Jujuy, Fries No. 676 (US) ; Humahuaca, 3050 m., Jujuy, Parodi No. 9705 (plants robust, G) ; Tehuelches, 250 m., Santa Cruz, Donat No. 91 (CAS, G); near Rio Grande, 6 m., Tierra del Fuego, Meria No. 7910 (G, US, UC). The plants of the last-cited collection are taller and more robust than the depauperate plants characteristic of the arid slopes of the high moun- tains and together with the two other collections noted as robust are more like typical T. concinna of the Pacific coast of North America. The last collection is anomalous in that it has ligules both subentire and 2-parted. A NEW MONOCEPHALOUS PARTHENIUM BY R. C. BARNEBY Wappingers Falls, New York Parthenium (§ Bolophytum) Tetraneuris Barneby, spec. nov., a P. alpino (Nutt.) T. & G. foliis latioribus trinerviis, periclinii squamis extimis (persistentibus ) subduplo latioribus, et preesertim achzenio fere duplo majori ala chartacea in auriculos obtusos sursum producta cincto facile separanda. A P. ligulato (Jones) Barneby preterea capitulis pedunculatis flosculisque fertilibus tubulosis et distantius aberrat. Herba acaulis e radice perenni verticali tortuosa lignea orta, cespites depressos 4—15 cm. latos efformans; foliis scapisque in summos caudicis 20 LEAFLETS OF WESTERN BOTANY [ VOL. ViANOAE multicipitis ramulos validos petiolis persistentibus creberrime indutos con- gestis, iis erectis oblanceolatis acutiusculis obtusisve (1) 1.5—4 cm. longis, lamina 3-nervia pube brevi appressa undique cano-sericea 2—4.5 mm. lata in petiolum brevem imo abrupte expansum amplexicaulem hirsutissimum sensim angustata; scapis monocephalis erectis 8—24 mm. longis hirtellis strigosisve; capitulis circa 5 mm. longis et zxquilatis; involucri squamis extimis persistentibus subbiseriatis ovatis 3—4.5 mm. longis, ad 4 mm. usque latis, extus carinatis; bracteis flosculos fertiles marginales suffulcrantibus (mox deciduis) orbiculari-truncatis, achznio flosculisque 2 sterilibus cum eo concretis saltem duplo latioribus; corolla pistillata late tubulosa 1 mm. longa, utrinque emarginata; achznio obovato-cuneato circa 4 mm. longo, 1.6—1.9 mm. lato, fusco, superne sericeo-hirtello, margine callosa inferne acuta ad medium in alam chartaceam erosam stramineam superne in auricu- los obtusos flosculo paulo breviores productam expansa; pappi paleis subu- latis inconspicuis ala subzequilongis. Cotorapo: locally abundant on white shale bluffs of the Arkansas River east of Portland, Fremont County, alt. 5400 ft., in late flower and advanced fruit, 1 June 1946, Ripley & Barneby No. 7662. Type in Herb. Calif. Acad. Sci. No. 332775; isotypes in Gray Herb., N. Y. Bot. Gard., U. S. Nat. Herb. Also on an isolated shale outcrop some miles to the east, near the confluence of Turkey Creek with the Arkansas, Pueblo County, No. 7666. The new species is closely related to the legendary and long- lost Parthenium alpinum (Nutt.) T. & G. (the type of Nuttall’s genus Bolophyta), being altogether similar in habit and pubes- cence, in its pedunculate heads, and in the tubular pistillate corollas. The original Bolophyta alpina is known to me only from the fragment in the Gray Herbarium, but this differs from the Colorado plant in its single-nerved, narrower, linear-spatulate. leaves only 1.5—2 mm. broad, in its narrower phyllaries (the largest of the outer ranks not exceeding 2 mm. in width), and especially in the smaller, glabrate, wingless achene only 2.2 mm. _long and about 1.2 mm. broad. In P. Tetraneuris the leaves are conspicuously triplinerved and 2—4.5 mm. wide, the phyllaries are all broader, the larger in the outermost ranks reaching 4 mm. in width, and the body of the copiously hirtellous achene is 4 mm. long, with the callous margin broadened upwards from the middle into a papery, erose wing produced beyond the seed-body into a pair of blunt auricles little shorter than the marcescent floret. The third member of the section, P. ligulatum (Jones) Barneby, stat. nov., first described as P. alpinum var. ligulatum Jones (Cont. W. Bot. 13:16,—1910), is a distinct species JANUARY, 1947] MONOCEPHALOUS PARTHENIUM 21 endemic to the Navajo Basin in Utah. It differs from both P. alpinum and P. Tetraneuris not only in the strap-shaped pistil- late florets, but also in the sessile capitula which nestle, almost hidden, among the hairy petiole-bases, and in the short, broadly spatulate leaves of a thicker texture and yellowish-green color. In an isotype of P. ligulatum (10 miles south of Theodore (i. e. Duchesne), Duchesne Co., Utah, Jones in 1908, N. Y. Bot. Gard.) and an approximate topotype (southwest of Duchesne, Ripley & Barneby No. 7806) the achene is hoary as in P. Tetra- neuris but wingless ; while the persistent phyllaries are relatively narrow as in P. alpinum. In an interesting paper entitled “The Story of Parthenium alpinum,” G. J. Goodman ( Madrofio 7: 115, sequ.,—1943) drew attention to the mystery surrounding the type-locality of the original Bolophyta, which has not been seen since its discovery by Nuttall in 1834. Goodman pointed out that a discrepancy exists between the season (“June’’) and part of the locality-data (“Three But [t] es’) published by Nuttall. The Wyeth trans- continental expedition spent the whole of June 1834 in Wyoming ; while the Three Buttes, if correctly identified with the volcanic eminences which rise from the floor of the Snake River valley near Pocatello, Idaho, were not reached until mid-July. Goodman came to no conclusion as to which part of Nuttall’s statement was more likely correct. However, lately acquired knowledge of the ecology of the other species of the section may perhaps throw some light on the problem. These curious dwarf Composite are highly specialized xero- phytes, simulating in their multicipital caudex clothed with im- bricated, hirsute, marcescent leaf-bases and in their reduced monocephalous inflorescence several species of Actinea § Tetra- neurts. As I have seen them, P. ligulatum and P. Tetraneuris are abruptly confined to small areas of gypseous shale at altitudes between 5400 and 5750 feet, where they are associated with spe- cies (Stanleya, Astragalus racemosus) which occur only on soils rich in selenium. Both flower in early spring. On June 1, 1946, P. Tetraneuris was already in advanced.fruit along the Arkansas, and a week later P. ligulatwm at Duchesne was so mature that Mr. Ripley and I had difficulty in finding even a few scraps in which the heads did not disintegrate at a touch. One may assume 22 LEAFLETS OF WESTERN BOTANY _ [VOL. V, NO. I with some assurance—so alike are the three species—that P. alpi- num is also a plant of naked shale benches at middle altitudes, and that Nuttall, as he claimed, collected the type in June. It is doubtful whether the peculiarly deciduous florets (which must have suggested the name Bolophyta) could have been obtained at a later date. In fact I would suggest that Nuttall’s locality is substantially correct, and that the species is to be looked for “near the sources of the Platte,” very likely in Converse or Natrona County, Wyoming. Suitable habitats for a Bolophyta (as well as buttes—perhaps even Nuttall’s Three) are plentiful in the region, and Nuttall was thereabouts in June. Phytogeographi- cally this area is most probable, linking as it does the distribution of the remotely disjunct species P. Tetraneuris and P. ligulatum into a pattern of endemism exhibited by other specialized western genera. An example from Astragalus (with which I happen to be familiar) may be given. Astragalus simplicifolius Gray is a local endemic of bluffs and buttes in central Wyoming, chiefly along the upper waters of the Platte, but has an outlying station (possibly of an aberrant form) on the Arkansas near Cafion City. Its closest relative, A. detritalis Jones, was first collected near Duchesne, Utah, where it is found in close association with Parthenium ligulatum on white shale bluffs. Like the Parthenium it is a rare endemic of the Navajo Basin. A word to collectors. The Bolophytas occur in local abun- dance, but are deceptively inconspicuous. At Portland, I myself passed over many hundreds of plants of P. Tetraneuris for a pathogenic or starveling state of the Actinea with which it was growing; though when Mr. Ripley had pointed out the small greenish heads hidden among the leaves the differences became obvious enough. It seems possible that T. S. Brandegee, during his very thorough herborizations about Cafion City in 18722>) may have been similarly deceived ; clearly he must have reached these or similar shale barrens to have come across the narrowly endemic Oxybaphus polytrichus (Standl.) Barneby.t Those seeking the lost P. alpinum in Wyoming will do well to scruti- nize any sterile Actinea along the upper branches of the Platte. _1Oxybaphus polytrichus (Standl.) Barneby, comb. nov. Allionia poly- tricha Standl., Cont. U. S. Nat. Herb. 12:346 (1909). In white shale near Portland, Fremont Co., Colorado, Ripley & Barneby No. 7663.—A most dis- tinct species, apparently confined to the gypseous formations along the Arkansas, which deserves a name in the conserved genus Oxybaphus. JANUARY, 1947] TYPE OF MALVASTRUM 23 TYPE OF THE GENUS MALVASTRUM BY THOMAS H. KEARNEY The writer proposes, for the reasons here set forth, Malvas- trum carpinifolium A. Gray (now referred to M. coromandeli- anum) as the lectotype of the genus Malvastrum, A. Gray. In establishing this genus, Gray’ did not designate a type but assigned to it eight species, listed in the order named: M. cocci- neum, M. grossulariefolium, M. Munroanum, M. Fremontii, M. Wrightti, M. carpinifolium, M. spicatum, and M. angustum. Malvastrum grossulariefolium and M. Munroanum .were transferred subsequently to Spheralcea, by Gray himself.? Mal- vastrum Fremonti belongs to a group, limited to California, Lower California, and adjacent islands, upon which E. L. Greene based his genus Malacothamnus.* In the writer’s opinion this is a valid genus, distinguished from Malvastrum by having carpels that are completely dehiscent (splitting into halves at maturity), muticous, unappendaged, and not rugose, as well as by the charac- teristic habit and appearance of the plants. Malvastrum an- gustum is the type of the genus Sidopsis, Rydb.* It has carpels very like those of Malacothamnus, but the plant is quite different in other characters and in geographical distribution. Sidopsis should, perhaps, be recognized as a monotypic genus.° The first species listed by Gray, M. coccineum, has been pro- posed by M. L. Green® as the lectotype of Malvastrum. It may be inferred that Gray (Plant. Fendl. p. 21) regarded M. cocci- neum as the typical species, and it is the first of the two species that were figured in his Genera [llustrata.* Garcke, however, pointed out long ago® that this species approaches Spheroma (Spheralcea) in its apically dehiscent carpels, not filled by the seed. Rydberg® transferred M. coccineum to Spheralcea, and the writer concurred.’® The sharp differentiation of the carpels 1 Plant. Fendl., Mem. Amer. Acad. Arts & Sci., N. S., 4:21, 22 (1849). 2 Proc. Amer. Acad. Arts & Sci. 22:291, 292 (1887). 8 Leaflets Bot. Obs. & Crit. 1:207, 208 (1906). 4 Fl. Prairies & Plains Centr. N. Amer. 541 (1932). 5Of South American species related to Malvastrum, but having com- pletely dehiscent carpels, many have been transferred to the genus Noto- triche, Turez. and others probably are referable to the genus 7J’arasa, Phil. 6 Intl. Rules Bot. Nomenclature ed. 3, 145 (1935). 7 Gen. Fl. Amer. Bor.-orient. Lllustrata 2:59, 60, pl. 121 (1849). Plate 121 does not show the dehiscent apical portion of the carpel, but this seems to be present, invariably. 8 Bonplandia 5:294, 295 (1857). 9 Bull. Torrey Bot. Club 40:58 (1913). 10 Univ. Calif. Publ. Bot. 19:5, 94 (1935). 24 LEAFLETS OF WESTERN BOTANY [ VoL. V, NO. I into a rugose-reticulate, indehiscent basal portion and a smooth and dehiscent apical portion (the latter very small in typical S. coccinea, but well developed in var. elata), and the presence of an endoglossum, seem fully to warrant this disposition, if any precise delimitation of these genera is to be attained. There- fore, if M/. coccineum be taken as the type, the name Malvastrum becomes a synonym of Spheralcea. The large group of species of which M. coromandelianum may be regarded as typical would then be left without a valid name, unless we follow O. Kuntze” in reviving Presl’s name Malveopsis, based upon an imperfectly described species which E. G. Baker** found to be closely related to Malvastrum capense Gray & Harvey. The name Malveopsis antedates Malvastrum by five years, but Malvastrum is conserved versus Malveopsis.** Although the South African species, including the one upon which Malveopsis, Presl was based, seem to be congeneric with such tropical and subtropical American species as M. coroman- delianum, further research may warrant their separation. In any event, it would seem preferable to choose as the type of Malvastrum one of the species listed by Gray when he published the genus. Only three of these belong properly to Malvastrum, if that genus be defined as having the carpels indehiscent or very nearly so, and not sharply differentiated apically and basally. The species in question are M. Wrightii A. Gray (now referred to M. aurantiacum (Scheele) Walp.), M. carpinifolium A. Gray (M. tricuspidatum A. Gray, now referred to M. coro- - mandelianum (L.) Garcke), and M. spicatuwm (L.) A. Gray. K. Schumann*® mentioned M. aurantiacum as “der Typus der Gattung,” but did not explain how he arrived at this conclusion. Although this is a true Malvastrum as the writer would define the genus, M. coromandelianum is much more widely distributed and is, in some respects, more representative of the group as a whole. It is, therefore, proposed that this species (as M. carpini- folium in the original publication) be chosen as the type. 11 Phymosia, Desv., Meliphlea, Zucc., and Ilamna, Greene, with com- pletely dehiscent, non-reticulate carpels and, for the most part, obliquely truncate, slightly decurrent (not symmetrically capitate) stigmas, probably should be excluded from the genus Spheralcea. 12 Rev. Gen. 3, pt. 3 :20 (1898). 13 Jour. Bot. Brit. & For. 32:186 (1894). 14 Intl. Rules Bot. Nomenclature ed. 3, 102 (1935). 15In Engler & Prantl, Die Natiirl. Pflanzenfam. III. 6:41 (1895). \ oe ee TE act > irda Riven gs oe ek a Vot. V No. 2 LEAFLETS of WESTERN BOTANY CONTENTS PAGE Pugillus Astragalorum VIII: Notes on Section Genistoidei 25 R. C. BaRNEBY ewe Vanetyiin Orobanche =... 05 5 eh A136 GerorcE J. GoopDMAN Studies in Carex—II ERY kl RMON A) cp Le eC AA eK, JouHn THomas Howey SAN FRraANcisco, CALIFORNIA Apri 25, 1947 LEAFLETS of WESTERN BOTANY A publication on the exotic flora of California and on the native flora of western North America, appearing about four times each year. Subscription price, $1.00 annually; single numbers, 40c. Address: John Thomas Howell, California Academy of Sciences, Golden Gate Park, San Francisco, California. Cited as LEAFL. WEsT. Bot. PUUUUUEPTUDP TATED PETE EP TERETE PUPA Tey INCHES ee eae ee ee eT MU LiL bile pilad May eu Ms Owned and published by Aticg Eastwoop and JoHN THomas Howe APRIL, 1947] PUGILLUS ASTRAGALORUM , 25 PUGILLUS ASTRAGALORUM VIII: NOTES ON SECTION GENISTOIDEI BY R. C. BARNEBY Wappingers Falls, New York ASTRAGALUS § GENISTOIDEI Astragalus § Genistoidei (T. & G.) Barneby, stat. nov. Homalobus § Genistoidee [sic] T. & G., Fl. N. Amer. 1: 351 (1838). Typical species : Homalobus campestris Nutt. Homalobus § Campestres Rydb., Bull. Torr. Club 50:26 (1923), based ‘on the same. I take up for Rydberg’s Campestres the early sectional name applied by Torrey and Gray to the first five species to be made known in the group. The section, which should be enlarged to include A. lancearius, discussed below, formed a part of Gray’s Astragalus ser. Phaca § Homalobi * Genuini, of Sheldon’s As- tragalus ser. Phaca § Homalobus, and of Jones’ Astragalus § Homalobi. In their Flora of North America (1838) Torrey and Gray published from Nuttall’s manuscripts three closely allied species of Homalobus collected in the Rocky Mountains: H. campestris, H. junceus, and H. orthocarpus. All are similar in general ap- pearance, characterized by their sparsely leafy, often rushlike, stems, free stipules, reduced leaves with mostly decurrent leaflets, and small, ochroleucous flowers with abruptly recurved banner. Owing to the fragmentary nature of the material obtained by Nuttall, the identification and evaluation of the types has proved from the first to be a problem of some difficulty. Since their original publication they have been very differently interpreted by Gray, Macbride, Jones, and Rydberg, all of whom published notes on this perplexing group, and, further complicated by the fact that all three specific names published in Homalobus were preoccupied in Astragalus, the synonymy has become pretty in- volved. An early confusion arose when Gray applied the name A. campestris, not to the Nuttallian original, but to the allied A. decumbens (Nutt.) Gray,’ which differs from all members of the group under discussion by its connate stipules. Apart from this error, perpetuated by Jones and others, Nuttall’s three Homalobi have generally come to be accepted as representing a 1Cf. Cronquist, A. The varieties of Astragalus decumbens. Leafi. West. Bot. 3: 250—254 (1943). Leaflets of Western Botany, Vol. V, pp. 25-40, April 25, 1947. 26 LEAFLETS OF WESTERN BOTANY [ VOL. Vv, NO. 2 single, polymorphic species, with H. orthocarpus as a rule main- tained as a variety. Rydberg, however, in his review of Homalo- bus,? continued to distinguish all three, as well as the related A. junciformis A. Nels., as specific entities. Study of recent col- lections of H. orthocarpus and field-experience of several races of H. campestris have convinced me that the key-characters em- ployed by Rydberg are for the most part illusory, but at the same time have brought to light unexpected criteria which have gone unnoticed hitherto. The result of these enquiries may best be presented in the form of a Key To ASTRAGALUS DIVERSIFOLIUS AND ITS ALLIES 1. Pod narrowly oblong, 3—4 mm. wide and 10—15 (17) mm. long; leaves 2—5.5 cm. long, the upper and sometimes nearly all re- duced to a thin, flat, linear or lanceolate, grasslike phyllodium 2—5 mm. (some rarely 1 mm.) wide, the lateral leaflets, when present, 2—4 in number, of similar shape but shorter, decurrent on the rachis; stems slender, flexuous, prostrate, not rigid or abruptly zigzag; plants of moist bottomlands in alkaline soil i AB OD SSN ans eS ABS NY Re a et Seb ne Se A Na “1. A. diversifolius 1. Pod linear to oblanceolate, mostly 2 mm., rarely up to 3.5 mm. wide or, if wider (in var. riphoides), then at least 2.5 cm. long; leaves 410 cm. long, the rachis filiform, canaliculate, some- times enlarged and hooked at apex, the lateral leaflets, when present, 2—10 in number, linear-filiform to subulate and com- monly involute or, if flat (in var. foliolatus), then all articulate to the rachis ; stems mostly stouter, ascending or erect, rigid and rushlike ; plants of dry soil on plains and foothills. (A. conval- larius, with four varieties. ) 2. Leaflets all or nearly all decurrent on the rachis, commonly lack- ing on the upper leaves, the terminal represented by an often falcate prolongation of the rachis. 3. Pod linear, linear-oblong or linear-lanceolate, widest at or below the middle, or the sutures parallel throughout; corolla ochroleucous or tinged with lurid purple. 4. Pod narrowly linear, 1.5—3 (3.5) mm. wide, not very strongly compressed at maturity, the valves extended by the seeds............-....2-- 2a. A. convallarius var. typicus 4. Pod linear-oblong, 4—5 mm. wide, flattened at maturity PIU AMER A OS reese ALY Na a 5 EO chk i ae A 2b. var. xiphoides 3. Pod linear-oblanceolate, narrowed below and widest near the obliquely deltoid apex; flowers bright purple................-.-- EL SRERRU GIS Oey rw eh SEP CS a ee 8 2c. var. scopulorum 2. Leaflets, including the terminal, present in all the leaves and. articulate ‘with, the’ rachis. 1220s ec nen scectmn pore 2d. var. foliolatus 2 Bull. Torr. Bot. Club 50: 261, sequ. (1923). APRIL, 1947] PUGILLUS ASTRAGALORUM 27 1. A. prversiFotius Gray Astragalus diversifolius Gray, Proc. Amer. Acad. 6: 230 (1864). Ho- malobus orthocarpus Nutt. ex T. & G. Fl. N. Amer. 1: 351 (1838), non _ A. orthocarpus Bss. A. campestris (Nutt.) Gray var. diversifolius (Gray) Macbr., Cont. Gray Herb. 65:35 (1922). A. junceus (Nutt.) Gray var. orthocarpus (Nutt.) Jones Rev. Astrag. 76 (1923). A. convallarius Greene var. diversifolius (Gray) Tidestr., Proc. Biol. Soc. Wash. 50:20 (1937). A. ibapensis Jones, Zoe 3:290 (1893). Atelophragma ibapense (Jones) Rydb., Bull. Torr. Bot. Club 40:51 (1913). Astragalus reclinatus Cron., Madrofio 7:79 (1943). Wyominc: Colorado of the West, Nuttall (P*) ; Platte Sources, Nuttall (G, perhaps part of the same collection) ; south fork of the Platte (? in Colorado), Geyer No. 2 (G). Ivano: Clyde, Blaine Co., Macbride & Payson No. 3183 (G); Springfield, Bingham Co., R. J. Davis No. 1531 (Poc) ; Dickey, Custer Co., Cronquist No. 3086 (Poc, type-collection of A, reclinatus). Utau: Juab, Juab Co., Goodding No. 1084 (G, NY); Ibapah, Tooele Co., Jones (PO, type-collection of A. ibapensis). From the collections obtained by Nuttall and Geyer (the last correctly identified by Hooker, Lond. Journ. Bot. 6: 214, 1847, as Homalobus orthocarpus), which show only the upper parts of the stem, it is impossible to appreciate the true character of A. diversifolius. Nuttall’s plant, which bears a ripe fruit, is in the same stage of maturity as, and exactly matches, that obtained by Davis in Idaho. In both these the phyllodia are rather narrow and the lateral leaflets, present only on the lower leaves (as noted but not collected by Nuttall) are few or almost wanting. Both, however, are unquestionably of the same species as the younger and more luxuriant plant collected by Cronquist (A. reclinatus), and the type of A. ibapensis Jones. The shape of the pod, rela- tively short and broad as compared with that prevailing in _ A. convallarius, is not perfectly diagnostic, as I formerly supposed when reporting (Proc. Calif. Acad. Sci. IV, 25: 148,—1944) A. diversifolius from southern Utah. The plant which I then had in mind (Ripley & Barneby No. 4967) has pods of almost exactly the same proportions as A. diversifolius, as little as 15 mm. long and up to 3 mm. wide, but in other respects it is charac- teristic A. convallarius var. typicus, to which I would now refer it. Likewise the length of the calyx-teeth, by which Macbride partly defined his var. diversifolius from typical “A. campestris” (i. e., convallarius), is equally unreliable, these organs varying within about the same limits in both species. The specific validity of A. diversifolius depends less on characters of the flower and 8 The symbols used here to indicate herbaria in which cited material is deposited are given on page 1 of this journal. It should be noted that the 2 de published there was numbered V whereas it should have been VI. 28 LEAFLETS OF WESTERN BOTANY [ VOL. V, NO. 2 fruit than upon the very striking habital peculiarities, the slender prostrate stems, the broad, thin, grasslike blades of the rachis and » leaflets, and on the distinctive habitat. From the few labels which bear ecological data, it seems that A. diversifolius is a plant of meadows (which always connote moist, and usually alkaline, soil in the Rocky Mountain states), or damp bottom lands, and this may partially account for its rarity in herbaria. Such places are usually poor in species and not much favored by modern collectors, though it was not always so, for early travellers, such as Nuttall and Geyer, habitually followed and camped near water. By reason of its habitat A. diversifolius might be regarded as a mesophytic ecotype of A. convallarius, but in other respects this seems improbable. If such were the case one would expect the broader and thinner, but not shorter, leaves, pods larger if anything (whereas they are actually smaller), and in general the plants would be more, not less, robust. It is surely significant of the strong impression made upon the mind by the living plant that twice in the last fifty years the species has been redescribed, yet in neither case did the author suspect any close relationship with A. convallarius, Jones comparing it with A. atratus Wats., Cronquist with A. decumbens and A. flexuo- sus Dougl. Along with them I have no hesitation in accepting A. diversifolius as a distinct species. The three trinomials cited in the synonymy are all invalid. Both A. campestris and A. junceus of Gray are later homonyms, while the legitimate A. diversifolius, being of much earlier date than A. convallarius, cannot, whatever the taxonomy of the plants involved, be reduced to varietal status under it. 2a. A. CONVALLARIUS Greene var. TyPICcUS Barneby Astragalus convallarius Greene var. typicus Barneby, nom. nov. Astragalus convallarius Greene, Erythea 1: 207 (1893). Homalobus cam- pestris Nutt. ex T. & G. Fl. N. Amer. 1:351 (1838). A. campestris (Nutt.) Gray, Proc. Amer. Acad. 1:229 (1864), non Linneus. A. de- cumbens var. convallarius (Greene) Jones, Contrib. West. Bot. 10:58 (1902). Phaca convallaria (Greene) Piper, Contrib. U. S. Nat. Herb. 11: 373 (1906), as to name. Homalobus junceus Nutt. ex T. & G. Fl. N. Amer. 1:351 (1838). A. junceus (Nutt.) Gray, Proc. Amer. Acad. 6:230 (1864), non Ledeb. ex Spreng. A. diversifolius Gray var. junceus (Nutt.) Jones, Cont. W. Bot. 8:13 (1898). A. diversifolius var. roborum Jones, op. cit. 10:61 (1902). A. junciformis A. Nels., Bull. Torr. Bot. Club 26:9 (1899). Homalobus junciformis (A. Nels.) Rydb., op. cit. 32: 666 (1906). A. junceus var. attenuatus Jones Rev. Astrag. 