Fe diel aad bess.) ate ret (eae Mela aerogenes seiyesbanchen. can Pty hoe x ase Se iesreacedatenero bets tafetenere” otett sheamtete net nga _ oe! ice ewan Seen teas tate Dl come ptvree m ohet Ae by he amma a Beam nah a eckghte ep) manatee gin ly Seen > “ aoe poses ne We» - — : 2 a ae a ie A ee ie eS mere aan ‘ -s = aera a Sel _ o - - a4 = . a» et “ t : iret aoe ~“ a pee 7 7 - ae ae nent f Pos oa = =. ~- i egeatiige . — i ae SAERAR EE ee — _ — Feiaeak si torcviaan cOT ad SOE ett te em 2 p . ’ 4 £ i ! bit ly VAD ie WA Ont DEAPREES OF WESTERN BOTANY VoLuME IV SAN FRANCISCO, CALIFORNIA 1944-1946 Se Owned and Published by | _ Atice Eastwoop anp Joun Tuomas Howert nes , We Printed by ' THE JAMES H. BARRY COMPANY SAN FRANCISCO ei “4 rn \ lal oA | bye oe ae : 4 ae 2c ee MOL. LV. No. 1 LEAFLETS of WESTERN BOTANY CONTENTS Mariposa, a Neglected Genus Rosert F. Hoover A New Manzanita from Baja California . ALIcE EAstwoop A New Fruticulose Eriogonum JoHN THomAs HoweELu A Note on the Distribution of Some Western Dicotyledons R. C. BARNEBY A Reconsideration of the Genus Miltitzia . JoHN THomas Howe. SAN FRANCISCO, CALIFORNIA Fesruary 2, 1944 PAGE 12 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. PACU UU aU Rees Re es i TTT aM ETAT AU aa a Owned and published by Auicz Eastwoop and Joun THomas Howe. NEW y¥< MARIPOSA, A NEGLECTED GENUS BOTAN! BY ROBERT F, HOOVER bi\ Mariposa (Wood) Hoover, gen. nov., Liliacearum. Calo- chortus sect. Mariposa Wood, Proc. Acad. Philad. 1868: 168. The genus Calochortus as currently accepted includes at least two, and probably more, sharply differentiated groups. Although Mariposa has generally been recognized as a section distinct from typical Calochortus, its truly distinctive features appear to have been overlooked ; at least, no mention of most of these features can be found in any publication. During the twenty-eight years of my residence in California, I have been particularly interested in that group of plants, and my observations have convinced me that two different genera exist where only one has been recog- nized. The two genera differ in the following respects. Leaves never grooved; stigma-branches slender, on a short style; seeds rather coarsely reticulate-pitted, not much flattened, dark purple or purple-tinged, at least in all Californian species ; chromosomes always EAL Gy Sou oU ND) Cee Cot 9 se) aR EL ee wpe eRe pee a oP en ROS SRE Calochortus Leaves all with a sharp groove extending the entire length on the upper side; stigma-branches stout, sessile on ovary?; seeds very minutely reticulate, the reticulations not forming evident pits, much-flattened, whitish or greenish; chromosomes usually in a multiple of seven or CIENCIA pe Ce A Ac le a to a on es Mariposa Although it must be admitted that I have been unable to examine mature seeds of all the species, no species has yet been seen which in any way combines the characters of the two groups as listed above. It therefore does not seem consistent with scien- tific methods of classification to place them in one genus. Prob- ably the generic characters have been overlooked because adequate material of living plants has not been available to botanists. If it is still considered desirable to unite the two groups in one genus, it should be remembered that there are many pairs of genera, at present universally regarded as distinct, which resemble each other more than do Calochortus and Mariposa, Examples are Agrostis and Calamagrostis, Vicia and Lathyrus, Medicago and Melilotus, Stellaria and Arenaria, Clarkia and Godetia, Lamium and Stachys, and Aster and Corethrogyne. If genera eA Information as to chromosome numbers was obtained from data pub- lished by Beal (Bot. Gaz. 100:528—547,—1939). 2 This key character refers to the condition in anthesis. In fruit, the capsule in most species of Mariposa containing no seeds and not becoming expanded, presents the appearance of a style. * Leaflets of Western Botany, Vol. IV, pp. 1-16, February 2, 1944. GARI 2 LEAFLETS OF WESTERN BOTANY [ VOL. IV, NO. I are to be based on the actual relationships of plants rather than on mere custom or tradition, a uniform procedure ought to be followed in all such cases. It may be noted that there is a third group of species occur- ring in California, represented by Calochortus Weedu Wood, which is different from both Mariposa and true Calochortus. These species have fibrous-coated bulbs and probably constitute another distinct genus. However, I have had very little oppor- tunity to study living plants of the Calochortus Weedu group and consequently am unwilling to venture a definite statement as to its generic status. Field and garden observations have convinced me that the species listed below are all valid. There is a considerable number of other species which undoubtedly should be transferred to Mariposa, but as I am not familiar with them, I refrain from making any change in their classification. Mariposa catalinz (Watson) Hoover, comb. nov. Calochortus cata- line Watson, Proc. Amer. Acad. 14: 268 (1879). Mariposa clavata (Watson) Hoover, comb. nov. C. clavatus Watson, Proc. Amer. Acad. 14: 265 (1879). Mariposa Leichtlinii (Hooker) Hoover, comb. nov. C. Leichtlinii Hooker, Bot. Mag. ser. 3, 26: pl. 5862 (1870). Mariposa lutea (Douglas) Hoover, comb. nov. C. luteus Douglas ex Lindley, Bot. Reg. 19: pl. 1567 (1833). Mariposa macrocarpa (Douglas) Hoover, comb. nov. C. macrocarpus Douglas, Trans. Hort. Soc. Lond. 7: 276 (1828). Mariposa pratensis (Purdy) Hoover, comb. nov. C. superbus var. pratensis Purdy ex J. T. Howell, Leafl. West. Bot. 1:12 (1932). Mariposa splendens (Douglas) Hoover, comb. nov. C. splendens Douglas ex Bentham, Trans. Hort. Soc. Lond. ser. 2, 1: 411 (1834). Mariposa superba (Purdy) Hoover, comb. nov. C. superbus Purdy ex J. T. Howell, Leafl. West. Bot. 1:11 (1932). Mariposa venusta (Douglas) Hoover, comb. nov. C. venustus Douglas ex Bentham, Trans. Hort. Soc. Lond. ser. 2, 1: 412 (1834). In addition to those listed above, there are two species of Mariposa which apparently have never been named. Both of them, so far as they have received any attention at all, have been confused with M. venusta and universally referred to that spe- cies. The three species, however, are readily distinguished by the following features. Stem with several (usually 3 to 5) bulblets at base; glandular area of petal lunate or nearly rectangular, transversely elongate; capsule be- FEBRUARY, 1944] MARIPOSA, NEGLECTED GENUS 3 fore dehiscence stout (narrowly lanceolate in outline), 4—6 cm. (ari? TE a AE Ne RS Cee ERT a tM: 5 2M AS eS M. argillosa Stem with one bulblet, rarely two or none; glandular area of petal rounded or somewhat quadrate, scarcely if at all transversely elongate; capsule before dehiscence rather slender (linear in outline), 6—8 cm. long. Petals without markings above the lowest third, with a red spot sur- rounding the glandular area and frequently a small round red spot immediately above it; seeds 5—6 mm. long, 2.75—3 Pa ey eek a 28 ae ee ee M. simulans . Petals with a conspicuous central dark spot and frequently also a lighter spot near the apex; seeds 4.5—5.25 mm. long, 1.75—2 bEbLSI RNP (GCet ae eR A PRA DE eRe Pe Nee uN oA DUE SIRES gt M. venusta Mariposa argillosa Hoover, spec. nov. Glauca; caule basi aliquot bulbos ferenti; petalis albis usque ad purpureis, prope medium macula rubra