76 (1923). APRIL, 1947] PUGILLUS ASTRAGALORUM 29 This, the common form of the species, is distributed over the plains and foothills from central Montana and southern Idaho (in the valley of the Snake River) to eastern Nevada, western Wyoming and throughout Utah south to the Colorado River. The Washington record of Phaca convallaria (Piper, Cont. U. S. Nat. Herb. 11: 373,—1906) is based on A. decumbens. A col- lection cited by Wooton & Standley (op. cit. 19: 365,—1915, as A, diversifolius) from Gallup, New Mexico, has not been ex- amined, but from the locality one might suspect it to belong to var. xiphoides below. The material of var. typicus is rather diverse, particularly in the length of the pod, which reaches limits of 1.5 and (in var. attenuatus Jones) 5 cm. in length. In outline the pod is always linear or sometimes gradually acuminate above and then linear- lanceolate. This is true of the types of both Homalobus junceus and H. campestris, which differ essentially only in the relative length of the calyx-teeth, and in the incidence of dark hairs on the tube, characters which Macbride (Cont. Gray Herb. 65: 35,— 1922) rightly considered superficial as criteria. Dark and light hairs occur in varying proportions on individuals of the same population, and the teeth may fluctuate between subulate and del- toid on the same hillside. The combination of nigrescent calyx with deltoid teeth, used to characterize A. junciformis A. Nels., is found here and there throughout the range of var. typicus. In the North American Flora (vol. 24, p. 257) Rydberg keyed out his Homalobus campestris as having the pod “tapering toward the base” and further described it (op. cit., p. 267) as “linear- oblanceolate . . . with a cuneate apex.” He had earlier defined the imperfect type of H. campestris (Bull. Torr. Club 50: 262,— 1923) as “in every respect” matching a plant from Colorado, C. F. Baker No. 242 (NY), which is in fact characterized by such a pod, and it is clear that he had transferred his concept of H. campestris to the race represented by this collection. Although his interpretation of Nuttall’s fragmentary type was faulty, Ryd- berg was justified in segregating the plant with the oblanceolate pod native to the upper drainage of the Grand River as an entity distinct from widespread “H. junceus’ ; it is described below as A. convallarius var. scopulorum. 30 LEAFLETS OF WESTERN BOTANY [ VOL. V, NO. 2 The petals of var. typicus are normally ochroleucous, but the banner is often veined, and sometimes suffused, with lurid purple. At maturity the seeds completely fill the cavity of the pod and the valves become extended, the cross-section being at this stage broadly elliptic or oval. In some robust specimens, among them the type-collection of A. junciformis A. Nels. (Point of Rocks, Wyoming, A. Nelson No. 3081, NY), the leaflets are expanded and up to 3 mm. wide; these might be thought to connect A. con- vallarius with A. diversifolius, but the plants are quite unlike the latter in other respects. An extreme form from northeastern Arizona seems worthy of separation as 2b. A. CONVALLARIUS Greene var. XIPHOIDES Barneby Astragalus convallarius Greene var. xiphoides Barneby, var. nov., a precedenti legumine latiori ensiformi 4—5 mm. lato, 2.5—3.5 cm. longo, maturo valde compresso, valvulisque tenuioribus subdiaphanis absimilis. ARIZONA: on detrital clay banks 5 miles northeast of Holbrook, Navajo Co., alc. 5200 ft., Ripley & Barneby No. 5246. Type in Herb. Calif. Acad. Sci., No. 324951. Isotypes G, NY. Among the extensive collections of the preceding, to which var. xiphoides is closely allied, none has been found to have pods more than 3.5 mm. wide, and this occurs only in a few extreme individuals. The Arizona plant is quite unusual in having the pods 4—5 mm. wide, and flattened at maturity, the valves not extended by the maturing ovules, while the texture of the valves themselves is thinner. In the shape of the pod it much resembles A. lancearius Gray, but the abruptly recurved banner and the outline of the keel-petals are characteristic of the A. convallarius alliance. Dr. T. H. Kearney informs me (in litt.) that the plant cited from Holbrook as A. junciformis in Kearney & Peebles, FI. Pl. Ariz. 477 (1942), represents the same form. 2c. A. CONVALLARIUS Greene var. SCOPULORUM Barneby Astragalus convallarius Greene var. scopulorum Barneby, var. nov., a precedentibus corolla purpurea (nec ochroleuca) ac presertim legumine oblanceolato infra apicem oblique deltoideum tota fere longitudine in caly- — cem sensim angustato recedens. Cotorapo: dry banks among junipers, 25 miles north of Loma, Garfield Co., 28 May 1943, fl. & fr., Ripley & Barneby No. 5478. Type in Herb. Calif. Acad. Sci..No. 332325. Colorado Territory, lat. 39°—41°, Parry in 1864 (P) ; Middle Park, Patterson in 1879 (P) ; Glenwood Springs, Gar- field Co., Osterhout in 1895 (NY) ; Cedar Edge, Delta Co., Baker No. 242 (NY) ; north slope of Grand Mesa, Mesa Co., Ripley & Barneby No. 5429. APRIL, 1947] PUGILLUS ASTRAGALORUM 31 As noted above, the var. scopulorum is in large part the Homalobus campestris of Rydberg, though the Utah material referred there by him (Bull. Torr. Club 50:262,—1925), as well as the actual type of that species, belong to the much commoner race which he treated as Homalobus junceus, i. e. typical conval- larius as defined above. From the latter it differs in the pod, which, though nearly linear when young, early becomes oblanceo- late, broadest near the obliquely deltoid apex and gradually narrowed downward into the calyx. It is, in fact, very similar to the pod of some forms of A. decumbens (Nutt.) Gray, and quite unlike the truly linear or linear-lanceolate one of var. typicus. In addition the corolla, when fresh, is of a bright magenta-purple, but the color is fugacious, and after a few months in the herbarium turns to a dull, tawny yellow ; so that this charac- ter should be used with caution in determining dried material. 2d. A. CONVALLARIUS Greene var. FOLIOLATUS Barneby Astragalus convallarius Greene var. foliolatus Barneby, var. nov., a var. typico nob., cui floribus ochroleucis leguminisque forma proxime accedit, imprimis foliis perfecte imparipinnatis, foliolis 4—6-jugis lineari- oblongis, omnibus (et terminali) rachi articulatis absimilis. Legumina 2.5—3.5 cm. longa, 3—4.5 mm. lata, utrinque abruptiuscule acuminata, recta vel leviter falcata ac pedicello torto sepe resupinata, primum valde compressa, demum valvulis extensis subteretia. Arizona, all but the last from Coconino County: open flats among juni- pers, in clay and sand, 16 miles southwest of Frazier’s Wells, alt. 5800 ft., 13 May 1943, fl. & fr., Ripley & Barneby No. 5229. Type in Herb. Calif. Acad. Sci., No. 324952. Isotypes G, NY, PO. Similar habitat, 13 miles southwest of Frazier’s Wells, alt. 6000 ft., No. 5223; calcareous plateau, near Hyde Park, alt. 5000 ft., No. 3404; west of Peach Springs, Mohave Co., alt. 4450 ft., No. 7017. The var. foliolatus is a well-marked race, apparently not uncommon on the limestone plateau between Peach Springs and the Grand Canyon. With its regularly pinnate leaves it does not immediately suggest A. convallarius to the eye, but the abruptly recurved banner, the slender, often contorted pedicels, and the form of the keel-petals and legume are truly of this spe- cies. Apart from the presence of leaflets jointed to the rachis in all the leaves, the variety is distinguished from var. typicus by the somewhat broader pod, but plants of the latter from southern Utah approach it in this respect. In habit and foliage the variety bears a striking likeness to A. pinonis Jones of eastern Nevada 32 LEAFLETS OF WESTERN BOTANY [ VOL. V, NO. 2 and adjacent Utah, and the ripe pods of the two are not dissimilar. But in that species the racemes are shorter, not elongating notice- ably in fruit, the pedicels are scarcely half as long (1—2 mm.), and never contorted, the keel is produced into a slender beak, and the pod is at no stage of its development compressed, but terete and somewhat inflated from the first. Although referred to Pisophaca by Rydberg, A. pinonts is no doubt a close relative of the junceous Homalobi. It seems likely that in Homalobus salide Rydb. (Bull. Torr. Bot. Club 32: 667,—1906) there exists a fifth race of A. conval- larius, standing in the same relation to var. scopulorum as does var. foliolatus to var. typicus, i. e., a form with regularly pinnate leaves. The type of H. salide (Salida, Colorado, Shear No. 3468, NY) is hardly sufficient for analysis, and I have seen neither duplicates nor further collections. In addition to the regularly pinnate leaves it differs from var. scopulorum in its slightly narrower pod which is not truly stipitate, as claimed, but merely narrowed downward into the calyx in the manner of that variety ; and the extension of the valves by the growing ovules is more pronounced, so that the pod becomes subterete at maturity. But this seems to be due to its narrowness rather than to any structural difference, and Rydberg’s eventual transfer of the species to Pisophaca is probably misleading. ASTRAGALUS LANCEARIUS Gray, Proc. Amer. Acad. 13: 370 (1878). A. Episcopus Wats., op. cit. 10:346 (1875), nomen confusum, Urau: 17 miles east of Kanab, Kane Co., Ripley & Barneby No. 4833. San Rafael Swell, Emery Co., No. 4719. During the sixty-odd years since the description of A. Episco- pus and A. lancearius, the existence in southern Utah and northern Arizona of two separate species has found general ac- ceptance in the literature. Tidestrom, it is true, reduced (FI. Nev. Utah 331,—1925) A. lancearius to Watson’s earlier propo- sition, but both Jones and Rydberg, in their respective mono- graphs, continued to recognize them as distinct. In his key to the Homalobi, Jones (Rev. Astrag. 67) claimed that A. Episco- pus could be separated from A. lancearius by its short peduncles, short bracts, and small calyx-teeth ; but in the type of A. lancearius (Beaver Dam, Arizona, Palmer No. 114, G) the calyx-teeth are / APRIL, 1947] PUGILLUS ASTRAGALORUM 33 minute and deltoid (as described for A. Episcopus by Jones), and the peduncles are relatively no longer, in both “species” far exceeding the subtending leaf. Rydberg, who discussed (Bull. Torr. Bot. Club 50: 184,—1923) the identity of A. Episcopus at some length, pointing out that it was founded on a mixed col- lection, concluded that it might be distinguished from A. lance- arius in having “shortly stipitate,” as opposed to “strictly sessile” pods, and so maintained it in the North American Flora. More recently Kearney & Peebles (Fl. Pl. Ariz. 476) have continued to recognize the two as occurring in Arizona, but they express a note of doubt as to whether they are truly distinct. This doubt is fully justified by study of a series of specimens and of the living plants. Like many of its relatives, A. lancearius is variable in just those characters which have been claimed as diagnostic. The calyx may be either shortly campanulate or somewhat cylindric, the teeth in either case varying from deltoid and obtuse to subu- late and acute, from 0.6 to 1.5 mm. in length. The legume, even in the same collection, varies between the lanceolate form with subsymmetrical, cuneate base as displayed by Parry No. 114, through the lance-oblong with obliquely cuneate base described by Jones as characteristic of A. Episcopus, to an extreme in which the ventral suture is so strongly arched at base as to form an oblique right-angle with the straight dorsal suture. Normally glabrous and sessile, the pod may be either strigose or abruptly narrowed at base into a stipe about 0.5 mm. long; in fact sessile, strigose pods and substipitate glabrous ones are both represented in the type-collection of A. Episcopus. Among these variables I have been unable to work out any correlation, and conclude that there is but a single species involved. For this single species I am inclined to adopt the later name, A. lancearius Gray, discarding A. Episcopus as a nomen con- fusum. The latter was named from a mixture of material col- lected in southern Utah in 1873 by Capt. Bishop. Presumably it was obtained somewhere near Kanab, where the Powell Grand Canyon Expedition had its headquarters, or conceivably at some point between that village and the Grand Canyon. There are now extant at least four sheets, distributed by Watson as “A. Episcopus n. sp.,’ and annotated in his hand. Of these four, 34 LEAFLETS OF WESTERN BOTANY [ VOL. V, NO. 2 that in the Torrey herbarium (NY) is, at least as to the fruit, A. Coltoni Jones, a species with long-stipitate pod and purple > flower ; that in the Gray herbarium is the form of A. lancearius with short-stipitate pod as to the fruiting part, and possibly also as to the flowering branch, but the two specimens mounted to- gether differ in the shape of the calyx, and the latter is probably A. Colton; while the two at the Philadelphia Academy include four elements, a loose pod, shortly stipitate and glabrous, evi- dently the same as that in herb. Gray., several racemes of sessile, strigose pods (well matched by part of my collection from Kanab, Utah), some flowers of A. Coltoni, and a fragment of another species with ochroleucous petals, insufficient for determination. Watson’s description, it must be admitted, covers only part of this assemblage of diverse plants, and his concept was perhaps less mixed than would appear from collections accepted by him as typical. The pod described as “very shortly stipitate” is no doubt that preserved at the Gray Herbarium, the “purple” flower can belong only to A. Coltoni, but Watson seems not to have taken into account that the pod might also be sessile and strigose. In the genus Astragalus, when a name is known to have been applied originally to a mixture, and where there exists fruiting material perfect enough for certain determination, I would favor, in principle, the restriction of the name to that element of the type. That course could be adopted here. But since there is avail- able a name, published only three years later, founded on an excellent single, not fragmentary and mixed type, which repre- sents, moreover, the commoner phase of a variable species, it seems reasonable to reject A. Episcopus in favor of the clearly defined, if later, A. lancearius. Astragalus decumbens ( Nutt.) Gray var. crispatus (Jones) Cronquist & Barneby, comb. nov. A. campestris var. crispatus Jones, Rev. Astrag. 75 (1923). Montana: Alta, Beaverhead Co., Jones (PO, type). IpaHo: 7 miles east of Tendoy, Lemhi Co., Christ & Ward No. 14778 (NY); 14 miles east of Leadore, Lemhi Co., Christ & Ward No. 14805, 14810 (NY). In his review of the A. decumbens complex, Cronquist (Leafl. West. Bot. 3:251,—1943) pointed out that A. campestris var. crispatus Jones was known only from the type-collection, but sug- APRIL, 1947] PUGILLUS ASTRAGALORUM 35 gested that when further known it might prove acceptable as a distinct variety of A. decumbens. Dr. Cronquist and I agree that the additional material lately acquired by the New York Botanical Garden from the Idaho slope of the Bitterroots fully confirms this view. The specimens cited differ from the widespread and perhaps too variable A. decumbens var. decumbens ( Nutt.) Cron. in the quality of the pubescence which, while still medially at- tached, is composed in large part of fine, twisted and woolly hairs, especially on the pod. That var. crispatus should intergrade with the typical variety is to be expected, and a third collection from near Leadore (Christ & Ward No. 14799, NY), though best referred to var. decumbens, shows a decided tendency in this direction ; for the pubescence is noticeably less appressed than in the prevalent form of var. decumbens, yet lacks the villous quality characteristic of var. crispatus. New NAME For A NEEDLEGRASS. Stipa pulchra Hitchce. var. cernua (Stebbins & Love) Beetle & Tofsrud, comb. nov. S. cernua Stebbins & Love, Madrofio 6: 137 (1941). Stipa pul- chra Hitche. and S. lepida Hitche. are of equal date (Amer. Jour. Bot. 2:301—303,—1915). They were separated from their Mexican relatives (1. e., nearest representatives within the genus ) with which they had been confused nomenclatorily. On many characters, S. lepida (glumes 6—10 mm. long, lemma 3—5 mm. long, awn 2—4 cm. long) has proven to be wholly distinct from S. pulchra (glumes 14—18 mm. long, lemma 6—10 mm. long, awn 5—8 cm. long). On the other hand, the recently described S. ceruna differs from S. pulchra only in having a slightly shorter lemma and a thinner awn with a long, flexuous terminal seg- ment.—A. A. Beetle and R. B. Tofsrud, Division of Agronomy, University of California, Davis, California. SuN SpuRGE IN SAN Francisco. In November, 1945, Miss Alice Mulford discovered Euphorbia Helioscopia L. in San Fran- cisco, California, on the southeastern slopes of Twin Peaks. It was found to be well established in a deserted vegetable garden where the rounded tops of the yellow-green plants were con- spicuous among other weeds. Heretofore this Old World spurge has been reported in California from Mendocino and Los Angeles counties.—J. T. Howell. 36 LEAFLETS OF WESTERN BOTANY __ [VOL. V, NO. 2 A NEW VARIETY IN OROBANCHE BY GEORGE J. GOODMAN University of Oklahoma, Norman Orobanche fasciculata Nutt. var. subulata Goodman, var. nov., speciei similis sed calycis lobis 6—8 mm. longis, 2-plo longioribus quam tubus. Specimens examined. OKLAHOMA: top of butte, 8 miles south of Watonga, Blaine Co., Apr. 19, 1935, Goodman No. 2375 (type, Bebb Herbarium, University of Oklahoma; isotype, N. Y. Bot. Gard.). TExas: bare limestone prairie, 1% mile northwest of Forestburg, Montague Co., Mar. 20, 1928, W. A. Bridwell (N. Y. Bot. Gard.). ’ A specimen from Arizona (Mormon Lake, Coconino Co., alt. 6,000 ft., June 9, 1898, D. T. MacDougal No. 78) in the New York Botanical Garden Herbarium I would place here, although Achey (Bull. Torr. Bot. Club 60:449,—1933) cites the same number in the Gray Herbarium as var. lutea. The only specimen of O. fasciculata that I have seen from Oklahoma is the one mentioned above. It might be added that this calyx character breaks down one of the key characters frequently used in separating O. fasciculata and O. uniflora. STUDIES IN CAREX—II BY JOHN THOMAS HOWELL In August, 1944, just east of the summit of Mono Pass in Mono County, California, I collected a specimen of a low mono- stachous Carex which superficially resembled C. subnigricans Stacey, but which, it occurred to me, might perhaps belong to some other species since it grew on a rocky slope instead of in a meadow where that species is generally found. Perhaps, I thought, it might be C. Engelmannti Bailey which had been re- ported by Mackenzie from the southern Sierra Nevada in Tulare County (Erythea 8:18,—1922). As it has turned out, this par- ticular collection is merely a depauperate sterile specimen of C. Breweri Boott, of little or no consequence in itself, but, in de- termining it, several interesting facts have been discovered: (1) that C. subnigricans is properly a member of section Inflate instead of section Callistachys where it was originally referred ; (2) that C. Engelmannii does not occur in California; and (3) that C. subnigricans, originally believed to be restricted to the ~ APRIL, 1947] STUDIES IN CAREX 37 Sierra Nevada, has a widespread but highly interrupted distri- bution in the western United States. Stacey originally reported C. subnigricans as C. pyrenaica, a new record for California (Leafl. West. Bot. 1: 177), and later, when he recognized the California plant as a new species (ibid., 2:167), he still maintained that it belonged to section Callis- tachys: “superficially C. subnigricans resembles C. pyrenaica, but is stoloniferous instead of densely cespitose, and is probably more closely related to C. nigricans” (ibid., 2: 168, 169). Nearly all critical characters of C. subnigricans, however, indi- cate a much closer relationship to section Jnflate than to section Callistachys. In the latter section, the species are essentially cespitose or short-stoloniferous (although C. nigricans is some- times prolonged-stoloniferous ) while C. subnigricans is stolonifer- ous as is characteristic of species in section Inflate. In section Inflate and in C. subnigricans the leaf-blade is involute-filiform, but in section Callistachys the leaf-blade is flat or channeled even when very narrow. In C. subnigricans the pistillate scales are tenaciously persistent, whereas in section Callistachys they are early deciduous. Although in size and shape the perigynia of C. subnigricans are more like those found in section Callistachys, the perigynia in C. subnigricans are more membranaceous in texture and they do not spread or reflex at maturity. More- over there is not the least indication of a dorsal suture on the perigynium-beak in C. subnigricans and the suture is generally apparent in section Callistachys. Finally there is a conspicuous rachilla developed in C. subnigricans that usually equals or ex- ceeds the achene in length, but in section Callistachys this remark- able vestigial structure is very short if it is present at all. With the transfer of C. subnigricans to section Inflate, it is of interest to note differences between C. subnigricans and the two species heretofore recognized in the section, C. Breweri and C. Engelmannii.* The differences between the three are readily apparent in the following synoptical key: * Another species has recently been proposed for the section, C. rachillis Maguire (Brittonia 5:199,—1944), but, after a careful study of part of the type-collection, I believe that the plant is not specifically distinct from C. subnigricans. It differs in a few details (e. g., only occasionally does the callous spot develop back of the ligules, the rachilla is often shorter than the achene, the achene is scarcely so plump, etc.), but these differences might be expected in material from such widely separated stations as Mt. Dana in California and Gilbert Peak in Utah, the type-localities of C. subnigricans and @. rachillis respectively. The reference of OC. rachillis to section Inflate by Dr. Maguire offers unexpected corroboration to my conclusion on the proper position of OC. swbnigricans. 38 LEAFLETS OF WESTERN BOTANY _ [VOL. V, NO. 2 1. Leaf-sheath narrow, not strongly nerved ventrally, generally mark- edly oblique at mouth, leaf-blade without callous spot back of ligule ; spikes narrowly to broadly ovate; pistillate scales gradu- ally narrowed to acute apex, midrib generally reaching apex; perigynium rounded at base, subsessile or shortly stipitate ; style slender, dark brown; achene dull, not plumply obovate, apiculate, much smaller than perigynium. 2. Ligule indurate-thickened or somewhat corky, spikes 4—¥Y% staminate; pistillate scales 3-nerved; achene narrow-obo- vate, cuneate at base; rachilla rather stout, about equaling achene, sometimes jointed and bracteate.................... C. Brewers 2. Ligule not indurate; spikes 4—%%4 staminate; pistillate scales 1-nerved; achene elliptic-obovate (i. e., a little widened above middle) ; rachilla setiform, exceeding achene................ WS NC cee yer te Ree NL nebs npn Mn eicame aOR AES tebe Le C. Engelmannit 1. Leaf-sheath larger and looser, rather conspicuously nerved, ventrally subtruncate at mouth, leaf-blade frequently developing a spongy callosity back of ligule; spikes lanceolate to oblong-lanceolate, 1424 staminate; pistillate scales 1-nerved, elliptic-ovate, obtuse or subacute, the midrib usually not reaching the hyaline apex; perigynium narrowed at base, rarely a little rounded, strongly stipitate ; style stouter, brownish-black ; rachilla stout, attenuate upward, green, generally equaling achene or a little longer; achene shining, plumply obovate, obtuse with or without a slight apiculation, nearly filling and distending the perigynium PEs HORS Uae AN Se PUT i NF OAS ge I BENE oo ALL NY AS Se BB C. subnigricans As soon as it was clear that C. subnigricans belonged to sec- tion Inflate, I suspected that the specimens cited as C. Engel- mannu from California by Mackenzie (Erythea 8: 18,—1922) would probably be referable to Stacey’s species. An examination of the four cited specimens in the Dudley Herbarium discloses that three are C. subnigricans, but the fourth, Dudley No. 2447 from Kokopo Creek, Tulare County, is apparently a specimen of C. Breweri with sterile depauperate spikes like those in my speci- mens from Mono Pass. No collections referable to C. Engel- mannii have been seen from California, the southern limit of that species being on “alpine summits of the Cascade Mts.” in Oregon (acc. Peck Man. Higher Plants Ore., p. 149). Carex subnigricans was originally believed to be “a relic, existing only in a restricted range in eastern California and western Nevada” (Stacey, Leafl. West. Bot. 2: 169), but it has now been found in Oregon, Idaho, and Utah. Its known distri- bution aligns it with other Californian plants about which I have APRIL, 1947 ] STUDIES IN CAREX 39 written which recur in the Wallowa Mts. of Oregon or the moun- tains of Idaho (cf. Amer. Midl. Nat. 33: 485, 486; Leafl. West. Bot. 4: 168—170) ; and Hitchcock and Thompson, in reporting the discovery of Lewisia Kelloggii K. Brandg. in Idaho have added another significant plant to the growing list (cf. Leafl. West. Bot. 4:197). Only in recent years has C. submigricans been collected extensively anywhere, so it may be premature to make statements concerning its regional abundance, but it would appear that, although it has been found to be one of the common and characteristic species in subalpine meadows of the middle and southern Sierra Nevada, it seems to be rare and relictual beyond the Sierra Nevada. It is most common in meadows and on moist rocky slopes about timberline in the Canadian and Hud- sonian Life zones at elevations ranging from 8000 to 11,000 ft., but occasionally it occurs as high as 12,500 ft. in the Arctic-Alpine Zone. The bibliographic and distributional data that have been as- sembled for C. subnigricans may be summarized as follows: CAREX SUBNIGRICANS Stacey, Leafl. West. Bot. 2: 167 (1939) ; Her- mann in Contrib. toward FI. Nev., No. 17, p. 14 (1940) ; Cronquist, Madrofio 7:82 (1943) ; Davis, Contrib. toward a Fl. Idaho, No. 18, p. 33 (1945). C. pyrenaica: Stacey, Leafl. West. Bot. 1: 177 (1935) ; Hermann, 1. c., as to Nevada plant; Peck Man. Higher Plants of Ore. 149 (1941), in part; not Wahl. C. Engelmannit, Mackenzie references to California plants: Erythea 8:18 (1922) ; in Jepson FI. Calif. 1: 210 (1922) ; in Abrams III. Fl. Pac. States 1: 289 (1923) ; in Jepson Man. FI. Pl. Calif. 165 (1923) ; in N. Amer. FI. 18: 23 (1931) ; not Bailey. C. rachillis Maguire, Brittonia 5:199 (1944). Ca.irorniA, all collections except the first from the Sierra Nevada. Inyo County: head of Cottonwood Creek, White Mts., Duran No. 2589; Rock Creek Lake Basin, Peirson No. 10828, 11400, 12208; head of Lone Pine Canyon east of Mt. Muir, C. W. Sharsmith No. 3293, 3372. Tulare County: East Lake, J. T. Howell No. 15765: between Reflection Lake and Harrison Pass, J. T. Howell No. 15821, 15853; Little Five Lakes, J. T. Howell No. 17426, 17767 ; Chagoopa Plateau, C. W. Sharsmith No. 3828, J. T. Howell No. 17550; Crabtree Meadows, McCracken in 1936. Fresno County: Bullfrog Lake, J. T. Howell No. 16086. Madera County: Garnet Lake, J. T. Howell No. 16642. Tuolumne County: Mt. Dana, J. T. Howell No. 14519 (type), C. W. Sharsmith No. 2356; Gaylor Lakes, J. T. Howell No. 20389; Mono Pass, J. T. Howell No. 20617; Dog Lake, J. T. Howell No. 20043; Tuolumne Meadows, J. T. Howell No. 20054; Eliza- beth Lake, J. T. Howell in 1944; Kuna Peak, C. W. Sharsmith No. 2682. Mono County: Slate Creek Basin, Keck No. 4657, 4914; Mt. Dana, C. W. 40 LEAFLETS OF WESTERN BOTANY [ VOL. V, NO. 2 Sharsmith No. 102, J. T. Howell No. 20251, 20264. Tuolumne or Mono County: Sonora Pass, Eastwood & Howell No. 7527. Eldorado County: Haypress Meadow, J. T. Howell No. 21530.* Nevapa. Washoe County: Third Creek at 8500 ft., J. T. Howell No. 14136; near Mt. Rose, J. T. Howell No. 14154; southwest of Reno at 8900 ft., Henrichs No. 179. Orecon. Wallowa County: Peets Point, Wallowa Mts., Peck No. 17909. IpaHo. Blaine County: Devils Bedstead, Sawtooth Range, Thompson _ No. 13587. Custer County: Ryan Peak, Boulder Mts., Hitchcock & Muh- lick No. 10616. UtaH. Summit County: small basin west of Gilbert Peak, 11,300 ft., Uinta Mts., Maguire, Hobson & Maguire No. 14668 (type-collection of. C. rachillis). All collections cited are in the Herbarium of the California Academy of Sciences. ANOTHER RusH IN CALiFoRNIA. Undoubtedly the establish- ment of Juncus acuminatus Michx. in the central San Joaquin Valley, California, is to be regarded as an occurrence of rela- tively recent date, but we have no data as to whether we should consider the introduction as naturally achieved (as by birds, for example) or as accidentally effected (as through the instru- mentality of man). According to Mrs. Ethel Hyde, University of California, Davis, who sent the specimen for determination, the plant was collected “in irrigated pastures about 5 miles north of Merced,” Nelson in 1940. Dr. F. J. Hermann, critical student of American Glumiflore, writes: “I was very glad to get the California collection of Juncus acuminatus. I have no other Cali- fornia record for the species . . ., although there are quite a few from Arizona, Oregon, and, of course, Washington.” What I have regarded for several years as Californian speci- mens of J. acuminatus are in Herb. Calif. Acad. Sci. from Weaverville, Trinity County, Kleeberger in 1879, and from Pudding Creek, Mendocino County, Mason in 1929.—John Thomas Howell. | * Because C. subnigricans is so distinctive vegetatively among the steno- phyllous sedges of California, I have little hesitancy in referring to that species a fragmentary specimen consisting only of leaves and rootstock made on Desert Creek, 10,000 ft., Sweetwater Mts., Mono Co., Alexander & Kellogg No. 45814. This is a noteworthy collection since it is only the second from California outside of the Sierra Nevada, the other being Duran’s from the White Mts. Another notable plant common to the White and Sweetwater eae is Polemonium chartacewm Mason (cf. Leafil. West. Bot. 3: 256, Two other recent California collections may also be noted here: Fourth Recess of Mono Creek at 11,000 ft., Fresno Co., J. T. Howell No. 22607 ; slope northeast of Gilmore Lake, 8,800 ft., Eldorado Co., J. T. Howell No. 22958. Vou. V No. 3 LEAFLETS of : WESTERN BOTANY CONTENTS PAGE Marin County Miscellany—IV . . . . .. . 4/3 JoHN Tuomas Howe. Private Herbariums Donated to the California Academy Te SUR, BON an SR Auice Eastwoop Additions to the Hepaticae of California . . . . . 48 DorotHy SUTLIFFE A New Aster from the High Sierra Nevada . . . . S5O Cari W. SHARSMITH Notes on the Flora of Montana ... i ual or aka ee C. Leo Hitcucock AND C. V. eee ema A New Species of Agrostis from California . . . . 56 Jason R. SwWALLEN This number published with funds from the California Botanical Club SAN FRrANcIsco, CALIFORNIA Jury 11, 1947 : NEW Y¢ BOTANI QARDI LEAFLETS of WESTERN BOTANY A publication on the exotic flora of California and’on the native flora of western North America, appearing about four times each year. Subscription price, $1.00 annually; single numbers, 40c. Address: John Thomas Howell, California Academy of Sciences, Golden Gate Park, San Francisco, California. Cited as LEAFL. WEsT. Bort. AA A INCHES He Te eed ee ML Mamita pins! o> Owned and published by Atice Eastwoop and JoHN THomas Howk tr JULY, 1947] MARIN COUNTY MISCELLANY—IV 41 MARIN COUNTY MISCELLANY—IV BY JOHN THOMAS HOWELL SECOND SPRING IN THE BuRNED AREA Following the widespread fire in Marin County, California, in September, 1945, I watched with interest all signs of recovery during the spring of 1946 and reported what I saw in the Sierra Club Bulletin (Vol. 37, No. 7: 18—23). The rapidity with which the vegetation reclothed the burned slopes was most impressive, and now, after a second season of rainfall and growth, the speed of recovery seems no less phenomenal. Grassland has now effaced all effects of the burn and chaparral approaches normalcy and is again offering characteristic resistance to cross-country hikers, but the once-beautiful Douglas fir woodland is, as it will be for many years to come, an unhappy reminder of a most disastrous fire. Visits to both Carson Ridge and Mt. Tamalpais disclosed the chaparral shrubs flourishing either as vigorous crown-sprouts or as numerous seedlings. Already the crown-sprouts are blossom- ing in Adenostoma fasciculatum, Pickeringia montana, and Erio- dictyon californicum, while buds on Heteromeles arbutifolia promised flowers before midsummer. The manzanitas and oaks did not flower this season, but “embryonic panicles” are now formed on Arctostaphylos glandulosa for next spring’s flowers and six-foot sprouts of Quercus Wislizeni var. frutescens and Q. durata look ready for flowers and fruits. The numerous her- baceous annuals which covered the slopes and flats last year were very sparse or entirely lacking in 1947. Where there had been thousands of plants of Phacelia suaveolens in 1946, none or only a few appeared in 1947, and on another choice slope, last year’s Silene multinervia, Mimulus Rattanti, Antirrhinum Hookeri- anum, and Campanula angustiflora were this year entirely re- placed by weedy Senecio sylvaticus. Seedlings of trees and shrubs are flourishing nearly every- where. On the serpentine, Arctostaphylos montana and Ceano- thus Jepson abound and seedlings of Cupressus Sargentii are well established with 2 to 4 inches of growth. Beyond the serpen- tine, other species of Ceanothus and Arctostaphylos have abun- dantly seeded and many little plants of Arbutus Menziesii are Leaflets of Western Botany, Vol. V, pp. 41-56, July 11, 1947. BYViAINI QAR: 42 LEAFLETS OF WESTERN BOTANY [ VOL. V, NO. 3 growing in burned woodland. Seedlings of several shrubs are already in flower: Lupinus Douglasu var. fallax,* Ceanothus foliosus, and Diplacus aurantiacus were especially attractive. The lupine, which is usually an occasional member of the chaparral, was notably gregarious on gravelly slopes along the Simmons Trail between the Mountain Theatre and Barths Retreat and offered the finest Tamalpais flower show of the year. In late spring, the blue of the lupine was rivaled by golden masses of Eriophyllum lanatum which were widespread on slopes of hill and canyon. The fire-flower of the year, however, was that of the squaw grass, Xerophyllum tenax, which bloomed on the north side of the mountain and on the slope above Lily Lake. I prefer the local common name, “fire-lily,” because in our region the plants rarely flower except after fires, although farther north blossoms are numerous every year. Following the Mt. Tamalpais fire in 1929, the lily bloomed profusely the next spring, but this time the season of bloom occurred the second spring following the fire. It may be that the season of the fire determines the time of bloom- ing—in 1929 the fire was in late June, in 1945 in late September. Although the rainfall has been less than normal this year, the flowering stalks appeared unusually vigorous and handsome. The maximum height of stems is given by Jepson as 6 feet, but one stem on the north side of Mt. Tamalpais this year measured 8% feet. Certainly we have adequate compensation for the long flowerless intervals between fires when such beauty arises from the ashes of a burn. GopeT1a Notes Depending on whether the inflorescence is erect or nodding in bud, there are two types of Godetia related to G. amaena (Lehm.) G. Don in Marin County. Besides this difference the plants also exhibit differences in the size of flowers and in the time of blooming, those with the erect inflorescence having much larger flowers that blossom about a month later. Because of these apparent and rather fundamental differences, it seems desirable to distinguish the plants by name. * Lupinus Douglasii Agardh var. fallax (Greene) J. T. Howell, stat. nov. L. fallax Greene, Erythea 2: 119 (1894). This lupine of Mt. Tamalpais and the Douglas lupine of the Santa Lucia Mts. have usually been combined and treated as a variety of L. albifrons Benth. However, L. Douglasii dif- fers not only in its much longer floral bracts which exceed the flower-buds but also in its plumper, rounder seeds which are not laterally flattened. The Tamalpais variant may be distinguished from typical L. Douglasii by its much broader subobtuse leaflets. ‘ OE JULY, 1947] MARIN COUNTY MISCELLANY—IV 43 Recent genetic work on our coastal godetias has been reported by Gunnar Hiorth and I believe that his accounts (Ztschr. f. Induktive Abstam. u. Vererbungslehre 79: 199—219,—1941, and 80: 289—349,—1942) furnish us with the clue that indicates names for our Marin plants. According to Hiorth (op. cit. 79, p. 203 ; 80, p. 295), the plants of this relationship with erect inflo- rescence are referred to G. amana and G. Whitneyi (Gray) T. Moore and those with nodding inflorescence are named as new, G. nutans. Although Hiorth would divide G. amena and G. Whitneyi by the Golden Gate, the former to the south, the latter to the north, it does not seem desirable to alter the currently accepted floristic interpretation of these species, at least for the present. Hence, I am applying the name G. amena to the Marin plants with erect inflorescences, and not G. Whitneyi as Hiorth would do. According to Hiorth (after Hakansson), G. nutans with fourteen haploid chromosomes is genetically quite distinct from G. amena and G. Whitneyi each with seven chromosomes. This plant, so well marked by its drooping inflorescence, has a natural range extending through the North Coast Ranges of California to Roseburg, Oregon, and to the northern Sierra Nevada. As a taxonomic entity, however, Hiorth does not properly publish G. nutans—he gives no formal description and indicates no type, so at best the name is subnudum and hence is not acceptable. Fortunately, there is a validly published specific name that is available, G. lassenensis Eastw. (Leafl. West. Bot. 2:281,— 1940), a name applying to the Sierra Nevada plant but specifi- cally applicable to the one in Marin County. In typical G. lassenensis, however, the stigma-lobes are linear while in the Marin County plant the stigma-lobes are broadly elliptic to ovate. It is evident from C. L. Hitchcock’s revision (Bot. Gaz. 89: 338, 342) and from Jepson’s more recent treat- ment (FI. Calif. 2:578) that our plant is referable to G. amcana var. concolor Jeps. (Fl. W. Mid. Calif. 334,—1901) which may now be called G. lassenensis var. concolor (Jeps.) J. T. Howell, comb. nov. In Marin County I have it from Mt. Tamalpais, Fairfax Hills, Carson Ridge, Lagunitas Canyon, and San Rafael Hills. Altogether Hiorth recognizes five varieties or subspecies in his “nutans” complex, but var. concolor is not among those indi- 44 LEAFLETS OF WESTERN BOTANY _ [VOL. V, NO. 3 cated unless it is allied to the small-flowered forms discussed from Humboldt County, California, or southern Oregon, but not named. Two large-flowered forms, one from the northern Sacra- mento Valley and adjacent Sierra Nevada, the other from the Coast Ranges north of San Francisco Bay, have been previously named as varieties of G. ame@na and may now be transferred to G. lassenensis: Godetia lassenensis var. albicaulis (Jeps.) J. T. Howell, comb. nov. G. amena var. albicaulis Jeps., Univ. Calif. Publ. Bot. 2: 329 (1907). Godetia lassenensis var. sonomensis (C. L. Hitchc.) J. T. Howell, comb. nov. G. amena var. sonomensis C. L. Hitche., Bot. Gaz. 89:338 (1930). MorRNING-GLORY OBSERVATIONS Among the different kinds of Convolvulus which I have found in Marin County, one is intermediate in habit between the tufted and nearly stemless species and those with elongate twining stems. Because the peduncles are sometimes 2-flowered and the leaves somewhat resemble those depicted in Dr. Abrams’ photograph of C. occidentalis Gray (Contrib. Dudley Herb. vol. 3, pl. 66), I thought I might have specimens of Gray’s species which was based on a collection by Dr. Gibbons from “near San Francisco.” However, since my collection seemed specifically different from the usual Marin County form of C. purpuratus (Greene) Greene which Dr. Abrams treated as a synonym of C. occidentalis (op. cit., p. 353), I felt it necessary to examine the type-specimen which was loaned to me by Dr. Lyman B. Smith from the Gray Herbarium. From a study of this specimen I have decided that my her- baceous trailer is not C. occidentalis but is referable to C. poly- morphus Greene. This species is not uncommon in northern California and adjacent Oregon, but before this it has not been reported south of Lake County in the California Coast Ranges. Heretofore more southern collections may have been confused with caulescent forms of C. subacaulis (H. & A.) Greene, but in that the more sepaloid floral bracts closely subtend the calyx while in our Marin County plants the more foliaceous bracts are smaller and distinctly removed from the base of the calyx. After a study of the type of C. occidentalis, I am in accord with Dr. Abrams’ decision to refer it to that variable group of JULY, 1947] PRIVATE HERBARIUMS 45 California morning-glories which have been called C. luteolus Gray or C. purpuratus (Greene) Greene, a complex which, if accepted as a single polymorphic species, must be now known as C. occidentalis Gray, the oldest acceptable name. Dr. Gibbons’ specimen, although marked by unusually ample broad-lobed leaves and by slender ovate-lanceolate acuminate-tipped sepals, seems to correspond most closely to Greene’s segregate C. fruti- cetorum and here I am accepting it as belonging to that variant. This typical variant of C. occidentalis, with stems, leaves, inflo- rescence, and sepals more or less puberulent, has not been found in Marin County, but two glabrous variants have, those de- scribed as C. luteolus var. purpuratus Greene (with leaf-lobes acute and bracts entire), and C. saxicola Eastw. (with leaf-lobes obtuse or subacute and bracts lobed or entire). These may be renamed as follows: Convolvulus occidentalis var. purpuratus (Greene) J. T. Howell, comb. nov. C. luteolus Gray var. purpuratus Greene Man. Bot. Reg. S. F. Bay 265 (1894) ; C. purpuratus (Greene) Greene, Pitt. 3:332 (1898). Convolvulus occidentalis var. saxicola (Eastw.) J. T. Howell, comb. nov. C. saxicola Eastw., Bull. Torr. Bot. Club 30:495 (1903). C. luteolus Gray var. saxicola (Eastw.) Jeps. Man. FI. Pl. Calif. 779 (1925) ; C. purpuratus (Greene) Greene var. saxicola (Eastw.) Jeps. Fl. Calif. 