spe flavescenti-circumdata tinctis, areola glandulosa transverse orthogonia vel lunata; capsula lanceolata; seminibus 5.5—6 mm. longis, 3 mm. latis. Stem with several bulblets at base; herbage glaucous; petals white to purple, with red spot near middle, the red spot often surrounded by pale yellow, the glandular area transversely rectangular or lunate; capsule lanceolate in outline; seeds 5.5—6 mm. long, 3 mm. wide. Mariposa argillosa resembles M. venusta in the color and markings of the petals, but the shape of the glandular area of the petals and the shape of the capsule indicate a closer relationship to M. lutea, to which, remarkably enough, no botanist appears to have referred it. In some localities it grows in company with M. lutea, but under such circumstances remains genetically dis- tinct from that species. Dried specimens, in which the shape of the glandular area is not easily visible, are readily confused with M. venusta, and are usually found labelled as that species in herbaria. This species is always found in hard clay soil in areas of volcanic or metamorphic rocks. The following specimens have been seen, indicating a distribution in the California Coast Ranges from San Mateo County to San Luis Obispo County. San Mateo Co.: Portola, Elmer No. 4763 ; Redwood City, Abrams No. 7499. Santa Clara Co.: near Coyote, Ferris No. 823. San Benito Co.: Arroyo Dos Picachos, J. T. Howell No. 13814, Hoover No. 3488 (type). San Luis Obispo Co.: San Luis Valley, Summers in 1882. 8 The measurements given refer only to fully developed seeds not dis- torted by crowding. 4 LEAFLETS OF WESTERN BOTANY [ VOL. Iv, NO. I Mariposa simulans Hoover, spec. nov. Habitu M. venuste accedens; petalis albis extus szpe dilute rosaceis, circum areolam glandulosam rutilis et sepe maculam rutilam proxime supra areolam ferentibus; seminibus 5—6 mm. longis, 2.75—3 mm. latis. Plant with appearance of M. venusta; petals white, often pink on the outside, the glandular area surrounded by bright red, and often also with a small red spot immediately above ; seeds 5—6 mm. long, 2.75—3 mm. wide. Mariposa simulans is obviously closely related to M. venusta. The seeds of the two species differ in shape, but plants in flower can hardly be distinguished except by the color markings of the petals. In the latter respect, M. simulans resembles M. cataline, a species which is quite different in characters of the fruit. This superficial resemblance to /. cataline has suggested the proposed specific name. The distinctness of M. stmulans from M. venusta may be subject to doubt, but it should be noted that, although both species occur in the same region, no intergrading plants or apparent hybrids between them have ever been collected. This is a local species, known to occur only in a limited section of San Luis Obispo County. I have observed it in the field only once, but have since studied it under cultivation. The following collections have been seen: La Panza, Keck No. 2815 (type, Herb. Univ. Calif.) ; Trout Creek, Condit in 1908 ; old road to Pozo, Eastwood No. 15144. A NEW MANZANITA FROM BAJA CALIFORNIA BY ALICE EASTWOOD Arctostaphylos australis Eastwood, spec. nov. Fruticosa, divaricate intricateque ramosa, dense foliosa; caulibus senioribus glabris, junioribus puberulis; foliis subpallidis, ellipticis, apice acutis, basi obtusis, puberulis, in juventute rubescentibus, maximis 3 cm. longis et 2 cm. latis, petiolis circa 5 mm. longis; paniculis in alabastro erectis, ramis gracilibus, rubris, bracteis minutissimis; floribus albis, 4—5 mm. longis, segmentis calycis glabris, membranaceis, rubescentibus; fructu glabro, 4 mm. diametro, nuce solida. Type in Herb. Univ. Calif., Los Angeles, collected in flower in Pine Canyon near San Antonio Mesa, Baja California, March 27, 1940, by Carl Epling and Wm. Robinson, and in rather young fruit by the same collectors on the ridge above Pine Canyon near San Vincente, Baja California, April 25, 1940. The striking inflorescence erect in bud with almost thread- like branches and minute bracts at once suggests affinity of this plant with A. insularis Greene from Santa Cruz Island, Cali- FEBRUARY, 1944] NEW FRUTICULOSE ERIOGONUM 5 fornia. It differs, however, in the puberulent indument on leaves and young stems, the smaller flowers, the paler green smaller leaves, and the small berries each containing a single nut. The young leaves have a rosy tinge and the budding panicles are a dark red. A NEW FRUTICULOSE ERIOGONUM BY JOHN THOMAS HOWELL n\ Eriogonum Ripleyi J. T. Howell, spec. nov. Fruticulus depressus ex radice lignea elongata crassiuscula, ramis divergentibus prostratis vel ad- scendentibus, cortice tenuissime rugulosa nigro-grisea vestitis, et ramulis abbreviatis folia in rosula terminali ferentibus supra basibus foliorum veterum persistentibus conferte vestitis infra et ramulis raris elongatis folia separata ferentibus tomento albo vestitis, caulibus 5—15 cm. longis, ad 5 mm. diametro; foliis 2—6 mm. longis, 0.5—1 mm. latis, anguste oblanceo- latis, valde revolutis, tenuiter villosis supra, dense tomentosis infra, acutis, angustatis in petiolum brevem basi late expansum et membranceo-margi- natum; pedunculo ebracteato perbrevi, 1—3 mm. longo, glabro, solitarium involucrum ferenti; involucris campanulatis, 3 mm. longis, 3 mm. latis, tenuissime villosis vel subglabris, late submembranaceo-marginatis, promi- nenter lobatis, lobis 3 ad 5, inzequalibus, triangularibus, ad 1 mm. longis, seepe mucro tis, involtucris basi late cuneatis; bracteolis linearibus sub- villosis et yee glanduloso-papillatis ; pedicellis sparse pilosis; peri- anthiis eximiis, segmentis albis medio rubido-costatis, margine crenulatis, plus minusve undulatis, subemarginatis, intus levissime glanduliferis prope basin, segmentis exterioribus suborbicularibus, 3—3.5 mm. longis, 3.5—4 mm. latis, interioribus minoribus, late obovatis, 3 mm. lomgis, segmentis perianthii basi paulo crassiusculis et conjunctis circa 0.5 mm.; filamentis prope basin dense villoso-tomentosis ; ovario glabro; achenio ignoto. Type: Herb. Calif. Acad. Sci. No. 311,671, collected on steep declivities of sandy clay on the edge of a sandstone mesa where it was associated with pinyons, 13 miles southwest of Frazier’s Wells, el. 6000 feet, Coconino County, Arizona, May 13, 1943, by H. D. Ripley and R. C. Barneby, No. 5226. The most remarkable character of E. Ripleyi is the short stalk which carries a solitary involucre and which is without bracts either at the base or at the top. If there were bracts at the base of the structure, it might be interpreted as the so-called pedicel which is the distinctive mark of the subgenus Ganysma, but since no bracts are discernible and since the character of the foliage and flowers does not suggest an affinity to that group, the struc- ture is undoubtedly an ebracteate peduncle. The only other 6 LEAFLETS OF WESTERN BOTANY _ [VOL. IV, NO. I species in Eriogonum that I know which has a similar structure is E. cespitosum Nutt., a rather variable species widespread in the western United States. Although these two species are characterized by a structure so unusual, I do not believe that it indicates an especially close relationship since there seem to be fundamental differences in the flowers and foliage. In the flowers of E. cespitosum, the perianth-segments are united below and drawn down into a