3:124 (1939). PRIVATE HERBARIUMS DONATED TO THE CALIFORNIA ACADEMY OF SCIENCES BY ALICE EASTWOOD In the annual reports of the Botanical Department of the Cali- fornia Academy of Sciences all donations to the herbarium have been recorded since 1912, the year of the reéstablishment of the department after its destruction in 1906. Some small herbariums by amateur botanists have been given and no records have ever been published. It seems desirable that these should be rescued from oblivion because of especially important specimens which they contained and brought to the herbarium. The first was that of Miss Evelina Cannon, who died some years before the fire and left her herbarium to her niece, Miss Carrington. She donated it to the California Academy of Sciences 46 LEAFLETS OF WESTERN BOTANY __ [VOL. V, NO. 3 soon after the fire. Miss Cannon had been a successful music teacher in New York City and had come to San Francisco to be with her niece. She loved plants as well as music and had been a member of the Torrey Botanical Club. It was about the time when the California Botanical Club was founded that Miss Cannon came to San Francisco and she became one of its most enthusiastic members. The native flowers were then so abundant and all so new to her that she spent her spare time roaming around the city and collecting for her herbarium. In the list of species found in San Francisco, published in Zoe in 1892, Miss Cannon contrib- uted many. Among them was the rare Sanicula maritima Kell., originally found in Alameda and at that time extinct. She col- lected it ina marshy place in the Potrero about where the City and County Hospital of San Francisco now covers the ground. I was delighted to find in her herbarium a specimen of Cypripedium californicum Gray which I collected in one of the many densely wooded gulches on the south side of Mt. Tamalpais in May, 1898. My specimen went with all the others in the 1906 fire, but here it was preserved as I had given it to Miss Cannon because of her interest in orchids. It has again been collected probably in the same locality by another hardy explorer who gave the specimen to Miss Ethel Wickes, the artist, and the picture is among her collection of water color paintings of the native flowers of Cali- fornia. In Jepson’s Flora, the collection of the plant is incorrectly credited to the artist. I have forgotten the name of the collector and have been unable to find the reference. San Francisco is the type-locality of many species, especially those collected and named by Chamisso and Eschscholtz. Representatives of all were in our herbarium as well as all other species known to have been here. Ours were all destroyed so that the only specimens we now or ever will possess are in Miss Cannon’s herbarium. Helianthella Can- none Eastwood was named in her honor. No specimens can now be found in the Presidio, the type-locality, nor in the Bay View Hills in South San Francisco where it also flourished. When W. W. Carruth, the well-known organist, was a boy, he often came to the old herbarium before the fire to assure himself of the identification of the flowers which he collected for his her- barium. He collected chiefly in Oakland where he now lives and where most of the wild flowers have long since been destroyed by the growth of the city. Arctostaphylos pallida Eastwood, which JULY, 1947] PRIVATE HERBARIUMS 47 still can be found on the hills back of Piedmont, was described from one of his specimens. He had a small collection of the flowers of orchids from the greenhouse of A. H. Hills, the senior member and one of the founders of Hills Brothers Coffee firm. Mr. Hills was an enthusiastic collector of orchids and Will Carruth had per- mission to collect the flowers, almost all of which were named. He gave up botany for music and, after a course in Paris and at Yale University where he graduated as Bachelor of Music, he is now on the faculty of Mills College and organist at Sherith Israel Synagogue and First Church of Christian Science, San Francisco. In 1922, Mr. George R. Kleeberger gave his herbarium to the Academy. It had been stored for many years in San Jose. He collected in Weaverville and the Yollo Bolly Mountains, Trinity County, California, in 1880, and among his specimens from the mountains is the first collection of Oreobroma Heckneri Morton not published until recent years. This species is one of the love- liest of all the species known to present-day gardeners as lewisias. In our herbarium we have several specimens of the species from localities in Trinity County. Amsinckia grandiflora was named by him and published by Asa Gray as a variety of A. vernicosa H. & A. Both are members of the group of Amsinckia with smooth nutlets. The type of A. vernicosa collected by Douglas in the Santa Lucia Mountains is a small-flowered species quite unlike the beautiful large-flowered A. grandiflora. In Suksdorf’s publi- cation on Amsinckia it is credited to Kleeberger as a species. While at Yale University studying under Prof. D. C. Eaton, he obtained some specimens of plants collected on the Clarence King Expedition, also some specimens from a set collected and distrib- uted by Kellogg and Harford. Among these was a specimen of Eriogonum Kelloggti Gray collected by Kellogg on Red Moun- tain, Mendocino County, near the border of Humboldt County. I collected it there many years ago but my specimens were de- stroyed and this isotype is the only one that we now have. While at Yale he became interested in mosses and hepatics and pur- chased the herbarium of Coe F. Austin, who named many species of Hepatice, specimens which are in the Kleeberger herbarium and probably the types. The E. K. Abbott herbarium was donated after his death by his widow. Dr. Abbott was a well-known physician of Salinas and botany was his hobby. His collections came chiefly from 48 LEAFLETS OF WESTERN BOTANY __[VOL. V, NO. 3 the Salinas Valley, the adjacent Santa Lucia Mountains, and the Monterey region, the last the type-locality of plants collected by Menzies, Douglas, Coulter, Hartweg, and others. What a welcome accession it was to our small herbarium! Malvastrum Abbottu Eastwood, which Dr. Abbott collected in 1889 among willows along the Salinas River, was named in his honor. A col- lection from France, probably obtained by exchange, brought the first botanical specimens from Europe to our herbarium. That they came from the Valley of the Marne, where the Germans in World War I were first successfully repulsed, enhanced their value to me. The above herbariums were acquired many years ago but this very year we have received over four thousand mounted speci- mens in the herbarium of Mrs. H. C. Cantelow. It comprises mostly Californian plants but also includes many from western America and Alaska. Among the valuable specimens are the first- collected plants of Phacelia Dalesiana J. T. Howell and isotypes of Lewisia Cantelovii J.T. Howell and ‘Lupinus Dales@ Eastwood. ADDITIONS TO THE HEPATIC OF CALIFORNIA BY DOROTHY SUTLIFFE While connected with the Department of Botany, University of California, Dr. M. A. Howe published his excellent treatise on the Hepatice and Anthocerotes of California (Mem. Torr. Club vol. 7, 1899). Between that time and the year 1920, interest in this fascinating subject seemed to lag in California. In the latter year the collection at the California Academy of Sciences was begun and since then we have added twenty-one of these plants to those previously known within our limits. We now have in our collection about 4000 specimens which include all the species known to be growing in California. We also have some of Austin’s type-specimens which came to the Academy in the Kleeberger Herbarium in 1922. Additions to the formerly known California hepatics made by workers and correspondents of the California Academy of Sciences are given in the following list. All determinations have been made or confirmed by our leading American hepaticologists. Riccia BryricHIANA Hampe. Salmon Lake, Sierra Co., October, 1927, Sutliffe No. 76. Jury, 1947] HEPATIC OF CALIFORNIA 49 Riccia FLuITANS L, Lily Lake, Marin Co., September, 1921, Sutliffe No. 27. ASTERELLA SACCATA (Wahl.) Evans was collected in July, 1946, at Long Lake, Rock Creek Lake Basin, Inyo Co., at an elevation of 10,600 ft., by John Thomas Howell, No. 704. Several species of Asterella are fairly common in California, but this is the first report of A. saccata in the state. MARSUPELLA SPHACELATA (Giescke) Dum. var. ERYTHRO- RHIzA (Limpr.) Schiffn. East base of Banner Peak, Madera Co., Aug. 5, 1941, Howell No. 564. JUNGERMANNIA LANCEOLATA Schrad. Lower Salmon Lake, Sierra Co., October, 1921, Sutliffe. JUNGERMANNIA RIPARIA Tayl. Lower Salmon Lake, Sierra Co., October, 1921, Sutliffe. JUNGERMANNIA SCHIFFNERI (Loitl.) Evans. Plate Flat near Willoughby Mine, Sierra Co., October, 1927, Sutliffe No. 76. Narpia GEOscyPHA (De Not.) Lindb. Water Wheel Trail down Tuolumne River, Tuolumne Co., August, 1928, Sutliffe No. 8. 4 NarpiA scaLaris (Schrad.) S. F. Gray. Big Lagoon, Hum- boldt Co., July 3, 1936, Howell No. 388. LrEIocoLEA GILLMANI (Aust.) Evans. East Lake, Tulare Co., August 7, 1940, Howell No. 518. LEIOCOLEA OBTUSA (Lindb.) Buch. Plate Flat, near Wil- loughby Mine, Sierra Co., October, 1927, Sutliffe No. 76. This was re-collected by Mr. Howell, No. 586, near Badger Lake, Madera Co., Aug. 2, 1941. LopHoziA ALPESTRIS (Schleich.) Evans. High Sierra above Yosemite Valley, 1923, Michaels No. 67. LopuHozia ExcisA (Dicks.) Dum. Lake Lagunitas, Marin Co., March 7, 1922, Mrs. M. L. Campbell. LorpHoz1a HornscHUCHIANA (Nees) Schiffn, Lower Salmon Lake, October, 1921, Sutliffe. DIPLOPHYLLUM OBTUSIFOLIUM ( Hook.) Dum. Monumental, Del Norte Co., 1923, Eastwood No. 68. CrepHALoziA LAMMERSIANA (Hiutb.) Spruce. Gasquet Valley, Del Norte Co., Apr. 11, 1934, Howell No. 3106. CEPHALOZIELLA STELLULIFERA (Tayl.) Schiffn. var. GRACIL- LIMA Douin. Lake Lagunitas, Marin Co., Feb. 22, 1922, Mrs. M. L. Campbell No. 54. 50 LEAFLETS OF WESTERN BOTANY __ [VOL. V, NO. 3 ANTHELIA JULACEA (Lindb.) Dum. Between Iceberg and Ediza lakes, Madera Co., Aug. 6, 1941, Howell No. 585. PLEUROCLADA ALBESCENS ( Hook.) Spruce. Above Reflection Lake, Tulare Co., Aug. 2, 1940, Howell No. 510. ScAPANIA BarTLiINnGII (Hampe) Nees. Glen Aulin Camp, Tuolumne Co., Aug., 1928, Sutliffe No. 13. SCAPANIA GRANULIFERA Evans. Miss Eastwood’s new spe- cies, collected in 1923 in Del Norte Co. A NEW ASTER FROM THE HIGH SIERRA NEVADA BY CARL W. SHARSMITH Aster Peirsonii C. W. Sharsmith, spec. nov. Herba perennis, dense cespitosa, 2—6 cm. alta; folia basalia plurima, firma, linearia acutaque, plana vel plerumque conduplicata, 1.5—4 cm. longa, 1—2.5 mm. lata, glabra vel sparse glanduloso-scabridula; caules floriferi 1.5—5.5 cm. alti, sparse vel dense glanduloso-scabriduli monocephali; folia caulina 3—6, 1.5—2.5 cm. longa; involucrum turbinatum vel hemisphericum, 7—11 mm. altum, bracteis in 3 seriebus instructis, omnibus maturo disco brevioribus, lanceo- latis vel lineari-lanceolatis, glanduloso-scabridulis, presertim exterioribus, numquam pubescentibus; flores radii 8—18, ligulis 14—18 mm. longis, 1.5—2.5 mm. latis, ceruleis vel quasi purpureis; flores disci 26—48, corolla disci 5.5—8 mm. longa, paulatim sursum versus ampliata, infra sparse pubescente; styli rami longi exserti, 2—2.8 mm. longi, appendicibus usque duplo longioribus quam pars stigmatica, attenuatis, quasi acutis; achenia teretia vel leviter compressa, 10-nervata, 4—4.5 mm. longa, infra plerumque glabra, supra sparse pubescentia vel pene glabra; pappi sete plerumque sordide, 8—9 mm. longe. Densely cespitose perennial 2—6 (—8.5) cm. high with few to several erect or assurgent stems from tufts of crowded basal leaves on branches of a stout erect taproot; root-crown roughened and black with persistent remnants of dead leaves ; basal leaves firm, linear, acute, flat to usually con- duplicate, the veins, especially the lateral, usually not evident, 1.5—4 (—5) cm. long (average about 2.5 cm. long), 1—2.5 (—3) mm. wide (the narrower conduplicate leaves teretish, as little as 0.5 mm. wide), glabrous to sparsely glandular-scaberulous; floriferous stems 1.5—5.5 (—8) cm. high, sparsely to densely glandular-scaberulous, usually anthocyanous, each terminated by a solitary head, the cauline leaves like the basal, 3—6, sessile, 1.5—2.5 cm. long, reduced upward, or all much reduced and the stem scape- like ; involucre turbinate to hemispheric, 7—11 mm. high, 10—18 mm. wide (pressed) ; involucral bracts in about 3 series, somewhat recurved at tips (fresh), the outer about half shorter than the innermost, all shorter than the mature disc, lanceolate to lance-linear, mucronately acute to acute to soft-acuminate, glandular-scaberulous, especially the outer, never pubescent, — the outer green or anthocyanous with green tips, the inner anthocyanous throughout or above the base ; ray-flowers 8—18, the ligules (boiled) 14—18 JULY, 1947] A NEW ASTER 51 mm. long, 1.5—2.5 mm. wide, sky-blue to violet or purplish; disc-flowers 26—48 ; disc-corolla 5.5—8 mm. long, sparsely pubescent below, gradually ampliate upward, lobes deltoid to ovate, 0.5—1 mm. long, puberulent ; style- branches long-exserted, 2—2.8 mm. long, the slender and acutish appendages from slightly longer to twice as long as the stigmatic portion; achenes terete to slightly compressed, usually 10-nerved, 4—4.5 mm. long, at least lower half usually glabrous, above sparsely pubescent or puberulent to almost glabrous ; pappus 8—9 mm. long, the 25—40 slender bristles white or usually sordid. Distribution. Subalpine and alpine, Sierra Nevada in Tulare ' County, Inyo County, and southern Fresno County, California. Inhabiting meadows and granitic gravels in glaciated and un- glaciated terrane ; infrequent to occasional in occurrence. Specimens examined. (These unless otherwise indicated are deposited in the University of California Herbarium.) Fresno County : northwest base of University Peak by highest of Kear- sarge Lakes, Kearsarge Pass region, altitude 11,200 ft., August 16, 1937, Sharsmith No. 3253 (type). Inyo County: cirque near Consultation Lake, Lone Pine Canyon, 12,200 ft., Sharsmith No. 3272 ; cirque east of Mt. Muir, Lone Pine Canyon, 12,200 ft., Sharsmith No. 3299; Cottonwood Lakes, 11,000 ft., Alexander & Kellogg No. 3318. TuLArE County : between Reflection Lake and Harrison Pass, 12,000 ft., Howell No. 15869 (Herb. Calif. Acad. Sci.) ; headwaters of Kern River, 2 miles above Milestone Creek, 11,500 ft., Bacigalupi No. 1755 (Dudley Herb. Stanford Univ.) ; vicinity of Mt. Whitney, Dudley No. 2477 (in part only, Dudley Herb. Stanford Univ.) ; Mt. Whitney region, August, 1938, Jessie Saunders (Herb. Calif. Acad. Sci.) ; north slope Cirque Peak above Army Pass, 12,250 ft., Sharsmith No. 3396; summit of Boreal Plateau, southwest of Siberian Outpost, 11,400 ft., Sharsmith No. 3431. The close relationship of Aster Peirsonii to A. alpigenus (T. & G.) A. Gray is obvious. It possesses in common with A. alpigenus the monocephalous inflorescence, glabrous achenes (or only sparsely pubescent above), and slender stylar ap- pendages up to twice as long as the stigmatic portion. It differs markedly from both A. alpigenus and A. alpigenus subsp. Ander- soni (A. Gray) Onno in the glandular scaberulous involucral bracts and peduncle, and much shorter, narrower, commonly con- duplicate, often glandular-scaberulous leaves. This very dis- tinctive species is named in honor of Frank W. Peirson, keen observer and student of the high Sierra Nevada flora. 52 LEAFLETS OF WESTERN BOTANY [ VOL. V, NO. 3 NOTES ON THE FLORA OF MONTANA BY C. LEO HITCHCOCK AND C. V. MUHLICK University of Washington, Seattle The writers spent the period from June 24 to August 24, 1945, in botanical exploration in Montana where collecting was done chiefly in the Big Snowy, Belt, Castle, Crazy, Absaroka, Pioneer, and Beaverhead mountains. Among the 2400 numbers obtained, several plants were apparently collected in Montana for the first time. The Little Belt Mountains, a small range to the north of White Sulphur Springs, Meagher County, proved to be the most interesting region visited. These mountains are rather isolated from adjacent ranges, and, although their higher peaks extend but little above 9000 feet, the limestone which outcrops in many places supports some of the rarer plants of the state. We were fortunate in getting into the Little Belts a few days in advance of the many bands of sheep which are grazed there during July and August, and we therefore found the vegetation undisturbed. The high ridge extending east of King’s Hill to Yogo Peak was a blaze of color, due, in part, to the abundance of such rarities as Aquilegia Jonesu Parry (Hitchcock & Muh- lick No. 12332), Eritrichium elongatum (Rydb.) Wight (No. 12333), Astragalus aboriginum Richards. (No. 12338), and Pedicularis Hallii Rydb. (No. 12352). The last was especially plentiful throughout the subalpine meadows of the entire range and was encountered in several other ranges of the state as well (cf. Hitchcock & Thompson, Leafl. West. Bot. 4: 204,—1945). On Rimrock Ridge, one of the larger cliffs of limestone on the southwestern edge of the range, Kelseya uniflora (Wats.) Rydb. was found in fair amount (No. 12234). This plant was originally described from material collected at Gate of the Moun- tains, some distance north and west of the Little Belts. So far as I know, it has previously been known in Montana only from the type-locality, and in Idaho only from the Lost River Moun- tains, another limestone range. The plants were still in flower, and it was obvious that the petals are pink rather than white, as Rydberg and others have supposed them to be. Another unusual find in the Little Belts was Corallorhiza Wisteriana Conrad. This orchid apparently has never been re- JULY, 1947] NOTES ON THE FLORA OF MONTANA 53 ported as occurring in Montana, although Williams (Am. Midl. Nat. 18: 841,—1937) records it for Colorado, but for no other western state. We found it in considerable quantity in Douglas fir forest just below Rimrock Ridge, No. 12218, and about one mile southwest of William’s Mountain, No. 12179. Rydberg stated that Viola orbiculata Geyer was “occasional in the western part of the State,” and cited a specimen from Silver Bow County. We found this violet in the woods one mile southwest of William’s Mountain, No. 12176. Although Rydberg (Fl. Rocky Mts. 649,—1917) listed Drosace albertina Rydb. for Montana, Robbins (Am. Midl. Nat. 32: 147,—1944) reduced Rydberg’s species to synonymy under Androsace Lehmanniana Spreng., giving as the distribution “". . northern Rocky Mountains, principally Alberta (and Mon- tana?).” He cited no specimens from Montana. We collected the plant in Half Moon Canyon, Big Snowy Mountains, Fergus County, No. 11947 and 12040, so there need be no question of its presence in Montana. Another apparent record for the state obtained in the Big Snowy Mountains was Asplenium viride Huds., No. 11946, on mossy limestone rocks above stream three miles from mouth of Half Moon Canyon. Townsendia montana Jones was not included in the flora of Montana by Rydberg or by Blankinship, and Larsen (Ann. Mo. Bot. Gard. 14:17,—1927) cited no specimens from the state. We found it in abundance in the Pioneer Range, Beaverhead County, on the limestone ridge connecting Sheep and Black Lion mountains, No. 13003. In this same general locality we also found Gentiana propinqua Richards. (small meadow at south base of Sheep Mountain, No. 12887) and Parnassia Kotzebuei C. & S. (on moss near snow bank on the northeast slope of Black Lion Mountain, No. 12919). I can find no record of the former plant in the state, whereas the Parnassia is listed for Montana by Blankinship (Mont. Agri. Coll. Stud. 1:62,—1904) only from one collection made on the Gallatin River. Two collections of Campanula uniflora L. were secured : one, in the Pioneer Mountains at Lake Waukena, head of Rock Creek, No. 13063 ; the other in the southern end of the Bitterroot Range (or Beaverhead Range as it is listed on most maps) on the Conti- 54 LEAFLETS OF WESTERN BOTANY _ [VOL. V, NO. 3 nental Divide above Ajax Lake, Beaverhead County, No. 12705. Rydberg (op. cit., 823) gives as the range of this species “Greenl.-Lab.-Colo.-Utah-Alaska,” and I have found no record of its having been found in Montana. Other apparent new state records include: Angelica pinnata Wats., meadowland two miles southeast of Cooke City, No. 13610 (identified by Lincoln Constance) ; Penstemon humilis Nutt. ex Gray, one-fourth mile below Ajax Mill, Bitterroot (Beaverhead) Range, Beaverhead County, No. 12655 (identified by D. D. Keck) ; Mimulus breviflorus Piper, on moss-covered rocks three miles west of Pintlar Lake, Anaconda Range, Beaverhead County, No. 12792; and Alchemilla pratensis Schmidt, about four miles north of Castle City, eastern slope of Castle Mountains, Meagher County, No. 12081. Two very obnoxious weeds that apparently are due to spread unchecked in the state are Euphorbia Esula L. (one infestation of several acres is to be found five miles west of Ovando, Powell County, No. 11544) and Hypericum perforatum L. A fairly large colony of this most undesirable pest was observed on the eastern edge of the Bison Range, two miles southwest of St. Ignatius. Since the plants were long past flowering, no speci~ mens were preserved. Erigeron flabellifolius Rydb. was collected on a talus slide in the Beartooth Mountains, ten miles north of pass on Red Lodge- Cooke City highway, near the Wyoming-Montana border, No. 13516. According to Arthur Cronquist, who verified the determination, our plants represent the second known collection of the species. The following willows, all identified by C. R. Ball, were reported as new to the state or otherwise noteworthy: Salix melanopsis Nutt. var. tenerrima (Henderson) Ball, on gravel bars of Blackfoot River eight miles west of Lincoln, Powell County, No. 11615; S. monochroma Ball, in wet meadow along Checkerboard Creek, northeast base of Castle Mountain, Meagher County, No. 12139; S. brachycarpa Nutt. var. Sansoni Ball, head of first stream to south of Sheep Creek, Pioneer Range, Beaver- head County, No. 12886; and S. Dodgeana Rydb., in moist spots along limestone ridge connecting Sheep and Black Lion moun- tains, Pioneer Range, No. 12985. JuLy, 1947] ALIEN ASTRAGALUS 55 In 1944 we made a collection of Castilleja (No. 10498) on the divide between Alpine Creek drainage and Twin Lakes, Saw- tooth Primitive Area, Blaine County, Idaho. We noted that there were two forms of the plant, one red-bracted and the other yellow- bracted, and that all intermediate degrees were manifest. A sheet of this material was sent to Dr. Marion Ownbey, who identified the specimens he received as C. lutea Heller and C. rhexifolia Rydb., but questioned any intergradation. Accordingly, the entire collection was sent him and he agrees that there is con- clusive evidence that these two dissimilar species hybridized, pro- ducing the “hybrid swarm” to be expected from such a cross. The collection was large enough to furnish two complete series, one of which is at Washington State College. AN ALIEN ASTRAGALUS IN WASHINGTON In the August, 1946, issue of these Leaflets (4:279), Mr. R. C. Barneby calls attention to what he regards the first intro- duction of Astragalus falcatus Lamk. into Washington. It is quite probable that the specimen cited from Pullman and collected by G. N. Jones in 1929 came from, or was an escape from, the U. S. Department of Agriculture Plant Introduction Garden at Pullman. The then Division of Forage Crops had a staff member there as early as about 1906 and seeds of A. falcatus were sent there on five successive years from 1906 to 1910. Back of 1906, records of the distribution of seed of the species are not now available, but its first introduction was S. P. I. Number 1448 in December, 1898. Between 1898 and 1929 at least fifteen intro- ductions of the seed were made and quite a number of seed samples were sent for experimental purposes to several states. Taxonomists should not overlook the fact that nearly 160,000 introductions of plants and seeds have been made from all over the world and a great many times that number of samples have been distributed for experimental purposes. It is probable that very few of these have escaped from cultivation in such numbers as to warrant their being in a local flora—P. L. Ricker, Plant Industry Station, Beltsville, Maryland. 