slender solid stipe-like base which is diagnostic of the subgenus Eueriogonum, and although this structure is sometimes quite short in certain forms of E. cespitosum, it is always present. In E. Ripleyt, no such struc- ture is indicated: the perianth-segments are united below, but they form a shallow and hollow bowl that is broadly cuneate or rounded above the callous base. Flowers of this kind are found in several groups of the subgenus Oregonium. The arrangement of leaves in E. cespitosum and E. Ripley is quite different, but since both species are depressed dwarfs with much-condensed habits, the divergence in this character is almost concealed under a marked habital simulism. In the more loosely branched forms of E. cespitosum, however, the character of the shoot may be discerned: a naked stem bearing a congested terminal tuft of leaves, a type of shoot which is pro- duced by the excessive elongation of the first internode of the season’s growth and the rosulate abbreviation of all others. As in the character of the perianth-base, this type of branch- ing in E. cespitosum is highly characteristic of the subgenus Euertogonum. In E. Ripleyi, the character of the vegetative shoot is quite different and I believe holds the most obvious clue to its relation- ship. The usual type of leaf-arrangement for the species is found on the numerous spur-like branches, where the leaves are rosulate- congested because of the very short internodes. The stems below this tuft of leaves are entirely covered by the persistent imbri- cated bases of leaves of past seasons and it is only on the main stems of the plant after these long-persistent bases have been finally sloughed off that a blackish-gray finely furrowed bark is apparent. Besides this shortened kind of shoot which gives to E. Ripleyi its characteristic appearance, there is another kind which appears to be occasional or rare. In this the internodes FEBRUARY, 1944] SOME WESTERN DICOTYLEDONS i are relatively elongate so that the leaves are quite discrete and the whole is clothed by a white tomentum. This is a vigorous type of shoot representing a season’s growth and it may arise directly from an old basal stem or from the end of one of spur- branches. It does not appear to produce an inflorescence, but after the season’s growth it passes into the usual slow-growing type of stem which alone seems to be floriferous; and from the axils of its spaced leaves, in subsequent seasons secondary spur- branches may arise. In the abbreviated branches of E. Ripley, I can see only a remarkable condensation of the plant axis that is scarcely ex- ceeded in Eriogonum even in the cushion-plants of alpine sum- mits, a character, however, only specific in significance; but in the rarer elongate type of shoot, I can perceive a direct phylo- genetic connection between E. Ripley: and such southwestern dwarfs as E. bicolor M. E. Jones and E. pulchrum Eastw. of the E. microthecum group. In these, both kinds of branches also occur, but the more common typical kind is the elongate sort which is so rare in E. Ripleyi. Also there are marked corre- spondences between the plants in the leaves, involucres, and flowers. The revolute leaves with expanded bases are quite similar in the three; the membranous margin of the involucre in E. Ripleyi has a textural equivalent in E. bicolor; and the bowl- like flower-base which is so shallow in EF. Ripley: is much deeper and more or less bulbous-inflated in the other two. Hence, as far as the ebracteate peduncle is concerned, I would interpret it as yet another morphologic expression of the marked conden- sation and reduction which seems to characterize E. Ripleyi, and I should judge that in this outstanding character, FE. Ripleyi might bear to the E. microthecum group much the same relation that E. cespitosum bears to the E. umbellatum group. A NOTE ON THE DISTRIBUTION OF SOME WESTERN DICOTYLEDONS BYR. G, aac eh \ Among the plants from the southwestern deserts collected during recent years by Mr. H. D. Ripley and the writer there have inevitably turned up a few records of special interest, whether on account of the rarity of the species involved or as evidence of 8 LEAFLETS OF WESTERN BOTANY [ VOL. IV, NO. I an extended range. Some of these have already found their way into print through monographic studies, but others, particularly those belonging to groups of which our knowledge is more or less up to date, seem unlikely to be published for many years, and it is in order to place them on record that these notes have been prepared. It has also seemed a convenient opportunity to set down a few observations on the habitat or peculiar attributes of the less common species, particularly when such details are not generally known. The numbers cited below are derived from the collections of Mr. Ripley and the writer, and corresponding specimens have been deposited in the herbarium of the California Academy of Sciences. ATRIPLEX GRACILIFLORA M. E. Jones. Colorado: alkaline clay knoll 16 miles north of Loma, Mesa Co., alt. 4950 ft., No. 5471. This most distinct and interesting species is known to me only from one or two collections from southeastern Utah, but like many plants endemic to the Navajo Basin it extends eastward into Colorado. Our material agrees with Jones’ diagnosis in being annual, in spite of Rydberg’s suggestion (Fl. Rocky Mts. & Adj. PI. 248,—1917) in his key to the genus that it belonged with the ‘shrubby or perennial species. This error may have arisen from the fact that Jones, in searching for an affinity for A. graciliflora, stated that “this remarkable plant must rank near to the shrubby A. canescens [James] in the fruit character, though so unlike it in all other respects.”” This observation seems to smack of the superficial, for the only feature common to the fruiting structure of A. canescens and A. graciliflora lies in the decurrence of the bracts along the pedicel of the pistillate flower. In A. graciliflora, the orbicular, obscurely sinuate fruiting bracts are strongly con- nate throughout their whole circumference except for a minute terminal pore through which the style is exserted in anthesis, and in this character the species recalls the monotypic genus Zuckia Standl., while the entire fruit resembles to a marked degree the samara of a Ptelea. STANLEYA VIRIDIFLORA Nutt. California: Smoke Creek, northeast of Viewland, Lassen Co., alt. 4400 ft., No. 5761a. Stanleya viridiflora, although mentioned by Jepson (FI. Cal. 2:20, 1936) as having been collected near the boundary of California FEBRUARY, 1944] SOME WESTERN DICOTYLEDONS 9 in Esmeralda County, Nevada, has not actually been reported from the state. The nearest station recorded by Rollins in his monograph of the genus (Lloydia 2: 124,—1939) is in south- eastern Oregon or perhaps adjacent Nevada, in any case some 180 miles distant from the present locality. In Lassen County, exactly as I have seen it in the neighborhood of Elko, Nevada, S. viridiflora occurs exclusively on denuded slopes of alkaline clay such as are the home of species of Miltitzia A. DC. and Scutellaria nana Gray. SoPHORA SERICEA Nutt. Utah: Enterprise, Washington Co., alt. 5750 ft., No. 4968. Though widely distributed over the plains east of the Rockies from South Dakota to Texas, and extending westward through Arizona as far as Yavapai County, S. sericea has apparently remained unknown north or west of the Colorado River. In the western part of its range the species often occurs along roadsides, in disturbed places or in fallow ground where, although well-established, it has the air of being a comparatively recent immigrant. Specimens in herbaria may often be found among undetermined Astragalus. PARRYELLA ROTUNDATA Wooton. Arizona: Willow Springs north of Cameron, Coconino Co., alt. 4850 ft., No. 4876. In the summer of 1942 a few specimens of this curious legume were discovered by Mr. Ripley at the foot of the great escarpment known as the Echo Cliffs at a point about a mile south of Willow Springs. They grew in crevices of hard red sandstone pavement, forming low, gnarled bushes of rounded outline and only two or three decimeters high. In early June, when the related but very different Parryella filifolia T. & G., which grew nearby on dunes along a watercourse, was only then beginning to expand its first racemes, P. rotundata was already in mature fruit, and no flowers were obtained. According to Kearney and Peebles (Fl. Pl. Ariz. 452,—1942), the species, although first collected half a century ago, has been known only from the type-locality near Winslow, where Wooton stated that he encountered but a single plant, and it thus appears to be one of the rarest of our western Fabacee. Well-developed leaves with their very numer- ous and crowded orbicular leaflets tapering upwards along the rachis somewhat resemble the fronds of Notholena sinuata var. crenata Lemm. 10 LEAFLETS OF WESTERN BOTANY [ VoL. IV, NO. I CymopTerus GILMANI Cov. & Mort. Nevada: Spotted Range, southwest Nye Co., alt. 4500 ft., No. 3418; north end of Desert Range, Clark Co., alt. 3500 ft., No. 3337; Pintwater Range, Clark Co., No. 2913. Also observed on Sheep Mt. near Jean, Clark Co. Although not uncommon throughout the low limestone mountains of southwest Nevada, this species seems to have been reported only from Death Valley in adjacent Cali- fornia. On one deeply eroded calcareous hill at the foot of the Pintwater Range, C. Gilmani was noticed as the dominant species on hot banks of detrital gravel mixed with clay: with it grew Acamptopappus Shockleyi Gray and Penstemon petiolatus Bdg. As they approach maturity the large handsome fruits are some- times suffused with rose or purple, though normally the wings at least are white. CyMOPTERUS BASALTICUS M. E. Jones. Nevada: foothills of the Snake Range, 8 miles west of Baker, White Pine Co., alt. 5800 ft., No. 3566. An exceedingly rare plant, long known only from the type station west of Wawa in Beaver County, Utah, where it was said to occur at an altitude as high as 7000 feet. In Nevada it is confined to an area of sterile, alkaline clay hills, where it is associated with Hermidium alipes Wats., Artemisia spinescens D. C. Eat., and Penstemon dolius M. E. Jones. CymorTerus Ripteyi Barneby. This species is evidently more widely distributed in southern Nevada than was at first supposed, having now been collected in Lincoln County (near Crystal Springs, No. 4410 and 4411; Pahranagat Mts., No. 4405) and in yet another locality in Nye County (east of Warm Springs, No. 4433). It is strictly confined to sand-dunes or deep sandy soil of open valleys, and sometimes occurs in great abundance. CymoprTerus Jonestt C. & R. Nevada: detrital slopes be- neath cliffs along the Muddy River, 2 miles south of Caliente, Lincoln Co., No. 2517. Rather common in the juniper belt of the Highland Range, whence it had seeded down into the gorge of the Muddy River: otherwise known only from southwestern Utah. SWERTIA ALBICAULIs (Griesb.) O. Kze. (? var.). Nevada: Toquima Range, between Belmont and Manhattan, Nye Co., on sandy slopes among pines, alt. 7400 ft., No. 3670 (fl.) and 3999 (fr.). According to St. John’s recent account of Swertia (Amer. FEBRUARY, 1944] SOME WESTERN DICOTYLEDONS II Midl. Nat. 26: 1—29,—1941) this plant keys out to the dich- otomy embracing S. albicaulis and S. modocensts St. J., but it does not entirely agree with the description of either. With its strongly pubescent herbage, opposite leaves and solitary linear- elliptic fovea, it would doubtless fall, however, within the specific limits of S. albicaulis as treated by Jepson (FI. Calif. 3:94,— 1939), and without fuller understanding of St. John’s segregates from this species it seems undesirable to describe further forms. In any case it is interesting to note that, with the exception of S. gypsicola Barneby, it is the only Frasera of its type as yet encountered in Nevada, and moreover in a station strikingly remote from the northern Sierra Nevada, where its near relatives are so highly developed. AScCLEPIAS CuUTLERI Woodson. Utah: along the San Juan River west of Bluff, San Juan Co., alt. 4500 ft., No. 5382. Asclepias Cutlert, described quite recently from northeastern Arizona, has not been reported outside that state. On the dunes near Bluff, where it is associated with Eremocrinum albomargi- natum M. E. Jones, Astragalus sabulonum Gray, and Euploca albiflora (Raf.) Jtn., it is rather abundant and of characteristic habit. The few stems which rise from a slender though deep- seated vertical root are abruptly bent near the surface of the ground and spread horizontally over the sand, while the narrowly linear leaves are somewhat distichously arranged in short, flat sprays. The pendent follicles are indeed remarkable. NAMA PUSILLUM Lemmon. Nevada: Sheep Mt. near Jean, No. 3328 ; near Garnet, No. 3340. Also observed near Glendale— all localities in Clark Co., alt. ca. 2500 ft. Apparently known hitherto only from the deserts of southern California, and no- where common.’ The related N. densum Lemm., which is widely distributed on alkaline dunes of the interior and extends south- ward in Nevada at least to Callaway (No. 3638) and the Pan- cake Range (No. 3651) in Nye County, was inexplicably omitted by Tidestrom from the Flora of Utah and Nevada. StacHys RotHrocxi Gray. Utah: valley of the Virgin River near Orderville, Kane Co., alt. 5400 ft., No. 4806. Stachys Rothrockui is recorded by Epling (ap. Fedde, Repert., Beih. 80, 1934) only from the plateau region of northern Arizona and New Mexico lying between the Colorado River and the Mogollon 12 LEAFLETS OF WESTERN BOTANY _ [VOL. IV, NO. I Escarpment, so the present locality marks a somewhat surprising extension of range. At Orderville the species occurs on knolls of stiff “gumbo” clay such as have been described by J. T. Howell (Leafl. West. Bot. 3: 137,—1942) as the home of Phacelia cepha- lotes Gray, Astragalus ampullarius Wats., and Eriogonum sub- reniforme Wats., and it is interesting to record that the same rare species, with the exception of the phacelia, are again present, forming another island of a peculiar and limited association. The Stachys has a distinctive manner of growth. At this station (as well as near Williams, Ariz.) it was observed that the perennial root runs horizontally at considerable depth, giving rise at inter- vals to slender, at first simple, herbaceous stems, which thus appear to spring solitary from the earth. For six inches or so the erect primary axis is subterranean, with few, elongated, very slender yet evidently quadrangular internodes, but at its emer- gence it becomes abruptly more robust and commonly produces one or more pairs of opposite, ascending branches. The