56 LEAFLETS OF WESTERN BOTANY __[VOL. V, NO. 3 A NEW SPECIES OF AGROSTIS FROM CALIFORNIA BY JASON R. SWALLEN U. S. National Museum, Washington, D. C. John Thomas Howell recently sent to the National Herbarium for verification several specimens of an Agrostis he collected on Point Reyes Peninsula which appeared to be a new species. Com- parison with the related species of Agrostis shows that it was pre- viously unknown, and on the generous request of Mr. Howell it is described below. Agrostis aristiglumis Swallen, spec. nov. Annua; culmi erecti, 5—15 cm. alti, glabri; vagine internodiis longiores, glabre; ligula 2—2.5 mm. longa, decurrens, scabridula; lamine plane, 1.5—6 cm. longz, 0.5—2.3 mm. late, scabre ; panicule 3—6 cm. longe, 5—8 mm. latz, densz, ramis inferi- oribus ad 2 cm. longis, basi nudis ; glumz acuminate, 5—6 mm. longe, prima 1-nervis, secunda 3-nervis, aristis 1—2 mm. longis; lemma 3.2—3.5 mm. longum, firmum, scabrum, callo barbato pilis 0.5 mm. longis, nervis minute excurrentibus, dorso aristatum, arista 6—7 mm. longa, geniculata, scabra, infra geniculam tortili; palea hyalina, enervis, 1—1.2 mm. longa; antherz circa 1 mm. longe. Annual; culms 5—15 cm. high, erect, glabrous; sheaths all longer than the internodes, glabrous; ligule 2—2.5 mm. long, decurrent, scaberulous ; blades 1.5—6 cm. long, 0.5—2.3 mm. wide, flat, scabrous; panicles 3—6 cm. long, 5—8 mm. wide, dense, the branches appressed, the lowermost as much as 2 cm. long, the longer ones naked toward the base; glumes 5—6 mm. long including the awns, these 1—2 mm. long, acuminate, scabrous, the first 1-nerved, the second 3-nerved ; lemma 3.2—3.5 mm. long, rather firm, scab- rous, the callus bearded on the sides, the hairs 0.5 mm. long, the nerves excurrent in short scabrous rather delicate awns as much as 0.5 mm. long, awned from the back just above the middle, the awn 6—7 mm. long, genicu- late, twisted below the bend, scabrous ; palea well developed, broad, hyaline, nerveless, 1—1.2 mm. long; anthers about 1 mm. long. Type, U. S. National Herbarium No. 1869693, collected on slope above Drakes Estero, west of Mt. Vision, Point Reyes Peninsula, Marin County, California, May 14, 1947, by John Thomas Howell (No. 23149). Duplicate type in the California Academy of Sciences Herbarium. Agrostis aristiglumis is allied to A. microphylla and A. Hen- dersoni, having awned glumes and the nerves of the lemma ex- current. It differs from both of these, however, in the presence of a well-developed palea and the firm scabrous lemma. According to Mr. Howell “it is restricted to a slope of loose gravelly soil on an outcrop of diatomaceous shale of the Monterey series” asso- ciated with Agrostis Blasdalei, Festuca dertonensis, and Aira caryophyllea. ts wd? « 3 VoL. V No. 4 LEAFLETS of WESTERN BOTANY CONTENTS PAGE A New Penstemon from Wyoming . . . . . . S57 Daviw D. Keck Beem regon oaxkiirages: 6. i ee a A SS Morton E. Peck Distributional Notes and Minor Novelties . . . . 61 R, C. BARNEBY Further Studies of Broad-Leaf Erodium . . . . . 67 Joun THomas Howe. Additional Notes on the Grass F igh in Marin County, California P PRUE Dane Lm an Cm tenn « JoHN THomas Howe. SAN FrRANcIScO, CALIFORNIA OctTosBer 31, 1947 LEAFLETS of WESTERN BOTANY A publication on the exotic flora of California and on the native flora of western North America, appearing about four times each year. Subscription price, $1.00 annually; single numbers, 40c. Address: John Thomas Howell, California Academy of Sciences, Golden Gate Park, San Francisco, California. Cited as LEAFL. WEsT. Bort. AU ee pritgun INCHES pvvavanyvonapaagyovuepnagycoeaen agony vagy canga toga en anny Owned and published by Auicge Eastwoop and JoHNnN THomas Howektui iJ Te ae fae OcToseEr, 1947 ] A NEW PENSTEMON 57 A NEW PENSTEMON FROM WYOMING BY DAVID D. KECK Carnegie Institution of Washington, Stanford University, California Penstemon Paysoniorum Keck, spec. nov. Caulibus pluribus, 8—15 (—25) cm. altis, e caudice ramoso, undique glaberrimis ; foliis crassis, mar- gine plus minusve crispo, imis lineari-oblanceolatis, 3—6 cm. longis, 3—8 mm. latis, petiolatis, obtusis, superis similibus, lanceolatis, sessilibus, in inflorescentia ad bracteas abrupte reductis; thyrso multifloro, secundo, 4—10 cm. longo; calyce 5—7 mm. longo, segmentis ovato-lanceolatis, inferne scarioso-marginatis, superne ad apicem acuminatum gradatim reductis; corolla 15—20 mm. longa, cceruleo-purpurea, paulatim ampliata, obscure bilabiata, glaberrima ; loculis antherz late divaricatis, anguste oblongis, de- hiscentibus ab apice fere ad basim, 1.3—1.6 mm. longis, moderate brevi- terque albo-pilosis ; filamento sterili parce flavo-barbato in dimidio supero. This plant grows in southwestern Wyoming on sandy creek bottoms, alkaline shale bluffs, and dry hills, among sagebrush at altitudes of 6500 to 7500 feet. It flowers in June. The following collections define it. Sublette Co. : between Eden and Big Piney, E. B. & L. B. Payson No. 2572 (Ph, RM?). Lincoln Co.: dry hillsides between Opal and Kemmerer, June 19, 1923, Payson & Armstrong No. 3221 (RM, type; isotypes, Ph, US); 10 miles east of Opal, Ripley & Barneby No. 7863 (CAS); Flat Iron Butte, Rollins No. 2354 (CI, SU). Uinta Co.: 6 miles north of Ft. Bridger, Rollins No. 2314 (CI, SU); Ft. Bridger, 1873, Carter (Ph) ; LeRoy, A. Nelson No. 4585 (RM) ; east of Ragan, Ripley & Barneby No. 7837 (CAS) ; Cottonwood Creek, 10 miles north of Lonetree, Rollins No. 2799 (CI, RM, SU). A single plant from 5 miles east of Green River, Sweetwater County, at 6200 feet, Ripley & Barneby No. 7894 (CAS) belongs here although its appearance is radically different because its leaf- blades are elliptic to oval and up to 14 mm. wide. This small-statured, small-flowered member of the Glabri is amply distinct from its nearest relatives. In habit and shape of flower it resembles P. Fremontii T. & G. of the same general region, but that species is densely cinereous-puberulent through- out, with shorter-tipped sepals and differently shaped anthers, 1 The abbreviations apply to the following herberia: CAS, California Acad- emy of Sciences; CI, Carnegie Institution of Washington, Stanford Univer- sity; Ph, Academy of Natural Sciences of Philadelphia; RM, Rocky Mountain Herbarium, University of Wyoming; SU, Dudley Herbarium, Stanford Uni- versity ; US, United States National Herbarium, Washington. Leaflets of Western Botany, Vol. V. pp. 57-72, October 31, 1947. 58 LEAFLETS OF WESTERN BOTANY [ VOL. V, NO. 4 the cells being directly opposite and united by a long line of con- tact. Penstemon Paysoniorum is less clearly related to P. cyan- anthus Hook. subsp. subglaber (Gray) Penn., its neighbor to the north and to which it might seem to key, for that is a larger plant in every way and it lacks the much-branched caudex that gives Paysonioruan its tufted habit. Edwin Blake Payson, professor of botany at the University of Wyoming, whose death on May 15, 1927, in his thirty-fourth year cut short a singularly promising and productive career in systematic botany, was accompanied on many field trips by his wife, Lois B. Payson, and their names appear together on the labels of some widely distributed sets of beautifully prepared specimens. Payson collected and studied the present plant from the desert region of southwestern Wyoming. In his report “The species of Penstemon native to Wyoming” (Univ. Wyo. Publ. Bot. 1:88—103,—1924) he discussed its possible relationship with P. Caryi Penn., a species of the Big Horn Mountains known at that time very imperfectly from the type-specimen only. Penstemon Caryi differs in several characters, but above all its flowers are significantly larger and of a different shape, with a much more ample throat and limb. In the herbarium, Payson tentatively proposed to name the present plant in honor of his wife, Lois, but the name was never published. It now seems appropriate to honor the botanical industry of both Edwin and Lois Payson in the name of this beautiful little plant. CERTAIN OREGON SAXIFRAGES BY MORTON E. PECK Willamette University, Salem, Oregon The northwestern saxifrages have always presented consider- able taxonomic difficulties to students of the group, and there has been great diversity of opinion as to definable specific entities. Dr. Rimo Bacigalupi’s treatment of the genus in Vol. 2 of Abrams’ Illustrated Flora is the most satisfactory that has yet appeared and has cleared up much of the confusion, but so far as Oregon is concerned there are still some minor questions. The following observations may be worth recording: Saxifraga fragosa Suksd. subsp. claytoniefolia (Canby ) Baci- galupi seems to have little if any taxonomic value. We have ex- OcToBER, 1947 | OREGON SAXIFRAGES 59 amined numerous specimens from the type-locality and adjacent areas. Some are good claytoniefolia while others growing beside them are quite indistinguishable from typical fragosa and to- gether with the two there are many intergrading examples. On the other hand, we occasionally find the claytoniefolia form even as far south as Jackson County. There seems no very good reason for not recognizing S. Halli Johnson as a specific entity readily distinguishable from S. Mar- shallu Greene. The former has apparently a limited and rather sharply defined distributional range, occurring plentifully along the Willamette River and about Oswego Lake in northwestern Clackamas County, thence to the foothills and western slopes of the Cascades at moderate altitude and as far south as central Marion County. We have seen no specimens of S. Marshallii from this area, though it is found a little farther to the west in Washington and Polk counties. Saxifraga Halli has usually narrower and less deeply indented leaves, uniformly broader and shorter calyx-lobes, and filaments narrowly linear, only occasion- ally slightly dilated upward. This is in strong contrast to the broadly clavate or spatulate filaments with extremely slender bases that characterize S. Marshallit. On the rugged walls of Snake River Canyon in eastern Wallowa County occurs a variant, evidently, of S. Marshallii presenting several well-marked features that set it off from the typical phase of the species. This variant we may designate and define as Saxifraga Marshallii Greene var. divaricata Peck, var. nov. Planta plus minusve purpureo-tincta; foliis plerumque basi cuneatis; ramis inflo- rescentize valde divaricatis ; filamentis sursum paulum dilatatis. Herbage, including the inflorescence, more or less strongly suffused with purple; leaves mostly distinctly cuneate at base; branches of the panicle strongly divaricate; filaments only moderately dilated upward. Type: Peck No. 18158, on wet cliff of Snake River Canyon near the mouth of Battle Creek, Wallowa County, Oregon, March 28, 1934. A single specimen (Peck No. 17449) from Hilgard, western Union County, seems quite intermediate between the new variety and true S. Marshallii. Several closely related, puzzling forms of at best subspecific value have been lumped together by Dr. Bacigalupi as S. occi- dentalis Wats. subsp. rufidula (Small) Bacigalupi. These are 60 LEAFLETS OF WESTERN BOTANY [ VOL. V, NO. 4 Micranthes rufidula Small, M. Allenti Small, M. equidentata Small, and S. klickitatensis Johnson. This is probably the best way of disposing of them, though in some cases minor differences, more or less stable, can be pointed out. True S. occidentalis is a Rocky Mountain species not known from Oregon, but reaching northeastern Washington ; the variants are reported mainly from the Cascade Mts. and westward. Micranthes equidentata might well be given recognition as a named subspecies or variety on its own merits. We have seen it only from the Columbia Gorge, where its large, flat-topped, corymbiform panicles of relatively large flowers make it very conspicuous. It is the handsomest of our native saxifrages. A well-marked variant of S. occidentalis, as we understand it, occurs in the Wallowa Mts., which can scarcely be associated with the forms above mentioned. This may be designated and characterized as Saxifraga occidentalis Wats. var. wallowensis Peck, var. nov. Planta matura valde rubro-tincta; marginibus foliorum minus crasse crenatis, subter sparse fusco-tomentosis ; inflorescentia 2—6-ramosa, ramis elongatis suberectis, cymulis congestis, pedicellis calyce brevioribus ; filamentis an- guste clavatis. Plant at maturity deeply tinged with dark red; leaves finely and evenly crenately toothed, sparingly tomentose beneath with rusty hairs; flowers in 2—6 compact clusters, the pedicels not lengthening, the lower clusters on elongated, closely ascending branches of the scape; filaments narrowly clavate. Type: Peck No. 18542, from moist slope above Ice Lake, Wallowa Mts., Wallowa County, Oregon, July 4, 1934. The variety differs from M. rufidula in its much less coarsely crenate leaves, only sparsely rusty-tomentose beneath, in its few dense cymules on elongated strict branches of the scape, and in the uniformly barely clavate filaments. A New NAME IN THE GENUs SENEcIo. My attention has been called by Dr. S. F’. Blake to a publication of the name Senecio revolutus by T. Kirk in Student’s Flora of New Zealand, p. 348, 1899. Senecio revolutus Hoover (Leafl. West. Bot. 3:256,— 1943) may therefore be called Senecio pattersonensis Hoover, nom. nov., a name derived from Mt. Patterson, the type-locality of the species in the Sweetwater Mts., Mono County, California.— Robert F. Hoover, California Polytechnic College, San Luis Obispo. OcTOBER, 1947] NOTES AND MINOR NOVELTIES 61 DISTRIBUTIONAL NOTES AND MINOR NOVELTIES BY R. C. BARNEBY Wappingers Falls, New York Records extending the known distribution of a number of species have turned up during the determination of collections lately obtained by Mr. H. D. Ripley and the writer in the western States, and are reported in the following notes. To these have been added a few minor nomenclatural changes and infraspecific novelties. The collection-numbers cited are those of Ripley and Barneby, and duplicates have been deposited in the herbarium of the California Academy of Sciences. ERIOGONUM DIVARICATUM Hook. Nevada: calcareous foot- hills of the Schell Creek Range, 12 miles south of Major’s, White Pine Co., No. 6316. The specimens are immature and depauper- ate, but the pubescence, form of the flower, and distinctive axil- lary foliage indicate this species, known otherwise from northern Arizona, Colorado west of the Rockies, and Utah. ERIOGONUM VILLIFLORUM Gray. Nevada: Monitor Valley near Lone Mountain, Eureka Co., No. 6226; Ely, White Pine Co., No. 6301; Cathedral Gorge, Lincoln Co., No. 6345. The species has been reported once from eastern Nevada by J. T. Howell (Leafl. West. Bot. 3: 187,—1942), but in reality it is fairly common in calcareous valleys of the region, where it is often associated with E. Shockleyi subsp. candidum (Jones) Stokes. LEPIDIUM NANUM Wats. Nevada: Monitor Valley, west of Eureka, Eureka Co., No. 6204; north of Hamilton, White Pine Co., No. 6267; Sunnyside, Nye Co., No. 3611. To the four collections of this singular plant known to C. L. Hitchcock at the time of his revision of Lepidium, three additional stations were recently added by Maguire and Holmgren (Leafl. West. Bot. 3: 12,—1941). Since all the reported localities are in Elko Co., those noted here mark a considerable extension of range to the south and southwest, the last station lying 120 miles south of the Elko County line. PsORALEA LANCEOLATA Pursh. California: sandy fields near Doyle, Lassen Co., No. 5952. Widely distributed in the interior from Saskatchewan to Texas, northern Arizona and central Nevada, this rather variable species is omitted from the Cali- 62 LEAFLETS OF WESTERN BOTANY [VOL. V, NO. 4 fornia flora by Jepson. Our specimens are in flower only, so that the precise variety to which they are to be referred remains doubt- ful; since, however, the var. scabra (Nutt.) Piper is the prevail- ing form in the adjacent Nevada, it is probable that the California plant will prove to be the same. PETERIA THoMpPpsoN#& Wats. P. nevadensis Tidestr., Proc. Biol. Soc. Wash. 35: 183 (1923). Utah: southern Utah (prob- ably near Kanab), Mrs. Thompson in 1872 (New York Bot. Gard., isotype!) ; Springdale, Washington Co., No. 4932 ; Green- river, Emery Co., Jones in 1890 (N. Y. B. G.). Nevada: Las Vegas, Clark Co., Tidestrom No. 9083 (N. Y. B. G., isotype of P. nevadensis!) ; foothills of Pancake Range, near Silverton, Nye Co., No. 3647; Alkali Hot Springs, Esmeralda Co. (vidi!). Idaho: Hot Springs, east of Bruneau, Owyhee Co., No. 6511. Study of the cited collections shows that the characters by which Tidestrom segregated his new species from P. Thompson@ are not diagnostic. It is true that plants from southern Utah, grow- ing on detrital banks below sandstone cliffs, are taller than the prevalent Nevada form, which is commonly found along dry stony washes in basaltic soil ; but in the isotype of P. Thompsone the stem is only 23 cm. long, about average for the species, sensu ampliato. The flower is variable throughout the range, tending to become larger northward; certainly it is no smaller than in the Nevada plant, as claimed, the isotype of P. nevadensis itself bearing corollas of almost exactly the same size as in Mrs. Thomp- son’s original gathering. The calyx-teeth, described by Tide- strom as wider in P. Thompsone, are, in the isotype, scarcely half as wide (at least at base) as in typical P. nevadensis, while their length and width fluctuate widely in our collections. Simi- larly variation in the degree to which the glandular hairs on the calyx descend along the axis of the raceme (in general less northwards), in the length of the legume (4—7 cm.), and in the size of the leaflets, occurs here and there without correlation between any pair of factors; so that it seems impossible to main- tain P. nevadensis even as a variety. Up to the present P. Thompsone@ has been reported only from southern Utah and ad- jacent Arizona, and P. nevadensis only from the type-locality. The Idaho station extends the range of the species some 400 miles northward. OcTOBER, 1947] NOTES AND MINOR NOVELTIES 63 Linum Kincu Wats. var. sEpompEs Porter. Nevada: cal- careous knolls in Duck Valley, north of Pioche, Lincoln Co., No. 6346. This perennial flax, long known only from open pine- woods of the Wasatch Plateau, has been reported from Elko Co., Nevada, by Maguire and Holmgren (loc. cit.), who remark that it occurs there in a dry desert habitat. In Lincoln Co., 150 miles to the south and at an altitude not above 5600 feet, it ap- pears to be equally at home on slopes of clay mixed with fine limestone grit. Mentzelia Torreyi Gray var. acerosa (Jones) Barneby, comb. nov. M. acerosa Jones, Cont. West. Bot. 17:30 (1930). Idaho: Bruneau, Owyhee Co., Jones No. 25020 (type, Pomona College Herb.!) ; southwest of Hot Springs, Owyhee Co., No. 6515 ; southern Idaho, P. H. Hawkins, (N. Y. Bot. Gard.). Ment- zelia acerosa, which closely resembles M. Torreyi in habit, was wholly reduced to it by Darlington in her monograph of the genus (Ann. Mo. Bot. Gard. 21 : 158,—1934). It differs, however, in the size and color of the flower, and, considering its geographic isola- tion, can well be maintained as a variety. In typical MW. Torrey, as I have seen it in Churchill and Lander counties, Nevada (approxi- mate topotypes), in Jones’ specimens from Muncy, Nevada, and in several collections from eastern California, the petals are clear yellow and about 1 cm. long; while in the plant from the canyons of the Bruneau River the petals are scarcely half as long and of a curious burnt orange. Darlington mistakenly described the petals as about 5 mm. long for the whole species, and this error was re- peated by McVaugh in the Loasacez of Nevada (Cont. Fl. Nev. 27: 11,—1941); Jepson’s measurements (FI. Calif. 2: 528,— 1938), ““4—5 lines long,” are correct for typical M. Torreyi. CENOTHERA MINOR (A. Nels.) Munz. California: valley below Cedarville, Modoc Co., No. 6006. Although widely dis- tributed over the sterile volcanic plains of northern Nevada and transmontane Oregon, this species has apparently not been col- lected hitherto in California. Gilia micromeria Gray, reported by Jepson (FI. Calif. 3: 194,—1943) with some doubt from a single station in Inyo Co., also enters the state at this same point, occurring in some quantity on dunes in association with the CEnothera. 64 LEAFLETS OF WESTERN BOTANY _ [ VOL. V, NO. 4 LoMATIUM LATILOBUM (Rydb.) Math. Colorado: on cliffs, with Aquilegia micrantha Eastw., Colorado National Monument, southeast of Grand Junction, Mesa Co., No. 5443. Reported by Mathias and Constance (N. Amer. Fl. 28B: 253,—1945) only from southeastern Utah. Neither specimens nor the published descriptions do justice to the size attained by this striking um- bellifer, plants of which, composed of many densly cespitose crowns, form great pendent masses on shady ledges of the cliffs. SWERTIA MODOCENSIs St. John (?). Nevada: mountains west of Vya, northern Washoe Co., No. 6016. This is a pubes- cent plant, closely allied to, and perhaps best considered merely as a form of, S. albicaulis Dougl. A similar species has been col- lected in central Nevada (cf. Leafl. West. Bot. 4: 10,—1944), so that, although St. John reported no member of this alliance from the state, it seems likely that the S. albicaulis aggregate is more widely distributed in Nevada than has been supposed. LEPTODACTYLON c#sPITosuM Nutt. Nevada: clay knolls in Monitor Valley, 28 miles west of Eureka, Eureka Co., No. 6211; Monte Neva Hot Springs, White Pine Co., No. 6280. This species was first collected by Nuttall near Scott’s Bluffs, Nebraska, and has since been reported from Wyoming and Utah, in the latter state reaching as far south as the head of the Sevier River (south of Hatch, Garfield Co., No. 4794). The Nevada stations repre- sent a range-extension of some 200 miles to the west. It has been noted that the leaves, constantly described in the literature as alternate, are in reality opposite, the upper pairs often con- nate at the base and forming a series of imbricated cups investing the solitary terminal calyx. The tetramerous corolla, fawn-col- ored rather than “yellowish,” is vespertine, opening an hour or so after sunset (when it exhales a strong perfume of hyacinths) and closing in the early morning. On the brilliant white knolls of Monitor Valley the larger plants form dense pulvinate domes two feet in diameter, and in company with Lepidium nanum Wats., Eriogonum Shockleyi subsp. candidum (Jones) Stokes, Oxytropis oreophila Gray, a species of Phlox, and Erigeron compactus Blake form the most remarkable association of aretioid or vegetable-sheep life-forms known to me in western America. GILIA TENERRIMA Gray. Nevada: Pinto Summit, Eureka Co., No. 6245. Tidestrom (Fl. Nev. Utah 437,—1925) did not OcToBER, 1947] NOTES AND MINOR NOVELTIES 65 report Gilia tenerrima from Nevada, where it seems, notwith- standing, to be rather common in the mountains at middle eleva- tions. Holmgren (Handb. Vasc. Pl. N.-E. Nev. 143,—1942) states that it is frequent in the Ruby Mountains. It has also been found as far west as the Warner Mountains of southern Oregon (No. 6059), and is to be expected in the Californian extension of that range. Scutellaria nana Gray var. sapphirina Barneby, var. nov., a forma typica corolla saturate coerulea, nec ochroleuca, recedens. Nevada: clay banks in the foothills of the White Pine Mts., west of Little Antelope, White Pine Co., No. 6261. Type in Herb. Calif. Acad. Sci., No. 329680. In his monograph of Scutel- laria, Epling (Univ. Calif. Pub. Bot. 20:30,—1942) cites three collections of Scutellaria nana with blue flowers from three paral- lel ranges in eastern Nevada, the White Pine, Schell Creek, and Egan mountains. His distributional map (No. 7, op. cit. p. 28) clearly shows the isolation of this form from the rest of the species-range, and there seems no reason why it should not be accepted as a named variety. The typical form, ranging from northwestern and northern Nevada through southern Oregon to Idaho, and differing from the calciphile var. sapphirina only in the color of the flower, is confined, so far as I have observed it, to stiff soils overlying basaltic substrata. NEMACLADUS RIGIDUS Curran. Idaho: 5 miles southwest of Hot Springs, Owyhee Co., No. 6518. This, the relatively rare typical form, has been known only from northwestern Nevada and adjacent California and Oregon. We have it also from as far east in Nevada as the Toyabe Mts. (10 miles east of Austin, Lander Co., No. 6201) where, as in Idaho, it was associated with Eatonella nivea Gray on open gravel slopes. TOWNSENDIA INCANA Nutt. Nevada: Needle Mts., Lincoln Co., No. 6418; Cathedral Gorge, Lincoln Co., No. 6344. The range of Townsendia incana, as currently reported, is from Wyoming southward to Arizona and New Mexico. Like several other plants of similar distribution, it enters Nevada along the valley of the Muddy River, a region which phytogeographically has more in common with adjacent Utah than with the limestone mountains of southern Nevada immediately to the west. 66 LEAFLETS OF WESTERN BOTANY _ [ VOL. V, NO. 4 DimereEsIA HoweELi Gray. Nevada: steep gravel slide in the mountains 7 miles west of Vya, northern Washoe Co., No. 6009. Peck (Man. Pl. Oreg. 743) has described the corollas of this peculiar annual composite as having a purple tube and “light yellow” lobes ; in our specimens the florets were pure white in early anthesis, the tube only becoming purplish in age. A few periclinia were observed to be composed of three florets and three bracts, not constantly two of each as described, and the generic character should therefore be emended to this extent. EATONELLA NIVEA (D. C. Eat.) Gray. Idaho: 5 miles south- west of Hot Springs, Owyhee Co., No. 6517. A new record for Idaho. Eatonella nivea is primarily a species of the transmontane foothills of the Sierra Nevada, ranging from Inyo Co., California, to Lake Co., Oregon, and sporadically eastwards in Nevada at least to Eureka Co. (Beowawe, acc. Homgren, Handb. N.-E. Nev. 186), the Reese River Valley (north of Austin, vidi!), and the Toquima Range (Round Mountain, Nye Co., No. 3666). CHANACTIS MACRANTHA D. C. Eat. Idaho: Bruneau, Owy- hee Co., No. 6494. Known from southern Arizona and interior southern California north to Harney and Malheur counties, Oregon, but apparently not previously reported from Idaho. PSATHYROTES PILIFERA Gray. Nevada: clay hills west of Glendale, Clark Co., No. 6370. Described originally from speci- mens obtained near Kanab, Utah, this plant has since been col- lected only along the Colorado and Virgin rivers in southern Utah and northern Arizona. Like several other species of like distrib- tional pattern, it recurs in the valley of the Muddy River in southern Nevada. The rarity of the species may be less real than apparent, owing to its late period of flowering (June to September) at very low altitudes. LYGODESMIA GRANDIFLORA (Nutt.) T. & G. Nevada: bare clay hills in the Needle Mts., Lincoln Co., No. 4650. Apparently a new record for the state. GLYPTOPLEURA MARGINATA D. C. Eat. Idaho: southwest of Hot Springs, Owyhee Co., No. 6516. Not known previously from Idaho. Southern Oregon, throughout central Nevada to transmontane California, in low sandy, usually alkaline, valleys. OcTosER, 1947 ] BROAD-LEAF ERODIUM 67 FURTHER STUDIES OF BROAD-LEAF ERODIUM BY JOHN THOMAS HOWELL In 1943, Wagnon and Biswell (Madrofio 7: 118-125, 3 fig.) presented detailed taxonomic and range data on the two types of broad-leaf filaree that are found in California and identified them as Erodium Botrys (Cav.) Bertol. and E. Botrys f£. mon- tanum Brumhard. Although they noted several differences in foliage and flowers between the two, the differences in the fruits were quite impressive: “beak (rostrum) 9.5—12.5 cm. long; fovea surrounded by two plicae .. .”’ for the species, and “beak (rostrum) 5.5—8.5 cm. long ; fovea surrounded by one plica . . .” for the form. The authors, having seen only Californian speci- mens of the form (which was originally described from Cali- fornian and Chilean plants), were confronted at the close of their paper by questions on the origin, nativity, and essential identity of the American plants. In the collection of Lewis S. Rose deposited in the Herbarium of the California Academy of Sciences is a specimen of Erodium from Morocco in which the fruits are exactly like those of the Californian plant referred by Wagnon and Biswell to f. mon- tanum. This African collection is labeled Erodium Botrys var. obtusiplicatum Maire, Weiller, & Wilczek, and, from reference to the original publication (Bull. Soc. Hist. Nat. Afr. Nord 26: 120,—1935), it appears to be an isotype of the variety. The authors give for the type-locality “in lapidosis arenaceis prope Khenifra,” while the datum on the specimen (ex herb. Weiller. ) reads “entre Kenifra (sic) et Raobr Ladla.” On page 196 the further data are given: “dans les paturages pierreux sur les gres au pied du Moyen Atlas pres de Khenifra vers 800 m. d’altitude.” Hence, it seems evident that the Californian plant of this relation- ship, like typical E. Botrys, was introduced from the Mediter- ranean region of the Old World and probably from North Africa. A further study of var. obtusiplicatum has convinced me that the entity is specifically distinct from E. Botrys. Not only do the two kinds of plants exhibit the difference of flowers, fruits, and cotyledons so clearly described and illustrated by Wagnon and Biswell, but there are yet further differences in the fovee and plicz of the two which should be noted. In E. Botrys, the 68 LEAFLETS OF WESTERN BOTANY [ VOL. V, NO. 4 fovezee are frequently extended upward into a short crest-like collar at the base of the beak and the margins of the fovea and of the concentric plice are usually chartaceous and more or less lacerate, while the sides of the sculptured top of the fruit are quite devoid of the hirsutulous trichomes that are so numerous below. In the other plant the fovea and plica are not crested or chartaceous-margined, but the trichomes are generally distributed to the base of the beak, along the margins of the fovea and plica, and sparsely even onto the face of the fovea. Since this distinctive _ plant has apparently not been recognized as a species heretofore, it may be known as Erodium obtusiplicatum (Maire, Weiller, & Wilczek) J. T. Howell, stat. nov. E. Botrys var. obtusipli- catum Maire, Weiller, & Wilczek, loc. cit.; E. Botrys f. mon- tanum Brumhard in Fedde Repert. Spec. Nov. 2:118 (1906). Not E. montanum Coss. & Dur. In relationship, E. obtusiplicatum lies between E. gruinum (L.) L’Hér. which it resembles in technical characters of the fruit, and E. Botrys, which it approaches in character of foliage and flowers. Erodium obtusiplicatum is represented in Herb. Calif. Acad. Sci. not only by the type-collection from Morocco but also by the following American specimens: OreEconN: Salem, Marion Co., Peck No. 13716. CALIFORNIA: Hamilton, Glenn Co., Heller No. 11842; Oroville, Butte Co., Heller No. 11201; Kelsey, Eldorado Co., Eastwood No. 14128; Ione, Amador Co., Eastwood No. 10120; Mokelumne Hill, Calaveras Co., Blais- dell; Mariposa Co., Congdon in 1892; 2 miles north of Napa, Napa Co., Kearney in 1947; Drakes Estero, Pt. Reyes Peninsula, Marin Co., J. T. Howell No. 23075; east end of Alpine Lake, Marin Co., J. T. Howell No. 23182; near Searsville Lake, San Mateo Co., J. T. Howell No. 23049; Ojai, Ventura Co., Pollard in 1945. Cute: Santiago, Philippi No. 272; Cerro de Renca, Herb. Mus. Nac. No. 51236; Conchali, Herb. Mus. Nac. No. 51235; Concon (Médano), near Valparaiso, Philippi No. 272a. These Chilean collections are represented in Herb. Calif. Acad. Sci. by fruits from the Herbarium of the Museo Nacional de Historia Natural in Santiago and were sent by Dr. Carlos Mujioz Pizarro, Chief of the Phanerogamic Section, in answer to my in- quiry concerning the forms of broad-leaf Erodium in Chile. The exami- nation of the fruits from Dr. Pizarro indicates clearly that both E. Botrys and E. obtusiplicatum occur in Chile. I am very grateful to him for his help in establishing this fact. OcTOoBER, 1947] GRASS FAMILY IN MARIN COUNTY 69 ADDITIONAL NOTES ON THE GRASS FAMILY IN MARIN COUNTY, CALIFORNIA BY JOHN THOMAS HOWELL STATIONS FOR THE BLASDALE BENTGRASS Agrostis Blasdalei Hitche., one of the rarest Californian grasses, was known as recently as 1935 (Hitchcock Manual, p. 334) only from the region of the type-locality near Fort Bragg, Mendocino County. More recently I have collected it in Marin County at several stations which, together with my field numbers, may be indicated as follows: Point Reyes Peninsula (near the lighthouse, No. 21954; dune hill near the radio station, No. 21466, 21952; rocks north of Abbotts Lagoon, No. 22006; Drakes Estero, No. 23150) ; rocks east of Dillons Beach, No. 22053. A New GENUS IN CALIFORNIA Another Old World grass, Brachypodium distachyon (L.) Beauv. from the Mediterranean region, has apparently come to make its home with us. Although Brachypodium is related to Festuca and Bromus, its species look like those of Agropyron because of the very short pedicels of the rather large spikelets. Our particular immigrant occurred as a dominant annual in open grassland on Tiburon Peninsula, and, judging from its show of competitive vigor, I believe it will become widespread in Cali- fornia. Heretofore in America it has been noted only as a ballast plant in Oregon and New Jersey (Hitchcock Manual, p. 57).* ANEMOPHILY IN THE ANNUAL FESCUES The annual and perennial fescues have, in addition to their distinctive habits, certain morphological and physiological charac- teristics which further separate them and which have been vari- ously interpreted as denoting either a generic or sectional division in the group. The perennials with 3-staminate wind-pollinated flowers have been regarded as typical of Festuca L. and the annuals with 1- or 3-staminate self-pollinated flowers have been separated as the genus ’ulpia Gmel. or Festuca sect. Vulpia (Gmel.) Reichenb. Although such floral differences are im- * After I had written this record of Brachypodium for Marin County, Mrs. M. K. Bellue told me of two other occurrences of the grass in California as shown by specimens in the California Department of Agriculture Herbarium in Sacramento. Both specimens were collected in 1941, one from southeast pd a nd es Sacramento County, the other from Pleasanton, Alameda ounty. 70 LEAFLETS OF WESTERN BOTANY [ VOL. V, NO. 4 pressive and especially so when supported by a distinctive habital character, the structure of the spikelets in both groups showed a relationship so close that the sectional rather than generic sepa- ration seemed the more natural to many workers and, therefore, the preferable taxonomic interpretation. Hence, in April, 1945, when Fernald (Rhodora 47:106—108) adopted the generic segregation of the two groups, I became much interested in the floristic aspect of the problem since among our fourteen Marin County fescues we have five that are perennial and nine annual (and of the annuals seven native and two Old World). I did not have long to wait before I had field evidence that at least some of the Pacific Coast Vulpie are frequently wind- pollinated and have three stamens. Rhodora No. 556 was re- ceived at the Academy library on April 23 and on May 6 follow- ing in the San Rafael Hills I observed my first anemophilous Vulpia, Festuca reflexa Buckl. This was promising, but it was not until the spring of 1946 that observations were extended from F, reflexa to three related species, F. Eastwoode Piper, F. Gray (Abrams) Piper, and F. pacifica Piper. On Mt. Tamalpais, on Carson Ridge, and on Tiburon Peninsula, wherever these species were observed at anthesis, florets with three large normal anthers could frequently be found. At such times the exposed stigmas probably could also have been seen, but it was not until the present year that exserted plumose stigmas were also observed in F, refleva in the San Rafael Hills and in F. pacifica in the Fairfax Hills. More careful examination in the herbarium has revealed that most of these plants produce two kinds of flowers, wind-pollinated in the terminal inflorescence and self-pollinated in the lateral or basal inflorescences. Also it has been seen that in one inflo- rescence the upper or outer flowers may have exposed anthers while the lower and inner flowers may have included anthers. The exposed anthers are 2—4 mm. long while the included an- thers are very tiny and less than 1 mm. long. The flowers in these species seem to be uniformly 3-staminate, although, when cleis- togamous, one anther only may be functional. While it is quite obvious from field evidence that the species related to F. reflexa are frequently more or less anemophilous, it is equally clear in the indigenous F. megalura Nutt. and F. octo- flora Walt. and in the introduced F. dertonensis (All.) Aschers. OcToBER, 1947] GRASS FAMILY IN MARIN COUNTY 71 & Graebn. and Ff. myuros L. that the florets are generally, if not always, cleistogamous. Since these species and others related to them are the ones best known to eastern American and European botanists, these fescues are undoubtedly the ones that have been most closely studied in the formulation of Vulpia as a genus. The partly anemophilous western American annuals are much too near to be excluded from those that are entirely cleistogamous, and once the annuals are viewed together, all must surely be regarded as Festuca. Certainly in Marin County, California, Festuca requires and will receive a broad Hackelian interpre- tation. A New ComBINATION IN THE GENUS TELESONIX. In con- nection with a projected treatment of the Saxifragace@ of Idaho, the entity Telesonix heucheriforme Rydb. seems best treated as only varietally distinct from Telesomix Jamesti (Torr.) Raf. The combination Telesonix Jamesii ( Torr.) Raf. var. heucheriforme (Rydb.) Bacigalupi, comb. nov., is therefore now proposed. The complete synonymy is as follows : Therofon heucheriforme Rydb., Bull. Torrey Club 24:247 (1897); Telesonix heucheriformis Rydb., N. Am. Fl. 22:126 (1905); Boykinia heucheriformis Rosend. in Engl. Bot. Jahrb. 37, Beibl. 83:64 (1905) ; Saxifraga heucheriformis M. E. Jones, Bull. Univ. Mont. 61:32 (1910) ; Boykinia Jamesii var. heucheriformis Engl. in Engl. & Prantl Nat. Pflanzenf. ed. 2, 18a:120 (1930).—Rimo Bacigalupi, Dudley Herbarium, Stanford University. ASPERUGO IN CALIFORNIA. Asperugo procumbens L. has recently been collected ina hay meadow near Fort Bidwell, Modoc County, by John C. Hays. This weed, a native of Europe, has long been known in the eastern United States, and in recent years has been collected in Washington, Oregon, Utah, and Colorado. This collection, however, apparently represents the first for Cali- fornia. Specimens have been deposited in the herbarium of the Botany Division, University of California, Davis, and in the her- barium of the California Academy of Sciences, San Francisco.— Charles Heiser, Jr., Division of Botany, University of California, Davis. 72 LEAFLETS OF WESTERN BOTANY [ VOL. V, NO. 4 ~ EUPHORBIA PROSTRATA AIT. IN CALIFORNIA. I cannot find that this widely distributed American species has been reported heretofore from California, where, in August, 1945, it was dis- covered in the Ojai Valley, Ventura County, by Mr. H. M. Pollard. Although the plant is indigenous through the central and southern United States as far west as Arizona, it is un- doubtedly adventive in California. In the Ojai Valley it grows with the weedy E. supina Raf. (£. maculata of authors), but the two may be distinguished by the transverse ridges on the seeds which are sharp and roughened in E. prostrata and rounded and smooth in E. supina. Euphorbia prostrata has been identified as E. Chamesyce L. by Wheeler (Rhodora 43: 265-271) but - Croizat (Bull. Torr. Bot. Club 72: 312) has shown that E. prostrata is the correct name for this American species.—J. T. Howell Monecious JuNIPERS IN Mopoc County, CALIFORNIA. Moncecious junipers are rare so that well authenticated specimens should be made known. Recently specimens were received from Mrs. Florence Angwin, collected at Adin, Modoc County. Sev- eral trees of Juniperus occidentalis were observed bearing ber- ries and at the same time both pistillate and staminate flowers. The staminate flowers come very early and soon fall, therefore moncecious junipers may be commoner than is supposed since there are few observers when the trees are in bloom.—Alice Eastwood. CHRYSANTHEMUM BatsaAmita IN IpaHo. Chrysanthemum Balsamita L. is a native of Eurasia, formerly much cultivated for its. fragrance and reputed medicinal properties. It is occasion- ally found as a roadside weed in northeastern United States, and as far west as South Dakota, seldom spreading much from the original site of cultivation. A specimen from a canal bank near Burley, Cassia County, Idaho, was recently sent to the New York Botanical Garden by Dr. Ray J. Davis, with the note, “just be- coming established.”—Arthur Cronquist, New York Botanical Garden. e f ie? ee. VoL. V No. 5 LEAFLETS of WESTERN BOTANY CONTENTS A Revision of the Oreastrum Group of Aster ARTHUR CRONQUIST Pugillus Astragalorum IX: Novelties in Batidophaca Rydb. R. C. BARNEBY New Records for Scirpus A. A. BEETLE Concerning a California Cudweed JouNn THomAs Howe. A New California Castilleja JouNn THOMAS HOWELL SAN FRANCISCO, CALIFORNIA FEBRUARY 16, 1948 89 90 91 ‘LEAFLETS. aes | of Pate Tae WESTERN BOTANY Raadx A publication on the exotic flora of California ani on native flora of western North America, appearing about | ‘ Sd times each year. Subscription price, $1.00 annually; single 1 nu BS "bers, 40c. Address: John Thomas Howell, California Acad es of Sciences, Golden Gate Park, San Francisco, California. — . Cited as ae LEAFL. West. Bor. +t ; r = Owned and published by ALIcE Eastwoop and Joun THomas HOWELL hot; F SN ag Atay Ph pee FEBRUARY, 1948] OREASTRUM GROUP OF ASTER 73 A REVISION OF THE OREASTRUM GROUP OF ASTER BY ARTHUR CRONQUIST University of Georgia, Athens The small group of western American monocephalous sub- scapose perennial Asters with a taproot or erect caudex has been recognized as a natural assemblage for at least 50 years. In 1896 E. L. Greene proposed the genus Oreastrum to receive these species, transferring Aster alpigenus (T.&G.) A. Gray and A. Andersonu A. Gray to it, and describing a third species, Orea- strum elatum, which he stated to be “more distinct from the two preceding than they are from each other.” Four years later Rydberg transferred A. Haydeni Porter to Oreastrum. The lat- ter species had been treated by Gray and others as a smaller, inland phase of A. alpigenus. Greene then proposed the generic name Oreostemma to replace Oreastrum, recognizing four spe- cies, O. alpigenum, O. Andersonii, O. elatum, and O. Haydeni. The segregate genus Oreastrum or Oreostemma has usually been submerged in Aster by conservative botanists, without denying the naturalness of the group. In 1932 Onno reduced A. Andersonii to subspecific rank under A. alpigenus, and in this treatment he was followed by Peck in the latter’s Manual of the Higher Plants of Oregon (1941). Meanwhile, the reduced form which we have mentioned as A. Haydeni was treated by some as a species under Oreastrum, Oreostemma, or Aster, and was completely submerged in A. alpigenus by others. It has not previously received formal intraspecific recognition. In preparing a treatment of the Composite for Dr. Ray J. Davis’ forthcoming flora of Idaho, I was faced with the problem of the proper taxonomic status of A. Haydeni. It clearly war- ranted some distinction from typical A. alpigenus, yet was also obviously closely related. A detailed study of the whole group was thereupon undertaken, and about 1000 herbarium speci- mens from various institutions were assembled; only a small pro- portion of these specimens are cited in this paper. I wish to thank the curators of the several herbaria here listed for their kindness in loaning specimens for the work. The abbreviations given are those used henceforth in the text. Leaflets of Western Botany, Vol. V, pp. 73-92, February 16, 1948. om a my 74 LEAFLETS OF WESTERN BOTANY [VOL. V, NO. 5 C, University of California, Berkeley CA, California Academy of Sciences, San Francisco CIUC, Clokey Herbarium at the University of California, Berkeley CV, California Vegetative Type Map Herbarium, housed at Berkeley G, Gray Herbarium, Harvard University, Cambridge, Mass. IS, Idaho State College, Pocatello M, Missouri Botanical Garden, St. Louis NY, New York Botanical Garden PA, Academy of Natural Sciences of Philadelphia PO, Pomona College, Claremont, California RM, Rocky Mountain Herbarium, University of Wyoming, Laramie Th, J. William Thompson Herbarium, deposited at the University of Washington US, United States National Herbarium, Washington, D.C. UT, Utah State College, Logan W, University of Washington, Seattle WI, Willamette University, Salem, Oregon WS, Washington State College, Pullman While I do not wish to become involved at this time in a comprehensive study of the validity of the various generic segre- gates from Aster, it seems clear in the case of Oreastrum that the segregation is unwarranted. The monocephalous habit, reduced cauline leaves, and well-developed erect caudex do indeed mark a natural group, but the group is clearly related to A. occiden- talis, and extreme forms of A. alpigenus subsp. Andersonii even have creeping rhizomes like that species. While retaining the erect caudex, the group shows every transition from a strictly tap-rooted to a strictly fibrous-rooted habit, so much so that I have been unable to make any taxonomic use of the under- ground parts. The proper rank at which the group should be received in Aster need not be determined at this time; certainly it is no more than a section. My conclusions from the study are that A. alpigenus is a polymorphic species with three well-marked geographical sub- species (these having been treated specifically as A. alpigenus, A. Andersonit, and A. Hayden), that Oreastrum elatum Greene is probably specifically distinct, and that certain reduced alpine glandular plants of the southern Sierra Nevada constitute a distinct species recently described by Dr. Carl W. Sharsmith. KEY TO THE SPECIES Plants either glabrous or slightly woolly-puberulent, not at all glandular. Involucre not especially firm, the phyllaries often chartaceous at the FEBRUARY, 1948] OREASTRUM GROUP OF ASTER 75 base, but scarcely coriaceous; plants with at least a slight cottony or woolly puberulence on the involucre or under the heads, often becoming essentially glabrous at full maturity, commonly 0.3—4 dm. tall, rarely taller; mostly in alpine or subalpine sites, widely distrib- Ratan Wes kA wake ok a0 RAR Red RE ele ee 1. A. alpigenus Involucre very firm, the bases of the phyllaries conspicuously coriaceous; plants glabrous throughout, 3-7 dm. tall; local at relatively low elevations in northern California...................5+ 2. A. elatus Plants with the stems, involucres, and often also the leaves glandular, not otherwise hairy; alpine and subalpine sites in the southern Sierra A Saree bid iti a fv'd n)a'a e's ote VP amte Ree awh Heads k 3. A. Peirsonii SYSTEMATIC TREATMENT OF THE SPECIES 1. ASTER ALPIGENUS (T. & G.) A. Gray. Perennial with an erect, woody, sometimes branched caudex, fibrous- rooted or with one or several main roots; stem 3-40 cm. tall, or rarely to 7 dm. tall, usually decumbent, its leaves reduced, not very numerous; basal leaves persistent, linear to linear-elliptic or oblanceolate, acute to rounded, up to 25 cm. long and 15 