lowest, subterranean, minutely bracteate nodes frequently bear adven- titious roots. QuINcULA LoBATA (Torr.) Raf. Nevada: dry lake. bed between Crystal and Garnet, Clark Co., alt. 2050 ft., No. 3348. Quincula lobata has been fairly frequently reported from as far west as California (e. g., by Britton and Brown), but the western floras do not substantiate the record. The most reliable and recent account of its distribution as given by Kearney and Peebles (FI. Pl. Ariz. 792,—1942, as Physalis) suggests Yuma County, Ari- zona, as the western limit of its occurrence: no record at all has been seen from Nevada. This form has exceptionally large flowers. A RECONSIDERATION OF THE GENUS MILTITZIA BY JOHN THOMAS HOWELL On June 10, 1934, as I travelled northward from Susanville to Alturas on my first collecting trip into northeastern California, a species of Miltitzia was one of the most exciting and provoca- tive plants which I discovered—exciting because this was one of my early meetings with the genus in the field, provocative because the aspect of the plant with its lavender and violet corollas was FEBRUARY, 1944 | GENUS MILTITZIA 13 more like Phacelia Fremontii as 1 remembered that species in southern Californian deserts than like the Multitzia of my her- barium experience. Again four years later, on August 8, 1938, when botanizing on the saline flats of Long Valley, Mono County, California, I eagerly collected summer-dried fragments of Mil- titzia pusilla for a “new phacelia,” a plant quite beyond my experi- ence in Phacelia because it was a Miltitzia! And yet again, on May 15, 1941, the problem of Miltitzia again came to the fore, this time on the east side of Montgomery Pass, Nevada, where young plants of Phacelia gymnoclada presented an appearance so Miltitzia-like that I was at a loss in the field to be certain of their generic position. More recently, while attending to a systematic revision of Phacelia sect. Microgenetes, I was so impressed with the Miltitzia-like character of the group that a critical study of Miltitzia was determined imperative in order that the relative positions of Miltitzia and Phacelia sect. Micro- genetes might be definitely stated. Now, as a result of a detailed study of Miltitzia, I have concluded that it should be treated as a section in Phacelia, coequal with and most closely related to section Microgenetes. A review of the literature discloses that I am not the first to be puzzled by the proper position of species that have constituted the genus Miltitzia A. DC. The earliest species in the group was described doubtfully as a species of Eutoca R. Br. by Hooker and Arnott (1840), who remark that it differs “by the persistent but marcescent corolla . . .,” the character which to the present day remains the most critical for the group. In 1857, when de- scribing Emmenanthe parviflora, the second known species now referable to Miltitzia, Asa Gray discussed briefly the differences between Miltitzia and Emmenanthe Benth. and concluded that Miltitzia should be treated as a subgenus of Emmenanthe. Later, in treating the more numerous species in his Conspectus of the North American Hydrophyllaceae (1875, p. 312), he allied Emmenanthe sens. lat. to Phacelia, indicating clearly the real generic cleavage between his two subgenera by his subgeneric diagnoses (pp. 328, 329). In 1871, Torrey confused a collection of Miltitzia with the type collection of his Phacelia pusilla; and relatively recently, Brand annotated as Phacelia this same col- lection in the United States National Herbarium, although years I4 LEAFLETS OF WESTERN BOTANY [ VOL. IV, NO. I before it had served Gray as the type of Emmenanthe pusilla. The impressive similarity between Multitzia and Phacelia sect. Microgenetes led M. E. Jones to remark concerning Emmen- anthe glaberrima: “This would appear to belong to section Micro- genetes of Phacelia and probably does. ... Emmenanthe as now regarded stands on a slim foundation” (1908, pp. 51, 52). The most recent evidence of the confusing nearness of Miltitzia to Phacelia has been the renaming of an old widely distributed Phacelia as a new Miltitzia (Osterhout, 1926, p. 35). Most recent floristic workers in west American botany, as well as Brand, monographer of Hydrophyllacee, have recognized Mil- titzia as a distinct genus, the yellow marcescent persistent corolla being used to separate it from Phacelia and the turgid and ovoid capsule to separate it from Emmenanthe. Actually there is very little difference between Phacelia and Miltitzia on the one hand and Emmenanthe on the other in the character of the capsule, although this has been diagnostically stressed by Brand (1913, p. 58) and by Jepson (1943, p. 224), and in his original description of Emmenanthe, Bentham de- scribed the ovary as “oblongo-compressum” (1835, p. 281). Certainly the capsule of Emmenanthe is not “plano-compressa” as it is described by Brand (1. c.), and although in shape it is more elongate-oblong than in Miltitzia, the capsule of Miltitzia, as in many species of Phacelia, varies from ovate and oblong- ovate to truly oblong. Gray (1875, pp. 328, 329) and Jepson (1. c.) contrast the persistent style of Miltitzia with the deciduous style of Emmenanthe, but this is a distinction that is not main- tained in Miltitzia. However, the marked habital dissimilarity between Emmenanthe on the one hand and Phacelia and Miltitzia on the other is indicative of a real generic break which finds proper morphologic support in the character of ovules and pla- cente. In Miltitzia and Phacelia the ovules are characteristically attached by their sides to fleshy placente, while in Emmenanthe the pendent ovules are attached by their proximal ends to the wing-like margins of membranous placente. The reticulate seeds of Emmenanthe are strongly compressed and submeniscoidal, quite unlike the seeds of any Phacelia or Miltttzia that I know. A detailed study of all species of Miltitzia has disclosed that the marcescent corolla is the only character by which they may FEBRUARY, 1944] GENUS MILTITZIA 15 be separated from Phacelia. Even this is not diagnostic since in Phacelia there are several species characterized by marcescent corollas more or less persistent: P. saxicola A. Gray, P. Quickit J. T. Howell, P. marcescens Eastw. ; and there are several others in which the corolla is truly persistent: P. sericea (Grah.) A. Gray, P. Lyallii (A. Gray) Rydb., P. idahoensis Henderson, and P. lenta Piper. Although the bright yellow color of the corolla is very striking in several species of Miltitzia, this could scarcely be considered a generic character even if the color were constant and not frequently tinged with lavender or violet. In evaluating the characters by which Miltitzia might be allied either to Phacelia sect. Microgenetes or to Emmenanthe, I regard the character of the placentz and of the transversely corrugated seeds much more important than the character of the persistent flaves- cent corolla. Hence, Miltitzia is here presented as a section of Phacelia, it being accepted as a well-distinguished group of re- lated species that is closely allied to section Microgenetes from which it was undoubtedly derived. After my review of all entities formerly referred to Miltitzia as a genus, I regard that the following transfers are in order. A systematic revision of the section with full synonymy will be presented in a later paper. Phacelia sect. Miltitzia (A. DC.) J. T. Howell, stat. nov. Miltitzia A. DC. in DC. Prodr. 