mm. wide, I-nerved or when larger often several- nerved and somewhat veiny; herbage often woolly-puberulent when young, characteristically and more persistently so under the heads and on the margins of the involucral bracts; heads solitary; involucre 5-13 mm. high, turbinate to subhemispheric, its bracts more or less imbricate, somewhat chartaceous at the base and often conspicuously purple-margined, not espe- cially firm; rays 10-40, violet or lavender, 7-15 mm. long, about 2-4 mm. wide; disk-flowers about 5-9 mm. long, the tube usually ill-defined, the lobes commonly 0.6-1.4 mm. long; style-appendages 1-2.5 mm. long, 1-4 times as long as the stigmatic portion; achenes several-nerved, hirsute throughout or only at the summit, or occasionally quite glabrous; pappus of about 30-50 bristles, occasionally with a few short outer setz, these prob- ably phyletically derived from the uppermost achenial hairs, which indeed frequently simulate an outer pappus. Open places, mostly in alpine or subalpine situations, subspecies An- dersonii sometimes descending lower. Washington to southern California, east to southwestern Montana, western Wyoming, and northeastern Nevada. Type: Tolmie s.n., Mt. Rainier, Washington (G). KEY TO THE SUBSPECIES OF Aster alpigenus Achenes usually hairy to the base; plants of California and adjacent Nevada, north in the Cascades to southern Oregon, and, in forms, passing to subsp. typicus, to southern Marion County, Oregon................. ROR ih nia :'n sn. a0; vnrela levis alldbnin’s i pcactin se alah eNot a tes A. subsp. Andersonii Achenes glabrous below. Leaves mostly oblanceolate, tending to be obtuse or rounded, the largest ones seldom less than 5 mm. wide; Olympic and Cascade mountains of Washington, south in the Cascades to about latitude 44° in Deschutes County, Oregon, and east to the Blue and Wallowa mountains of northeastern Oregon............. B. subsp. typicus 76 LEAFLETS OF WESTERN BOTANY [VOL. V, NO. 5 Leaves mostly linear or linear-elliptic to linear-oblanceolate, commonly up to 3 mm. or occasionally 5 mm. wide; western Wyoming and southwestern Montana, across Idaho to the Blue, Wallowa, and Steens mountains of eastern Oregon, and to the mountains of Elko County: Nevadas. icc acetic aera se rier C. subsp. Haydeni 1A. ASTER ALPIGENUS subsp. ANDERSONII (A. Gray) Onno, Bibl. Bot. 26 (Heft 106): 15 (1932). Erigeron Andersonii A. Gray, Proc. Am. Acad. 6: 540 (1865). Aster Andersonii A. Gray, Proc. Am. Acad. 7: 352 (1868). Oreastrum Andersonii Greene, Pitt. 3: 147 (1896). Oreostemma Andersonii Greene, Pitt. 4: 224 (1900). Plants 3-40 cm. tall, or rarely to 7 dm., decumbent, or especially when larger sometimes erect, usually fibrous-rooted and sometimes even with creeping rhizomes, varying to evidently tap-rooted; leaves commonly linear- elliptic and acute, up to 25 cm. long and 13 mm. wide, varying to essentially like those of either of the other subspecies; heads mostly hemispheric, the involucre 5-12 mm. high; rays 12-35; style-appendages mostly 1.5-2.5 mm. long; achenes commonly hairy to the base. Most commonly in moist meadows, sometimes in drier, more exposed situations. From the San Jacinto Mts. of southern California, north in the Sierra Nevada and Cascades to southern Oregon, overlapping into extreme western Nevada, and extending to the coastal counties in the Siskiyou Moun- tains of northern California and southern Oregon; forms passing to subsp. typicus may be found as far north as Marion County, Oregon. Type: Dr. C. L. Anderson s.n., near Carson City, Nevada (C). CALIFORNIA: Abrams 12764, Half Moon Lake, Eldorado Co., September 3, 1930 (C, G, NY, Th, US, W); Brewer s.n., Lake Tenaya, in 1863 (US); Cope- land 486, Jonesville, Butte Co., July 26, 1930 (achenes glabrous below) (C, CA, G, M, NY, PO, RM, US, W); Hall 2574, Tahquitz Valley, San Jacinto Mts., Riverside Co., July 30, 1901 (C, M, NY, PA); Hall & Babcock 395, Dinkey Creek, Fresno Co., 1900 (C, M, NY, US); Heller 7136, south of Donner Pass, Nevada Co., August 10, 1903 (C, G, M, NY, PA, PO, RM, US); Heller 11650, Butte Meadows, Butte Co., August 17, 1914 (achenes glabrous below) (C, CA, M, NY, US, W); Heller 12225, Mt. Eddy, Siskiyou Co., August 26, 1915 (CA, G, M, NY, US, W); Lemmon 128, Sierra Co., 1875 (C, NY, US); Torrey 208, near Donner Lake, 1865 (G, NY, US); Tracy 6708, Snow Camp, Hum- boldt Co., July 4, 1924 (C). Nevapa: Howell 14236, Slide Mountain, Washoe Co,, August 5, 1938 (CA, W); Kennedy 994, Mt. Rose, Washoe Co., August 17, 1905 (C, M, NY, PA, US); Kennedy 1252, Galena Creek, Washoe Co., August 1, 1906 (C, M, NY, US); Mason 12373, Mt. Rose, Washoe Co., August 6, 1940 (C, G, M, NY, PO, UT, W, WS); Train 3248, Marlette, Sierra Nevada, Ormsby Co., July 8, 1939 (C1UC, UT, W). OrEGON: Peck 5785, Mt. Pitt, Jackson Co., August 25, 1916 (G, WI); Peck 16555, Four-Mile Lake, Klamath Co., July 2, 1931 (NY, Th, WI); Thompson 4588, 10 miles southwest of Waldo, Josephine Co., June 7, 1928 (Th). FEBRUARY, 1948] OREASTRUM GROUP OF ASTER 77 1B. ASTER ALPIGENUS subsp. TyPIcus Onno, Bibl. Bot. 26 (Heft 106): 16 (1932). Aplopappus alpigenus T. & G. Fl. N. Am. 2: 241 (1841). Aster alpigenus A. Gray, Proc. Am. Acad. 8: 389 (1872), sens. strict. Oreastrum alpigenum Greene, Pitt. 3: 146 (1896). Oreostemma alpigenum Greene, Pitt. 4: 224 (1900). Plants 5-20 cm. or rarely 30 cm. tall, decumbent, stout; leaves mostly oblanceolate, tending to be obtuse or rounded, up to 15 cm. long and 15 mm. wide, the largest ones seldom less than 5 mm. wide; heads mostly subhemi- spheric, the involucre 7-11 mm. high; rays mostly 18-40, 9-14 mm. long; achenes hairy above the middle or toward the summit; style-appendages mostly 1.5-2.5 mm. long. Olympic Mountains of Clallam Co., Washington; Cascade Mountains from about latitude 47° in Chelan Co., Washington, south to about lati- tude 44° in Deschutes Co., Oregon, and east to the Blue and Wallowa moun- tains of northeastern Oregon. OREGON: Sharsmith 3963, Eagle Cap Peak, Wallowa Co., September 24, 1938, in large part, but passing into subsp. Haydeni (C, G, UT, WI, WS); Elihu Hall 248, in 1871 (G, M, NY, PA); Howell 7135, north of Mt. Bachelor, Deschutes Co., July 2, 1931 (CA); Thompson 5087, Mt. Hood, July 22, 1928 (M, PA, US); Whited 367, Three Creek Lake, Crook Co., September 13, 1921 (G); Coville & Applegate 555, south of the Three Sisters, in Lane or Des- chutes Co., August 17, 1897 (US). WASHINGTON: Abrams 11504, Mt. Rainier, Pierce Co., July 24, 1926 (C, CA, G, M, NY, PO, RM, US); Meyer 1099, Obstruction Point, Olympic Moun- tains, Clallam Co., August 11, 1938 (G, M, Th, WS); Piper 2156, Mt. Rainier, August, 1895 (G, PO, US, WS); Thompson 7642, Mt. Stuart, Chelan Co., July 27-31, 1931 (M, PA, Th); Thompson 15095, Chinook Pass, Yakima Co., August 1, 1940 (CA, C1UC, G, M, NY, WS); Thompson 15083, Mt. Aix, Yak- ima Co., July 15, 1940 (G, M, NY, US). 1C. Aster alpigenus subsp. Haydeni (Porter) Cronquist, comb. nov. Aster Haydeni Porter in Hayden Geol. Rep. 1871: 485 (1872). Aster pulchel- lus D. C. Eat. in King Geol. Expl. 40th Par. 5: 143 (1871) not A. pulchellus Willd. Oreastrum Haydeni Rydb., Mem. N. Y. Bot. Gard. 1: 398 (1900). Oreostemma Haydeni Greene, Pitt. 4: 224 (1900). More slender than subsp. typicus; mostly 3-15 cm. tall, decumbent; leaves mostly linear to linear-elliptic or linear-oblanceolate and acute, up to about 10 cm. long and 3 mm. or occasionally 5 mm. wide; involucre tending to be turbinate, 5-13 mm. high; rays 10-31; style-appendages mostly 1-2 mm. long; achenes hairy toward the summit, or occasionally quite glabrous. Range as given in the key. Type: Hayden 15, Upper Falls of the Yellowstone River, August 27, 1871 (G, PA). A collection made by Coulter in Upper Teton Canyon, “Idaho”, on the Hayden expedition of 1872, has been confused with the type (US). Another Coulter collection, made in Madison Canyon, “Idaho” in 1872 has been mounted with the Hayden collection on the sheet at PA, and it is now scarcely possible to determine which specimens are which. In any case, they all appear to represent the same taxonomic entity. 78 LEAFLETS OF WESTERN BOTANY [VOL. V, NO. 5 IpAHo: Cronquist 1900, Mount Jefferson, Fremont Co., August 1, 1939 (IS, M); Cronquist 3014, 14 miles west of Challis, Custer Co., July 9, 1941 (G, IS, M, UT); Davis 1373, Mt. Harrison, Cassia Co., July 14, 1939 (IS); Hitchcock & Muhlick 10844, east of Castle Peak, Custer Co., August 8, 1944 (CA, G, UT, W, WS); Macbride & Payson 3741, Smoky Mountains, Blaine Co., August 13, 1916 (C, CA, G, M, NY, PO, RY, US); Thompson 14114, Boulder Creek, Blaine Co., August 4, 1937 (C, CA, G, M, US, W). Montana: Blankinship 715, Lake Hearst, Anaconda, Deerlodge Co., September 2, 1906 (C, M, PO, RM, US); Hitchcock & Muhlick 13078, Rock Creek, Pioneer Mountains, August 1, 1945 (UT, W, WS); Hitchcock & Muhlick 13430, Mt. Haystack, Stillwater Co., August 8, 1945 (UT, W, WS); Scribner 88, Little Belt Mountains, July 15, 1883 (G, PA). NeEvaDA: Maguire & Holmgren 22575, above Island Lake, Ruby Range, Elko Co., August 16, 1943 (G, M, NY, UT, W); Maguire & Holmgren 22687, below Lamoille Lake, Ruby Range, Elko Co., August 19, 1943 (NY, PO, UT); Watson 513, Clover Peak (G, NY, US). OreEcON: Bailey 7700, Steens Mountains, Harney Co., July 31, 1916 (WI); Maguire & Holmgren 26764, Fish Lake, Steens Mountain, Harney Co., August 1, 1946 (NY); Peck 17804, Mirror Lake, Wallowa Co., July 20, 1933 (NY) (similar specimen with same data at WI bears number 17525). WyomiInc: Goodding 432, Ten Sheep Lakes, Big Horn Co., July 31, 1901 (C, G, M, NY, PO, RM, US); Merrill & Wilcox 1097, Teton Mountains near Leigh’s Lake, July 26, 1901 (G, NY, RM, US); Nelson 1016, Union Pass, August 13, 1894, (G, M, NY, US); Nelson & Nelson 6325, Yellowstone Lake, August 6, 1899 (G, M, NY, PO, RM, US); Payson 2720, Piney Mountain, Sublette Co., July 12, 1922 (C, G, NY, PA, PO, RM, US); Payson 4575, Green River Lakes, Sublette Co., August 5, 1925 (G, M, PA, RM, WS); Payson & Armstrong 3817, east of Smoot, Lincoln Co., August 13, 1923 (G, M, PA, PO, RM); Williams 946, Two-gwo-tee Pass, July 27, 1932 (CA, G, M, NY, RM, UT); Williams 3134, near Powder River Pass, Johnson Co., June 27, 1936 (G, M, NY, WS). The three intraspecific units of Aster alpigenus beautifully illustrate the geographic subspecies as conceived by Wettstein. Each has its own characteristic aspect, and each has been con- sidered as a distinct species by reputable taxonomists; yet they have an over-all unity which distinguishes them as a group from related entities. Individual specimens from well within the range of one might pass for the other in the absence of geographic data; and they intergrade completely where their ranges overlap. The subsp. Andersonii is the most variable of the three. In size it ranges from no larger than typical subsp. Haydeni to fully as large as A. elatus, being then much larger than either subsp. Haydeni or subsp. typicus. There is a strong negative correla- tion of size with the elevation at which the plants occur, and it FEBRUARY, 1948] OREASTRUM GROUP OF ASTER 79 may eventually be desired to recognize two or more ecotypes, but intergradation is so complete that I feel unable to separate them without further field observation. The leaves are usually long-pointed, as in subsp. Haydeni, but vary to shortly rounded, as in subsp. typicus. The stem is characteristically decumbent, but especially in larger forms is sometimes quite erect. The achenes are typically pubescent to the base, but some specimens from northern California, which do not otherwise approach subsp. typicus, have them glabrous below as in that subspecies. Although the plants are ordinarily fibrous-rooted, distinctly tap-rooted and, transitional forms occur. Several collections, while retaining the erect caudex, are not only fibrous-rooted but have well-developed creeping rhizomes. Extremely reduced alpine specimens of subsp. Andersonii, although greatly resembling subsp. Haydeni, may readily be distinguished from the latter by their uniformly hairy achenes. There is furthermore a break of more than 200 miles between the known ranges of the two. Further exploration of the high mountains of northwestern and north-central Nevada, where the species is not yet known to occur, may well reveal inter- gradient forms. In contrast to the seeming distinctness of subsp. Andersonti from subsp. Haydeni, both of these entities pass freely into subsp. typicus in the areas of range-overlap. In addition, there seems to be either a considerable gene-flow or an original lack of full differentiation between subsp. Andersonii and subsp. typicus, since both the leaf-shape and the achenial pubescence of subsp. typicus are found independently in otherwise apparently not unusual specimens of subsp. Andersoni from northern Cali- fornia, far beyond the range of subsp. typicus. True intergrada- tion is also revealed by specimens collected by M. E. Peck near Breitenbush Lake, Marion or Linn Co., Oregon, numbers 18795 (WI, WS) and 18838 (C). These have some of the achenes glab- rous below the middle, while others in the same head are sparsely hairy to the base; the habit of the specimens varies from typical of subsp. typicus to nearly typical of subsp. Andersonii. A num- ber of other specimens from the Cascades of central Oregon are not clearly referable to either one of the two subspecies. The intergradation between subsp. typicus and subsp. Hay- 80 LEAFLETS OF WESTERN BOTANY [VOL. V, NO. 5 deni is even plainer. There is no doubt that Haydeni is on the average smaller and less robust, with narrower and more pointed leaves, and narrower-shaped heads, than typicus, but there are many individual specimens, and several whole collections which would be impossible to identify, or which might well be misiden- tified, were their geographical origin unknown. Most of the specimens of the widely distributed Allen 143, from Mt. Rainier, Washington (G, M, NY, PO, RM, US), would probably pass as subsp. Haydenz had they been collected in Idaho. Furthermore, each of several collections from the Blue and Wallowa moun- tains of northeastern Oregon runs the gamut of variation from typical subsp. typicus to typical subsp. Haydeni. Among these might be mentioned Ownbey 1848 (CA, G, M, NY, PO, US, WS). In spite of the obvious intergradation, however, a very large proportion of the specimens from the whole range of the two can be identified without question from their gross morphology, and the existence of regionally differentiated units can scarcely be denied. 2. Aster elatus (Greene) Cronquist, comb. nov. Oreastrum elatum Greene, Pitt. 3: 147 (1896). Oreostemma elatum Greene, Pitt. 4: 224 (1900). Not Aster elatus Bert. ex Steudel Nom. ed. 2, 1: 154; a hyponym not validly published. Fibrous-rooted perennial from an upright woody caudex, glabrous throughout, the whole plant appearing coarser and firmer than the related A. alpigenus subsp. Andersonii; stem 3-7 dm. tall, erect or decumbent at the base; basal and lowermost cauline leaves persistent, linear-elliptic, some- what veiny, 8-25 cm. long, 5-10 mm. wide; middle and upper leaves erect, few, and reduced; heads solitary; involucre 11-14 mm. high, subhemispheric, its bracts scarcely to evidently imbricate, very firm, the coriaceous yellow base nearly as long as the green tip, or longer; rays about 25, violet or lav- ender, 7-12 mm. long, about 2-3 mm. wide; disk-flowers about 7 mm. long, the lobes a little over 1 mm.; style-appendages about 2.5 mm. long, the stig- matic portion very short, only about 0.5 mm. long; achenes several-nerved, glabrous, or slightly hairy especially toward the summit; pappus of about 40 rather firm bristles and a few inconspicuous outer sete which may be more than 1 mm. long. Meadows at relatively low elevations in Plumas Co. and perhaps also Lassen Co. in northern California. Type: Mrs. R. M. Austin, Mt. Dyer (on the border of Plumas and Lassen counties), California, July, 1897, presumably in the Greene Herbarium at Notre Dame University; not seen. The Austin specimen cited below may well be part of the type-collection. FEBRUARY, 1948] OREASTRUM GROUP OF ASTER 81 CALIFORNIA: Austin 1361 (or 1367), Big Meadows, Plumas Co., July 27 (US); Howell 618, Butterfly Valley, Plumas Co., August 10, 1924 (CA); Lem- mon s.n., Plumas Co., July 27 (G). Aster elatus is known from only three or four collections from a limited area in northeastern California, so that its normal range of variation cannot yet be ascertained. It has a distinctive aspect, and I feel, as did Greene, that it is less closely allied to the several subspecies of A. alpigenus than they are to each other. It is only after some hesitation, however, that I treat it as a distinct species. Its relatively large size may be due merely to its growth at lower elevations than is typical of A. alpigenus, and there are two collections from southern Oregon (Howell 142, near Waldo, June 5, 1884, G, NY, US, WS; and Leach 19782, Josephine Creek, June 25, 1930, WI) which are fully as large as A. elatus, up to 7 dm., but seem typical otherwise of A. alpigenus subsp. Andersonii. The complete absence of pubescence from the herbage and involucre of A. elatus is noteworthy, but some specimens of A. alpigenus subsp. Andersonii are only very scantily pubescent. The most striking feature of A. elatus is its very firm involucre, with conspicuous coriaceous bases to the phyllaries, and it is this character, even more than the absence of pubescence, which leads me to consider it as a distinct species. The large size and general coarseness of A. elatus suggest that it may be an autoploid. Such a condition, with its resultant barrier to interbreeding and exchange of genes, would further substan- tiate the case for treating it as a distinct species. 3. ASTER PEIRSONII C. W. Sharsmith, Leafl. West. Bot. 5: 50 (1947). Dwarf perennial with an erect branching caudex and a taproot; stem, involucre, and often also the leaves copiously provided with short spreading gland-tipped hairs; basal leaves tufted and persistent, linear, often folded along the midrib, 1.5-5 cm. long, 1-3 mm. wide; cauline leaves few and re- duced; stems 2-7 cm. high, decumbent or erect, monocephalous; heads turbinate or turbinate-hemispheric when pressed; involucre 7-11 mm. high, its bracts imbricate, linear, tapering to a slender, often indurate point, strongly suffused with purple, the outer evidently green-tipped, the inner more shortly so or wholly purple; rays mostly 8-20, lavender or violet, 10-18 mm. long, 1.5-2.5 mm. wide; disk-corollas slender, 5.5-8 mm. long, the lobes about 1 mm. long or less; style-appendages 1.2-2 mm. long, the stigmatic portion less than 1 mm. long; achenes several-nerved, slightly pubescent; pappus of about 25-40 slender bristles. Alpine and subalpine meadows and granitic gravelly soil in the Sierra Nevada of Inyo, Tulare, and southern Fresno counties, California. 82 LEAFLETS OF WESTERN BOTANY [VOL. V, NO. 5 Type: Sharsmith 3253, northwest base of University Peak by highest of Kearsage Lakes, Kearsage Pass region, 11,200 feet, Fresno Co., August 16, 1937; deposited at the University of California. CaLiFornNiA: Alexander & Kellogg 3318, in moist flat, third lake, Cotton- wood Lakes, southwest of Lone Pine, Inyo Co., August 17, 1942 (C, UT); Frost s.n., Wright Creek, Township 15 south, Range 33 east, Mt. Whitney Quadrangle, Tulare Co., September 19, 1933 (CV); Howell 15869, between Reflection Lake and Harrison Pass, Tulare Co., August 1, 1940 (CA); Saun- ders s.n., Mt. Whitney region, August, 1938 (CA). The evident glandulosity and peculiarly pointed phyllaries of A. Peirsoni provide easy means of distinguishing it from A. alpigenus. In addition, A. Peirsoniz is distinctly narrower-headed and more cespitose than A. alpigenus subsp. Andersonii, the only form of A. alpigenus that occurs in its range. Its distinct- ness may be further indicated by the fact that I had it in manu- script as a new species before I was aware of Dr. Sharsmith’s interest in the plant, and that each of us laid particular stress on its glandular indument as a distinguishing feature. Dr. Shar- smith has studied it at several places in the field, and his obser- vations, together with the now 11 known collections, fully substantiate its specific status. That the characteristic gland- ulosity of Aster Peirsonii does not represent a tendency entirely foreign to A. alpigenus is shown by the fact, however, that occa- sional specimens of the latter have the pubescence slightly viscid. PUGILLUS ASTRAGALORUM IX: NOVELTIES IN BATIDOPHACA RYDB. BY R. C. BARNEBY Wappingers Falls, New York Centering in the butte country towards the southeastern corner of the Great Basin, there exists a group of dwarf, cespi- tose or matted Astragali with small flowers and essentially 1- locular pods which are remarkable for their habitat, being nearly or quite confined to ledges of sandstone escarpment or to small areas of pavement along the rim of box-canyons. They are among the rarest species in the genus, several (or varieties of them) being known only from a single locality, the rest from but few stations in a limited range. During our journeys in the Southwest, Mr. Ripley and I have come to recognize “ledge- FEBRUARY, 1948] PUGILLUS ASTRAGALORUM IX 83 pavement” as a distinctive type of habitat which seems to fur- nish a balance of environmental factors peculiarly suited to the survival of these eclectic little plants. Special concentration in the field on such ecological islands has brought to light the entities described below. I have referred to the species here treated as a “group,” without intending that term to convey phylogenetic overtones. It seems probable that there are several lines of inheritance in- volved, and that more than one natural section in the genus has contributed a member or two to the ledge-flora. An attempt to define these sections, entailing a review of the greater part of Rydberg’s genus Batidophaca, together with his section Hu- millime of Phaca and other miscellaneous species, lies outside the scope of this paper. I have contented myself here with point- ing out what is believed to be the nearest relative of each. In the preparation of these notes I have had the privilege of making comparisons in the herbaria listed, with the appro- priate symbol used in citation of material, below: California Academy of Sciences (CAS); Gray Herbarium (G); University of Minnesota (M); New York Botanical Garden (NY); Pomona College (PO). I wish to express my gratitude to those who have courteously made available the collections in their care. Astragalus cremnophylax Barneby, spec. nov., pube dolabriformi ac statura minima A. humillimo Gray comparanda, sed stipulis glabratis adversus petiolum liberis, leguminis valvulis rigidioribus, petiolisque flaccis (nec cum rachi diu persistentibus subspinosis) absimilis. Ab A. gilensi Greene, cui leguminis forma verosimiliter propius affinis, racemo paucifloro, habitu pulvinato, etc., longe distat. Herba diu perennis depresso-pulvinata subacaulescens, caulibus horno- tinis 2—5 mm. longis e caudice ramosissimo ligneo stipulis petiolisque marcidis crebre induto ortis, preter stipulas corollamque pube arcte appressa dolabriformi undique strigoso-cana; stipulis imbricatis late ovatis obtusis vel obtusiusculis 1.3—3 mm. latis, petioli tergo manifeste adnatis, amplexi- caulibus sed haud connatis, primum herbaceis apicem versus parce strigosis ciliatisque, mox scariosis castaneis glabratis; foliis 3—9 (12) mm. longis patulis, petiolo rachique profunde sulcatis; foliolis fere sessilibus sed manifeste articulatis confertim 2—3 (4)-jugis, suborbicularibus obovatis late ellipticis, obtusiusculis obtusisve, 1—2.5 mm. longis sepissime conduplicatis utrinque argenteis; pedunculis 3—5 mm. longis vel interdum subobsoletis; racemo 1—3 (vulgo 2)-floro, floribus patulis; bracteis scariosis late ovatis glabratis pedicello gracili circa 1 mm. longo subdimidio brevioribus; calycis ebracteolati pilis albis vel interdum nonnullis fuscis adspersis strigosi tubo 84 LEAFLETS OF WESTERN BOTANY [VOL. V, NO. 5 oblique obconico-campanulato 1.5 mm. longo demum rupto, dentibus subulatis 1 mm. longis; petalis lilacinis; vexilli abrupte retroarcuati 5.5 mm. longi lamina orbiculari retusa 4 mm. lata; alis 5 mm. longis, lamina obovato- elliptica obtusa 1.3 mm. lata, auriculo minuto incluso 3.5 mm. longa; carine 4 mm. longe laminis 2.5 mm. longis fere 2 mm. latis, marginibus inferioribus per 120° in apicem minute porrectum abrupte incurvis, superioribus leviter concavis; legumine sessili 1-loculari oblique ovoideo utrinque obtuso 3—4 (4.5) mm. longo, circa 2.5 mm. diametro, dorsaliter complanato vel inferne paulo sulcato-impresso, suturis dorsali subrecta, ventrali prominula valida per totam longitudinem in rostrum subobsoletum vel brevissime deltoideum deflexum arcuata, valvulis subcoriaceis obscure ruguloso-reticulatis strigoso- canescentibus; ovulis 4—6; seminibus 2—3 (4), circa 1.8 mm. longis, fulves- centibus. ARIZONA: in fissures of limestone pavement on the south rim of the Grand Canyon of the Colorado River, about 2 miles west of E] Tovar, Coconino Co., alt. 7050 ft., 3 June 1947, fr., Ripley & Barneby No. 8473. Type in Herb. Calif. Acad. Sci., No. 336060. Grand Canyon (“at the end of the railroad on sandy ledges’’), Jones in 1903, fl. (CAS, fragments, PO). Astragalus humillimus, as figured and described by Jones (Rev. Astrag. 