9: 296 (1845). Emmenanthe § Miltitzia (A. DC.) A. Gray, Proc. Amer. Acad. 10: 328 (1875). Phacelia lutea (H. & A.) J. T. Howell, comb. nov. Eutoca lutea H. & A., Bot. Beech. Voy. 373 (1840). Phacelia inundata J. T. Howell, nom. nov. Emmenanthe parviflora A. Gray, U. S. Pacif. RR. Reports 6: 84, tab. 15 (1857). Not Phacelia parvifiora Pursh (1814) nor P. parviflora Phil. (1895). Phacelia inundata, as I have observed it in northeastern California, is not a species confined to the littoral zone of desert lakes, but rather it is a remarkable and peculiar plant characteristic of summer-dried playas left exposed as lake-waters recede during the dry season. Although the plants do not grow in the water even in the vegetative stage, they are truly lacus- trine in that it is apparently essential for germination that their seeds be submerged for a period in the saline waters of evanescent shallows. Phacelia glaberrima (Torr.) J. T. Howell, comb. nov. Emmenanthe glaberrima Torr. ex S. Wats., Bot. U. S. Geol. Explor. 40th Paral. (King’s Exped.) 257 (1871). Phacelia adenophora J. T. Howell, nom. nov. Emmenanthe glandu- lifera Torr. ex S. Wats., l.c. Not Phacelia glandulifera Piper (1906). 16 LEAFLETS OF WESTERN BOTANY [ VoL. Iv, NO. I Phacelia tetramera J. T. Howell, nom. nov. Emmenanthe pusilla A. Gray, Proc. Amer. Acad. 11:87 (1876). Not Phacelia pusilla Buckl. (1843) nor P. pusilla Torr. (1871). That the flowers of this species are predominantly and typically tet- ramerous is one of the most intriguing facts disclosed by my studies in sect. Miltitzia. Surely here is one of the most advanced species of this highly specialized group of desert annuals: a lowly halophytic herb with flower-parts reduced, not only in size but also in number, an anomalous departure, it would seem, in Hydrophyllacee. Phacelia salina (A. Nels.) J. T. Howell, comb. nov. Emmenanthe salina A. Nels., Bull. Torr. Bot. Club 25: 381 (1898). E. foliosa M. E. Jones, Zoe 4: 278 (1893). Not Phacelia foliosa Phil. (1891) ; nor P. salina M. E. Jones ex Brand, Das Pflanzenr. IV. 251: 119 (1913), as a synonym. I am happy to acknowledge the help of Mr. C. A. Weatherby of the Gray Herbarium, whom I consulted about the proper specific epithet that should be used for this species. Phacelia scopulina (A. Nels.) J. T. Howell, comb. nov. Emmenanthe scopulina A. Nels., Bull. Torr. Bot. Club 25: 380 (1898). Phacelia inyoensis (Macbr.) J. T. Howell, comb. nov. Muiltitzia in- yoensis Macbr., Contrib. Gray Herb., n. ser. 49:41 (1917). Miltitzia pinnatifida Osterhout (Bull. Torr. Bot. Club 53: 35,—1926) is referable to Phacelia Ivesiana Torr. REFERENCES Bentham, G. 1835. Trans. Linn. Soc. vol. 17: 281. Brand, A. 1913. Das Pflanzenr. IV. 251 (Heft 59). Gray, A. 1857. U.S. Pacif. RR. Reports vol. 6: 84. 1875. Proc. Amer. Acad. vol. 10. Hooker, W. J., and Arnott, G. A. W. 1840. Botany of Captain Beechey’s Voyage, p. 328. Jepson, W. L. 1943. FI. Calif. vol. 3, pt. 2. Jones, M. E. 1908. Contrib. West. Bot. no. 12. Osterhout, G. E. 1926. Bull. Torr. Bot. Club vol. 53: 35. Torrey, J.,ex S. Wats. 1871. Bot. U. S. Geol. Explor. 40th Paral. (King’s Exped.), p. 253. Two Laxe County Recorps. In May, 1943, while on a collecting trip with Mr. M. S. Baker, I collected Micropus am- phibolus Gray (No. 18002) and Stylocline gnaphaloides Nutt. (No. 18003) in Lake County, California, where the plants grew together in shallow soil on an exposed outcrop of volcanic rock, 5 miles from Lakeport on the road to Hopland. These collections represent northward extensions of the known ranges of both. species, the Micropus from Marin County, the Stylocline from Monterey County.—J. T. Howell. Aw Vot. IV No. 2 LEAFLETS of WESTERN BOTANY CONTENTS PAGE The Botanical Collections of Chamisso and Pechschalrzm Calitornia. 209s) fe ew ALICE EAstwoop Notes on Plants of New Mexico... Sy do A. L. HersHey AND P. J. Levenpecker, in A New Phacelia from Southern Oregon . . . . .. 25 JoHn THomas Howe. mee y aticty, of. echinocereus:) 20s a ee eS BF GraHAM Herp Rumacnrerity: Wisceliaty: 5. 80 ee ye) ee a eT Joun THomas Howe. This number published with funds from the California Botanical Club SAN FrANcIscOo, CALIFORNIA Aprix 28, 1944 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. Bor. NU POU INCHES puuvayacegeenngernggancnpeeegysv0agO0UgQOOQH HOA QUOTA peng PU UTE EA] Owned and published by Auicr Eastwoop and JoHn THomMAsS Howe. APRIL, 1944] CHAMISSO AND ESCHSCHOLTZ 17 THE BOTANICAL COLLECTIONS OF CHAMISSO AND ESCHSCHOLTZ IN CALIFORNIA BY ALICE EASTWOOD Otto von Kotzebue of the Royal Imperial Navy of Russia made two trips dround the world and visited California on both. The first was in the ship Rurik, the main object being the dis- covery of the Northwest Passage. With him as naturalists were Adelbert von Chamisso, a noted poet and author, and Johann Friederich Eschscholtz, entomologist, zodlogist, and ship’s doctor. The Rurik arrived in San Francisco October 1, 1816, and re- mained one month. This was an unfavorable time of the year for collecting plants or insects, but judging by the number of distinct species of plants collected, it must have been a year of early rains. I have collected in San Francisco on Thanksgiving Day nearly a hundred species of plants in bloom on such a year,* when “". . smiling spring its earliest visit paid, And parting summer’s lingering blooms delayed.” The account of the plants and descriptions of the new species were published in Linnza, a botanical journal published by D. F. L. von Schlechtendal and founded in 1826. Chamisso and Schlechtendal were joint authors of most of the new species, but some were described by Chamisso alone. The locality cited was “ad portum San Francisco” and all were collected, according to Chamisso, in the hills and downs about the Presidio. Very few if any are to be found there today. In the following list of the new species the order in which they were published is followed. In Linnea, volumes 1, 2, 3, Chamisso and Schlechtendal were joint authors; in volume 4, Chamisso was the sole author, and in volume 6, the Labiate were described by George Bentham, the authority on the family, and the Composite by C. F. Lessing, who had done some notable work in Vernonia and the author of “Synopsis Generum Com- positarum.” VoLuME 1 CHEIRANTHUS ASPER (Erysimum capitatum (Dougl.) Greene), “in lit- tore,” wall-flower, p. 14. FRANKENIA GRANDIFOLIA, “in sabulosis,”’ yerba rheuma, a salt marsh plant, p. 35. *Cf. Eastwood, ‘Plants in flower in November and December, 1897,” Brythea 6:114 (1898). Leaflets of Western Botany, Vol. IV, pp. 17-32, April 28, 1944. NE BO 18 LEAFLETS OF WESTERN BOTANY [ VOL. IV, NO. 2 VOLUME 2 Rubus vitiro.ius, blackberry (leaves green on lower surface), p. 10. Rusus urRSINuS, blackberry (leaves downy on lower surface), p. 11. FRAGARIA CALIFORNICA, wood-strawberry, p. 20. Fragaria chiloensis also noted. HorRKELIA CALIFORNICA, p. 26. ROSA CALIFORNICA, p. 35. PorTaMmo- GETON AMERICANUS, pondweed, p. 228. ScROPHULARIA CALIFORNICA, bee- plant, p. 585. VOLUME 3 PoLtyGoNUM ParRONYCHIA, sand joint-weed, p. 57. HyprricuM ANA- GALLOIDES, marsh St. Johns-wort, p. 127. OroBANCHE CALIFORNICA, cancer- root, p. 134. VoLuME 4 LESSINGIA GERMANORUM, “in arenosis,” p. 203. This represented