82, Pl. 6,—1923), is almost wholly the species here proposed. It is still not possible to contrast in full the characters of A. cremnophylax with those of the true A. humillimus Gray, of which only Brandegee’s imperfect, now flowerless type from the Mesa Verde, Colorado, is extant (G, NY). The two species are admittedly alike in their diminutive stature and dolabriform pubescence, and the pods are of about the same size and out- line, though apparently of thinner texture and less flattened dorsally in A. humillimus. In the latter, however, the stipules are distinctly connate opposite the petiole, whereas in A. crem- nophylax they are, albeit broad and amplexicaul, free to the base. In the Arizona species the petioles are flaccid, persisting on the caudex-branches merely as weak, fibrous tatters after the first season; whereas in A. humillimus they are wiry and sub- spinescent, persisting (with the rachis) for several years un- changed, as in the Eurasian Tragacanthas. No doubt further characters will appear when A. humillimus becomes better known. While the closeness of its affinity to A. humillimus can only be guessed at, A. cremnophylax does possess one obviously close relative in A. gilensis Greene, essentially similar in pubescence, FEBRUARY, 1948] PUGILLUS ASTRAGALORUM IX 85 in shape and proportions of the petals, and especially in the pod. In both the legume is ovoid, flattened dorsally and a little de- pressed to definitely sulcate near the base, while the thick and prominent ventral suture is about evenly convex through its whole length, ‘so that the beak (small and obscurely differen- tiated from the body in A. cremnophylax ) is somewhat declined. Astragalus gilensis, very different in its ample foliage, scapiform peduncles and many-flowered racemes, is a plant of open pine- woods along the Mogollon Rim, ranging from southern Navajo County, Arizona, into southwestern New Mexico. In its type-locality A. cremnophylax is confined to a narrow strip of limestone pavement immediately overlooking the Can- yon1, where the young plants form a dense silvery moss flattened against the stones. As these mature, the matted foliage is raised on a mass of gnarled, woody caudex-branches into moundlike cushions 1.5—2 dm. in diameter which, since the maximum growth of the shoots in a favorable year cannot exceed 5 mm., must be of considerable age. The fresh flower is described by Jones as light purple, the banner speckled, the wings (as in A. gilensis) paler than the rest or white. Astragalus micromerius Barneby, spec. nov., stipulis alte connatis leguminisque forma Homaloborum Jones seu Batidophace Rydb. speciebus cespitantibus, racemoque paucifloro presertim A. sesquifloro Wats. affinis, sed ab eo pube hirsuta basifixa, necnon corolla, legumine foliisque saltem dimidio minoribus diversissima. Ab A. humillimo Gray imprimis petiolis flaccis deciduis, nec rigidis cum rachi persistentibus, aliisque notulis abhorret. Herba perennis ramosissima inferne suffruticulosa, tegetes depressas 1—6 dm. latas efformans, preter legumen corollamque pilis patulis basifixis undique hirsuto-villosa et canescens; caulibus annotinis nunc brevissimis, in summis caudicis ramulis congestis et stipulis imbricatis tota longitudine indutis, nunc extensis, ad 12 cm. usque longis (internodiis manifestis 5—20 mm. longis), omnibus gracillimis teretibus flexilibus sed haud abrupte flexu- osis, hinc inde ramosis; stipulis late ovatis 2-3 mm. longis, adversus petio- lum alte connatis, ad apicem deltoideum liberum plus minusve squarrosis, primum herbaceis mox scariosis, intus glabris; foliis patulis reflexisve, 4—9 mm. longis, petiolo flacco filiformi deciduo rachin subequanti vel paulo superanti; foliolis (1) 2—4-jugis, valde confertis, breviter sed mani- feste petiolulatis, anguste oblongo-ellipticis obovatisve obtusis, 1—3.5 mm. longis, plerumque conduplicatis falcatisque, terminali minimo; pedunculis divaricatis, 3-6 mm. longis; racemis brevissime 1—3-floris; bracteis ovato- lanceolatis 1-2 mm. longis, pedicellum gracilem patentem paulo super- 1Whence the name from the Greek, meaning watchman or overseer of the chasm. 86 LEAFLETS OF WESTERN BOTANY [VOL. V, NO. 5 antibus; calycis ebracteolati demum rupti tubo obconico-campanulato 2—2.5 mm. longo, dentibus subulatis acutis 1—-1.3 mm. longis; corolla albida vel pallide lilacina, alis ad apicem dilute, carina saturatius purpureo-tinctis; vexillo alisque subaquilongis, 5—5.5 mm. longis, illo obovato 3—3.5 mm. lato emarginato, medium versus leviter retroarcuato, harum lamina oblonga subrecta 1.3 mm. lata, auriculo retroflexo incluso 3.5 mm. longa; carine 4 mm. longe unguiculis laminisque equilongis, his triangulari-lunatis, mar- ginibus superioribus rectis vel superne leviter convexis, inferioribus per angulum rectum in apicem obtusissimum vel obtusiusculum arcuatis; legu- mine patulo sessili 1-loculari oblique ovoideo 4—6 mm. longo, 2.5—3 mm. lato, basi obtuso et aliquantulum obcompresso, ad apicem in rostrum del- toideum compressum abrupte angustato, suturis filiformibus ventrali promi- nula recta, dorsali convexa sed basin versus obscure sulcata vel depressa, valvulis membranaceis demum stramineis immaculatis, pilis adscendentibus vel fere appressis strigoso-hirsutis; ovulis sapissime 4; seminibus 1—3, oblongis, 2—2.5 mm. longis, ochraceis pulchre purpureo-guttulatis. New Mexico: on shelving ledges of sandstone cliffs at the foot of Pyramid Rock, near Rehoboth, McKinley Co., alt. 7100 ft., 17 July 1945, fl. & fr., Ripley & Barneby No. 7047. Type in Herb. Calif. Acad. Sci., No. 324950. Isotypes G, NY, PO, Inter- mountain Herb. Same locality, 16 May, flor., No. 5276; Satan Pass, 17 miles northwest of Thoreau, McKinley Co., alt. 7300 ft., 17 May, sterile, No. 5277; 3 miles north of Washington Pass, Tunitcha Mts., San Juan Co., alt. 2500 m., Goodman & Hitch- cock No. 3218 (M). This is a remarkable little species, recalling in its tiny leaves and flowers, and in its prostrate, intricately branched mats, the alpine forms of A. (Kentrophyta) tegetarius Wats. From all members of that section, however, it is immediately separated by the petiolulate and obtuse, not confluent and spinulose, leaflets, and by the decided dorsiventral compression of the legume. In the last character, as well as in the connate stipules, it approaches some members of the group known to Rydberg as Batidophaca, where it finds a natural affinity in A. sesquiflorus Wats. The latter, likewise an inhabitant of ledges and crevices of sandstone cliffs, where it forms equally broad and prostrate mats, differs in its strigose, malpighiaceous pubescence, much larger flower and pod, long-acuminate keel and several other characters. From A. humillimus and A. cremnophylax it is immediately distin- guished by its basifixed pubescence, incurved pod, and strongly caulescent stems. Astragalus micromerius is almost unique among the high desert species in its late period of bloom. FEBRUARY, 1948] PUGILLUS ASTRAGALORUM IX 87 Astragalus desperatus Jones var. conspectus Barneby, var. nov., a var. typico2, cui ceterius persimilis, racemo subcapitato fructifero vix elongato, calyce cylindrico longiori, corolleque saturatius purpuree longioris ac angustioris carina 10—11 mm. (nec 6.5—8 mm.) longa separanda. ARIZONA: crevices of sandstone pavement along the Little Colorado River, 4 miles east of Holbrook, Navajo Co., alt. 5000 ft., 1 June 1947, fl. and fr., Ripley & Barneby No. 8451. Type in Herb. Calif. Acad. Sci., No. 337309. The proposed variety differs from all of typzcus that I have seen in the combination of large flowers with subcapitate raceme, and in the longer, proportionately narrower calyx and petals. The typical variety, in which the calyx is always campanulate, varies considerably in size of corolla, and at its longest (e. g. Green River, Utah, Jones in 1895, NY) the banner may reach 14 mm. in length as in var. conspectus; but even here the keel (always a more constant feature) is only 8 mm. long. Likewise the raceme of typicus may occasionally fail to elongate in fruit, but this occurs only in the small-flowered extreme (the so-called A. desperatus var. petrophilus Jones), and is a symptom of starva- tion (cf. Barneby, Leafl. West. Bot. 4: 52,—1944). Both vigorous and depauperate individuals are represented in the type-collec- tion of var. conspectus, but, in spite of corresponding variation in corolla-size, the essential characters remain constant. There seems to be a minor, perhaps wholly inconsequential difference in the color of the flowers. In var. conspectus the petals are all violet-purple, whereas in var. typicus the wings are pale or white, strongly contrasting with the more deeply colored vexil- lum and keel. It seems likely that Mrs. Hough’s plant from Hol- brook, referred to A. desperatus by Kearney & Peebles (Fl. PI. Ariz. 482,—1942), will prove to belong here. Astragalus naturitensis Pays. var. typicus Barneby, nom. nov. 4. nat- uritensis Pays., Bot. Gaz. 60: 377 (1915), sensu stricto. A. arietinus var. stipu- laris Jones, Proc. Calif. Acad. Sci. II, 5: 654 (1895). A. stipularis Jones, op. cit. 655 (1895), nomen provisorium. Xylophacos stipularis (Jones) Rydb., Fl. Rocky Mts. 1603 (1917) (non op. cit. p. 504). CoLorapo: Naturita, Montrose Co., Payson No. 360 (RM, type of A. naturitensis). McElmo Creek, Montezuma Co., East- wood in 1892 (PO, type of A. arietinus var. stipularis). McElmo 2 Astragalus desperatus var. typicus Barneby, nom. nov. A. desperatus Jones, Zoé 2: 243. 1891, sensu stricto. 88 LEAFLETS OF WESTERN BOTANY [VOL. V, NO. 5 Creek, 5 miles west of Cortez, Montezuma Co., Ripley & Barneby No. 8405 (CAS, NY). In an account of Astragalus § Argophylli (Amer. Midl. Nat. 37: 474,—-1947) I had occasion to mention A. arietinus var. stipu- ' laris, since I thought that the fragmentary type, now flowerless, might prove to belong to some form of A. zionis Jones. The topotypes lately obtained, very characteristic on account of the broad, closely imbricated stipules and leathery pod, furnish proof of the real identity of Miss Eastwood’s plant, which has been a baffling problem to all students of the genus. They are found to compare favorably also with the type of A. naturitensis, a species equally rare and misunderstood, which Jones (Rev. Astrag. Index,—1923) and subsequently Rydberg (N. Amer. FI. 24: 319,—-1929, with note of interrogation) unjustly reduced to A. desperatus. The collection recently reported as A. naturi- tensis from the Mesa Verde now proves to represent a distinct variety, viz., Astragalus naturitensis Pays. var. deterior Barneby, var. nov., a var. typico nob. calyce breviori campanulato, corolla subdimidio breviori ochro- leuca, leguminis valvulis chartaceis (nec coriaceis) aliisque nonnullis notulis satis diversa. —A. naturitensis sensu Barneby, Leafl. West. Bot. 4: 51, fig. 16, 17, opp. p. 56,—1944; non Pays. Cotorano: ledges of sandstone cliffs on the south rim of the Mesa Verde, and on detrital slopes beneath the Cliff Palace, Montezuma Co., 21 May 1943, fl. and fr., Ripley & Barneby No. 5359. Type in Herb. Calif. Acad. Sci., No. 313535. Similar habi- tat, near the Sun Temple, No. 8397. The var. deterior is essentially quite similar to A. naturitensis proper, but it is a more delicate plant with subfiliform petioles and peduncles, the stipules are shorter and narrower, the pod is of thinner texture and less heavily reticulate, the calyx is campanulate as opposed to cylindric, and the corolla one half shorter. Were it not for a similar type of variation in the related A. sparsiflorus Gray (with its var. majusculus Gray) and in A. desperatus (as shown above), I would incline to think var. de- terior specifically distinct. Payson emphasized the bicolored flower of A. naturitensis, saying of it (ex char.): “corolla . . . con- spicuously bicolored, standard white . . . apical portion of the lateral petals and blunt keel red’’. If one substitutes “purple” for “red”, this description neatly fits the plant of McElmo Creek. FEBRUARY, 1948] PUGILLUS ASTRAGALORUM IX 89 By contrast the petals of var. deterior are all of a drab straw-color, faintly purple-veined. Since the last two species are related, have been constantly confused, and together form a somewhat isolated group in the genus, it has been thought useful to contrast their principal characters ina Key To A. desperatus, A. naturitensis, AND THEIR VARIETIES 1. Pod horizontally spreading or declined, abruptly narrowed at apex into a deltoid beak, the valves spreading-pilose with long hairs seated on a pustular base. 2. Calyx campanulate, including the teeth 3—6 mm. long; corolla (bicolored) relatively small, the keel 6.5—8 mm. long; racemes (except in starveling, small-flowered individuals) elongating and 2—7 cm. long in fruit; lower reaches of the Green and Grand rivers and their immediate tributaries in southwest Colorado, southeast Utah, and down the Colorado River to the Moenkopi Wash in Coconino Co., Arizona........ A. desperatus var. typicus 2. Calyx cylindric, 6.5—8 mm. long; corolla (purple throughout) longer, the keel 10—11 mm. long; racemes subcapitate, not exceeding 1 cm. long in fruit; southern Navajo Co., Arizona.................244 MMe A tion haus tartar asd ayah ayaa Pea A. desperatus var. conspectus 1. Pod loosely erect, gradually acuminate into a lanciform beak, the valves strigulose with short, filiform, appressed hairs. 3. Calyx cylindric, 6-8 mm. long; corolla (bicolored) large, the banner 14—15 mm., the keel 11 mm. long; pod leathery, the valves strongly rugose-reticulate on the angles; stipules large and conspicuous, 3—7 mm. long, 2—5 mm. broad; Montrose and Montezuma coun- ties; Colorado, at 5400—5800 ft. altitude... . 6.0... 00... bee enone aia ais a pie Me Ne ersigalnetn te er Rien Se owas A. naturitensis var. typicus 3. Calyx campanulate, 3.5—5 mm. long; corolla (ochroleucous) smaller, the banner 10—11 mm., the keel 7—8 mm. long; pod chartaceous, delicately reticulate; stipules smaller, 2—4 mm. long, 1.5—2 mm. wide; Mesa Verde, southwest Colorado, at 6700—6900 ft. altitude SE CRE rie ek bor ee eee tee A. naturitensis var. deterior NEW RECORDS FOR SCIRPUS BY A. A. BEETLE University of Wyoming, Laramie ScIRPUS SUBTERMINALIS Torr. First record for Wyoming: John Reed No. 1583, collected in Teton County, Jackson Hole, beaver ponds east of Moran, August, 1947. 90 LEAFLETS OF WESTERN BOTANY [VOL. Vv, NO. 5 ScIRPUS NEVADENSIS Wats. Second record for Wyoming: Beetle No. 5288, collected on alkali gypsum flats, south of Lara- mie, Albany County, August, 1947. Only previous record: E. Nelson No. 4987, collected at Berthaton, Sweetwater County, July, 1898. ScIRPUS RUBROTINCTUS Fern. f. CONFERTUS Fern. First record for Wyoming: Beetle No. 4806, collected on the North Platte River at Saratoga, Carbon County, July, 1947. Growing with normal plants. ScirPUS MUCRONATUs L. First record for California and North America: Bellue, in February, 1947, recollected in June and August, 1947, as a weed in rice fields near Biggs, Glenn County, California. Cf. Bellue in Bull. Calif. Dept. Agric. 36: 91-96 (1947). Authenticating herbarium specimens may be found in the author’s Scirpus collection which has been given to the Rocky Mountain Herbarium, University of Wyoming. CONCERNING A CALIFORNIA CUDWEED BY JOHN THOMAS HOWELL After first finding Micropus amphibolus Gray in Marin County west of Fairfax and seeing how very different it is from M. californicus F. & M., I was puzzled why the two could be confused for each other, as has been intimated by Jepson. Later I found the plant not only at other places in Marin County but also in Lake and Alameda counties, and as a result of this wider field acquaintance I came to suspect that there might be more than a specific difference between the two plants. After recent herbarium studies I have concluded that this rather tentative suspicion has a real basis and that M. amphibolus properly belongs in Stylocline. In Micropus L. the receptacle is usually broader than long (it may be flat or even centrally depressed) and it is more or less parted into short ray-like projections to each of which is at- tached a sac-like pale and an enclosed fertile pistillate flower. Because of this flattened receptacle, the fruiting pales are ar- ranged radially in one plane and not at all imbricate. In Stylo- FEBRUARY, 1948] A NEW CALIFORNIA CASTILLEJA 91 cline Nutt. the receptacle is longer than broad, oblong to nearly linear, and is only a little roughened by the points where the pales and flowers are attached. The pales are attached spirally along the receptacle and the winged margins are spirally imbri- cate in a subglobose, slightly conic head. Aside from this chief structural difference between the genera, the following second- ary differences generally obtain. In Micropus the pale becomes bony in fruit and its aperture is not hyaline-margined, the at- tachment of the corolla to the achene is decidedly lateral, the perfect central flowers are not subtended by pales, and the sterile central achenes have no pappus. In Stylocline the pale is mem- branaceous and its aperture is hyaline-margined, the attachment of the corolla is apical or somewhat lateral, the perfect central flowers are usually subtended by plane pales, and the sterile central achenes generally have a few delicate pappus-bristles. Because M. amphibolus agrees with Stylocline in these char- acters, the following name is proposed: Stylocline amphibola (Gray) J. T. Howell, comb. nov. Micropus amphibolus Gray, Proc. Amer. Acad. 17: 214 (1882). When originally describing this plant, Gray recognized its relationship to Stylocline but concluded it was nearer Micropus. Even in referring it to Micro- pus, he indicated his uncertainty in the specific name he assigned to it. The species is relatively rare and has been found in central California only in Alameda, Contra Costa, Marin, Sonoma, and Lake counties. A NEW CALIFORNIA CASTILLEJA BY JOHN THOMAS HOWELL Castilleja Leschkeana J. T. Howell, spec. nov. Herba perennis; caule circa 1 m. alto erecto, ramoso infra inflorescentiam, glabro basi sparse hir- sutulo medio pilescenti supra; foliis infinis oblongo-linearibus, 0.5-1.5 cm. longis, foliis mediis anguste elliptico-oblongis acutis integris vel prope apicem lobum unum brevem ferentibus, 6-7 cm. longis, 0.8-1.5 cm. latis, hirsutulis et paulo scabris margine, foliis supremis lanceolatis vel ovatis integris vel 1- vel 2-lobatis, 4-5 cm. longis, 1.5-2.3 cm. latis, subcinereis pilis densis et subvelutinis; inflorescentia densa rotundata, circa 6 cm. lata pilosa et paulo viscidulo-glandulosa, bracteis 2.5—4 cm. longis, 2-3 cm. latis, sub- cinereis et subvelutinis late cuneiformibus 3-5-fissis circa ad medium, lobo medio latissimo interdum denuo lobato dentatove, lobis rosaceis plerumque acutis; calyce 2-2.5 cm. longo, circa equaliter infra medium fisso ventro ws LEAFLETS OF WESTERN BOTANY [VOL. V, NO. 5 dorsoque, segmentis integris lobatis vel ad medium fissis anguste oblongis subacutis breviter denseque pilosis paulo viscidulis rosaceis; corolla 2.5-3 cm. longa, galea longitudine calycis vel paulum longiore, galea tuboque aquilongo, labio inferiore atrovirenti, 1.5 mm. longo, valde saccato corru- gatoque, dentibus lateralibus medio longioribus; stigmate breviter exserto, valde bilobo, lobis anguste oblongis; fructu ignoto. Type in Herb. Calif. Acad. Sci., No. 339324, collected in swampy ground behind the dunes on Point Reyes Peninsula, Marin County, California, June 4, 1947, J. T. Howell No. 23220. This tall and handsome Castilleja belongs to the group that is typified in California by C. miniata Dougl. but from that species and its relatives it differs in its flowers and the peculiar subcinereous pubescence of the upper leaves and floral bracts. In general habital appearance, the plant looks more like C. acu- minata (Pursh) Spreng. from Alaska and the Aleutian Islands than any other species along the Pacific coast. This resemblance may be more than superficial because C. Leschkeana, in its cold coastal swamp, is associated with several plants whose range ex- tends northward to Alaska, a boreal floristic relation that prob- able goes back to the southward dispersal of this flora and to its tenuous survival on the central California coast. Castilleja Leschkeana is known from only a single plant and its general rarity may account for the fact that so showy and dis- tinct a species has been unknown until now. Naturally one is reluctant to describe as new any entity known from only a single individual but it is representative of the peculiarly insular type of flora found on Point Reyes Peninsula in which many highly localized and rare plants have been found. Thus, on the same trip on which the Leschke paintbrush was discovered, the fol- lowing plants were found associated with it in the same swale, each represented, even after intensive search, only by a single plant or colony: Eleocharis pauciflora (Lightf.) Link, the first- known occurrence of this boreal and montane species on the coast of California; Carex Buxbaumii Wahl., otherwise known in the California Coast Ranges only from Pitkin Marsh, Sonoma County, and Big Lagoon, Humboldt County; and Juncus fal- catus E. Mey., the only occurrence known in Marin County. It is a pleasure to name this plant for my friend, Dr. Hans Leschke, critical and enthusiastic amateur and my agreeable companion on many botanical trips in Marin County. VoL. V | No. 6 ‘LEAFLETS of WESTERN BOTANY . CONTENTS PAGE WNotés on the Arizona Flora... «.) 0) 3) eo oe a 9B Rosert A. DARROW meilew Western Violet?