a new genus named in honor of G. E. Lessing, a famous dramatist, his great- nephew, C. Lessing, and C. F. Lessing, the author of the Composite collected by Chamisso. Myosotis Cuortstana (Allocarya Chorisiana Greene), “durationeque annua (vel bienni),” p. 444. An annual growing in wet places and very unusual in fruit or flower at this time of year. PHACELIA CALIFORNICA, p. 494. PHACELIA MALVZFOLIA, p. 499. CERATO- PHYLLUM APICULATUM (C. demersum L.), “in stagnis,” p. 506. VOLUME 6 The following Labiate by Bentham: THymus CHaAmissonis (Micro- meria Chamissoms Greene, Satureia Chamissonis Briquet), yerba buena, p. 80. Sracuys ajyucores, p. 80. StacHys CHAmiIssonis, p. 80. The following Composite by Lessing: DreLoparpus ERICOIDES (Aplopappus ericoides H. & A., Ericamerica ericoides Jepson), heather-like composite, pp. 114-117. Aster cALtiFornica (Erigeron glaucus Ker.), seaside daisy, p. 121. BAHIA ARTEMISIZFOLIA (Eriophyllum artemisiefolium Kuntze), lizard leaf, p. 253. TRoxIMON APARGIOIDES (Agoseris apargioides Greene), California dandelion, p. 801. FRANsERIA CHAMISSONIS, p. 507, var. MALVZ- FOLIA and var. BIPINNATISECTA, p. 508. CEPHALOPHORA DECURRENS (Hele- nium puberulum DC.), sneezeweed, p. 577. CoINOGYNE CARNOSA (Jaumea carnosa Gray), salt-marsh composite, p. 520. TANACETUM CAMPHORATUM, tansy, p. 521. ARTEMISIA CALIFORNICA, worm-wood, p. 523. OLIGOCEPHALUS PYCNOCEPHALUS (Artemisia pycnocephala DC.), sand worm-wood, p. 524. Also: Myrica CALIFORNICA Cham., wax-myrtle, p. 575. ESCHSCHOLTZIA* CALIFORNICA Cham., “in arenis sterilibus siccus ad portum San Francisco,” was first described and illustrated in Hor. Phys. Berol. in 1820 and redescribed more fully in Linnza, vol. 1, p. 553. The following species also were collected, not considered as new species; some had been already described by authors in earlier publications, and some given names of species already known and later described as new or mistakes corrected. VoLuME 1 SILENE NICZENsIs Ait. (Silene verecunda Watson), p. 41. PLANTAGO MARITIMA L., p. 167. HyprocoryLe vutearis L. (H. prolifera Kellogg or H. ranunculoides L.), p. 256. RT original spelling of this was Eschscholzia, Hor. Phys. Berol. 73 APRIL, 1944] CHAMISSO AND ESCHSCHOLTZ 19 VOLUME 2 PoTENTILLA ANSERINA L. (P. pacifica Howell), silver-weed, p. 24. VIOLA CANINA L, (V. adunca Smith), blue violet, p. 35. Tr1GLocHIN MARI- TIMA L., arrow-grass, p. 151. PHOTINIA ARBUTIFOLIA, toyon, Christmas berry, p. 541. CaASTILLEJA TOLUCCENSIS HBK. (C. latifolia H. & A.), Indian paint brush, p. 579. Mrimu.Lus GLutinosus Wendl., sticky monkey- flower, p. 586. f VoLuME 3 PoLycGoNUM ACRE L., dotted smart-weed, p. 51. RUMEX SALICIFOLIUS Weim., willow-leaved dock, p. 60. Cornus crrcINNATA L’Herit.? (C. cali- fornica C. A. Meyer), creek dogwood, p. 139. Veronica BEecca-BUNGA L., (V. americana Schwein), speedwell, p. 558. VoLUME 4 NajaAs FLEXILIS (Willd.) Rostk. & Schm., water plant, p. 501. Mvyrto- PHYLLUM SCABRATUM Mx. (M. exalbescens Fern., Rhodora 21 :120), water plant, p. 506. VOLUME 6 SoLipaAco PETIoLaARIS Ait. (S. spathulata DC.), golden-rod, p. 502. Sott- DAGO LANCEOLATA Ait. (.S. occidentalis Nutt.), p. 502. SoLmaco sTRICTA Willd. (S. elongata Nutt.), p. 502. Sormpaco puBERULA Nutt. (S. cali- fornica Nutt.), p. 502. GrINDELIA INULOIDES Willd. (G. maritima (Greene) Steyermark, Ann. Mo. Bot. Gard. 21:576), gum plant, p. 503. Mortna LINEARIS R. & P. (Baccharis Douglasii DC.), p. 505. BAccHARIS GLOMERI- FLORA Mx. (B. pilularis DC.), coyote-brush, p. 506. HErLr1aNnrHus PROS- TRATUS Willd. (Helianthella Cannone Eastw.), p. 511. AcHILLEA MAGNA L. (A. californica Pollard), yarrow, p. 520. ANTENNARIA MARGARITACEA R. Br. (Anaphalis margaritacea B. & H.), pearly everlasting, p. -524. GNAPHALIUM PILULARE Wahl. (G. ustulatum Nutt.; cf. I. M. Johnston, Contr. Gray Herb., n. ser., 70 :87,—1924) , cud-weed, p. 525. GNAPHALIUM DECURRENS Ives (G. californicum DC.), everlasting, p. 525. GNAPHALIUM _ CHILENSE Spreng., p. 525. Quercus aGrIFoL1a Née, coast live-oak, p. 538. CorYLUS AMERICANA (C. rostrata var. californica Ait.), hazel, p. 538. VoL_uME 10 Leguminose, pp. 590-591, by Th. Vogel: TrrroLtuM FIMBRIATUM Lindl. (T. Wormskjoldti var. fimbriatum Jepson, T. Willdenovii var. fimbriatum Ewan). Darea CuiirrortIANA Willd. (Astragalus Gambellianus Sheldon). SYRMATIUM, a new genus, with two new species: S. GLABRUM (Lotus sco- parius (Nutt.) Ottley) and S. romentosum (Lotus eriophorus Greene). VICIA SITCHENSIS Bong. (V. gigantea Hook.). THE COLLECTIONS OF ESCHSCHOLTZ In 1824, Kotzebue made a second voyage in the ship Pred- priatie (1. e., Enterprise) and Eschscholtz came as naturalist. They landed in San Francisco September 27 and departed No- vember 23, a stay of nearly two months. During this time Esch- scholtz visited Mission Santa Clara and Mission San Rafael and from there rode across valleys and hills to Sonoma and Bodega 20 LEAFLETS OF WESTERN BOTANY [ VOL. IV, NO. 2 Bay. He remained at the Russian settlement a week and returned by sea to San Francisco. Off Point Reyes they were driven ashore by a violent storm and finally entered the Golden Gate on October 12. In November another trip was taken on which they ascended the Sacramento River in small boats as far, supposedly, as Rio Vista. Eschscholtz does not give the exact place where his collections were made, but from the plants collected all but - one could have come from San Francisco. The new species were published in “Mémoires de l’Academie Impériale des Sciences de St. Petersbourg,” vol. 10, 1826. An abstract of this publication | was given in Litteratur-Bericht in Linnea, vol. 3, pp. 147-153. In these abstracts the description of the new species by Eschscholtz were republished and have been considered authentic: ABRONIA LATIFOLIA, “in arenos marit.” yellow sand-verbena. Hoirzia sguarrosa (Navarretia squarrosa H. & A.), “in arenos,’ skunk weed. POLEMONIUM CAPITATUM (Gilia Chamissonis Greene), in arenos.” SoLa- NUM UMBELLIFERUM, “fruticetis,’ nightshade. Rises TUBULOSUM “fructi- cetis.” This must have been Ribes malvaceum and not R. glutinosum Benth., the common wild currant of San Francisco, as the leaves are described as white-tomentose on lower surface. Ribes malvaceum has not been found in San Francisco. VELEZIA LATIFOLIA (Frankenia grandifolia C. & S.). ERIOGONUM ARACHNOIDEUM (E. latifolium Sm.). LonicEra LEDEBOURTI, “fruticetis,’ twin-berry. CEANOTHUS THYRSIFLORUS, “fruticetis,” California lilac. This was described as twice the height of a man. RHAMNUS CALI- FORNICA, “fruticetis,” coffee-berry. HENDECANDRA PROCUMBENS (Croton californicus Muell. Arg.), “in arenos.’ Lupinus CHAMISSONIS, “in are- nos,” blue sand lupine. Lupinus serIcEus Pursh (L. arboreus Sims), flowers described as yellow. On the first expedition, during the month of October, sixty- nine species were collected. Among them were two new genera and thirty-three new species. Three were synonyms having been previously described by other authors. On the second expe- dition, Eschscholtz named and described thirteen species, three among them previously described. Linnea has been my source of information concerning these important collections. To Dr. E. O. Essig I am indebted for the historical data which were published in the Quarterly of the Cali- fornia Historical Society, vol. 12, pp. 191-209 (1933) under the title “The Russian Settlement at Ross.” Dr. August C. Mahr’s “The Visit of the ‘Rurik’ to San Francisco in 1816” was pub- lished in the History, Economics, and Political Science series of Stanford University, vol. 2, no. 2. It has been useful in checking . APRIL, 1944] PLANTS OF NEW MEXICO 21 my own lists of the botanical collections and also drew my at- tention to the part devoted to Leguminose in vol. 10 of Linnea (vol. 11 being cited by mistake). The set of Linnzea in the Library of the California Academy of Sciences is incomplete and both volumes 10 and 11 are wanting. I am indebted to the Library of the University of California for loaning them. NOTES ON PLANTS OF NEW MEXICO—III A. L. HERSHEY AND P. J. LEYENDECKER, JR. State College, New Mexico During the past few years the authors have made plant collections in various parts of New Mexico. Among these col- lections, which are deposited in the New Mexico State College Herbarium unless otherwise stated, are some species which repre- sent new additions to the flora of New Mexico or new distri- butional records for the state. _ Calamagrostis Scribneri Beal was found growing in a moist meadow along the banks of a clear mountain stream in the Jemez Mountains, Sandoval County (Herb. Iowa State College). This plant ranges from Washington to Oregon and south to Colo- rado. This record extends its known range some 300 miles south. The clumps were quite robust, and in general the species is well adapted to its habitat in the Jemez Mountains. Several of the larger clumps had been grazed, yet it appeared that the plants were recovering, and there was some evidence of reproduction. Eragrostis hypnoides (Lam.) B. S. P. was collected by J. O. Bridges on sand bars in Lake B. H. Hall at Elephant Butte Dam near Hot Springs, Sierra County. During the summer of 1940 the water in this reservoir was very low, and many sand bars were exposed. On these sandy islands this creeping meadow grass occurred rather abundantly. At the present time the reser- voir is practically filled to its maximum and it is doubtful whether this species will be reported again from this locality for some time. Agrostis Bakeri Rydb. was collected by Q. Hare near Cowles and was identified by J. R. Swallen. This collection (U.S. Nat. Herb.) extends the known range for this grass several hundred miles southward. 22 LEAFLETS OF WESTERN BOTANY [ VOL. IV, NO. 2 During the summer of 1941 the junior author chanced to collect an unusual spiderwort four miles south of Corona, New Mexico. The plant, which is only four to six inches tall, did not compare with any of those listed in Wooton and Standley’s Flora of New Mexico. The plant was identified by Dr. Edgar Ander- son of the Missouri Botanical Garden as Tradescantia Wrightu Rose & Bush. The species appears to be endemic to the Guada- lupe Mountains and the area in Lincoln County where this speci- men was collected. Although this is the second report of this species in New Mexico, the authors consider it of enough interest to be included in these notes. During the past few years several collecting trips were made by the authors along the United States-Mexico boundary in the San Luis Mountains of southwestern New Mexico and among the species collected was Quercus Toumeyi Sarg. (Hershey- Leyendecker No. 2654.) This low oak with small leaves which are rusty pubescent on their under surfaces grows in small clumps on the eastern slopes of the San Luis Mountains. This repre- sents an eastward extension of its range in the United States. The authors would also like to include in this report some notes recently submitted by Dr. E. O. Wooton on two species of Salicornia which were not included in Wooton and Standley’s Flora of New Mexico. “Salicornia utahensis Tidestrom, Proc. Biol. Soc. Wash. 26:13 (1915). Described from material collected by Kearney and Shantz, three miles north of Grantsville, Utah, January, 1913. Specimens (in U. S. Nat. Herb.) from New Mexico are: between Tularosa and White Sands, Oct. 20, 1916, R. L. Piemeisel No. 6483; salt flats, Carlsbad, June 30, 1924, W. T. Lee No. 170; salt lake southeast of Carlsbad, Aug. 12-20, 1924, Standley No. 40343, abundant; flats about Laguna Grande de la Sal, east of Carlsbad, Aug. 1937, W. B. Lang. The plant grows only in alkaline soil and may be expected in such locations any place in the state. It is one of the species called samphire or glasswort in the textbooks. It is a certain indicator of alkali too concentrated for most other species to endure. “Salicornia rubra A. Nelson. A single specimen of this species was collected by Standley at the salt lake southeast of Carlsbad, Aug. 12-20, 1924 (Standley No. 40344), who says of APRIL, 1944] PLANTS OF NEW MEXICO 23 it ‘annual, erect, bushy, glaucous, common.’ This and the pre- ceding species were collected at the same time. They are easily separable, as S. rubra is a rather small, slender-stemmed annual, _ with elongated, slender spikes of many flowers, while S. utahensis is a coarser, larger, perennial plant with short, thickened, clavate spikes, bearing few flowers in what looks like the thickened end of the stem.” Suckleya Suckleyana (Torr.) Rydb. was collected in the summer of 1942 near Clayton in Union County (Hershey No. 2418) and also near Ima in Quay County (H. Bradford & Hershey No. 2655). The latter locality represents a farther extension of the range of this species southward in the United States. Sisymbrium Irio L. is listed in Tidestrom and Kittell’s Flora of Arizona and New Mexico as “rare, waste ground”’ (p. 63). For the past several years the authors have observed the growth of this winter annual in southern New Mexico and have found it very abundant in the Rio Grande Valley. During the winter the plant makes a rapid growth and later produces seed in great abundance. It has become a very common weed in southern New Mexico, west Texas, and parts of Mexico. Swainsona salsula ( Pall.) Taubert has become quite common in Dona Ana County, along the Rio Grande, and in deserted fields where irrigation water is available (Hershey No. 2400). This noxious legume is not widely distributed at present, but because of its vegetative reproduction by creeping root stocks it warrants observation in irrigated areas of the state. Conium maculatum L. was reported from the Mesilla Valley, Dona Ana County, by F. R. Fosberg in 1940.* While engaged in a survey of poisonous plants occurring in New Mexico, the authors have collected this toxic plant in many localities. It is common in the mountains of Otero (Hershey No. 2490), Grant (Hershey No. 2658), Catron (Hershey No. 2458), and Colfax counties. Further investigations will probably show it to be well established in most of the moist, mountainous regions of the state. Euphorbia antisyphilitica Zucc. was collected in the San Andreas Mountains of Socorro County by D. Laughter (No. * Fosberg, F. R., ‘“‘The Aistival Flora of the Mesilla Valley Region, New Mexico,’’ Amer. Midl. Nat. 23:58. 24 LEAFLETS OF WESTERN BOTANY _ [VOL. Iv, NO. 2 2171) in 1940. A number of the plants were reported as grow- ing in gravelly soil on the east slopes of the mountains, a locality where the authors have not been. Probably no other collections have been made of the species in New Mexico. Microrhamnus ericoides Gray has been collected on the foot- hills east of the Guadalupe Mountains in Eddy County (Hershey No. 2633). E. O. Wooton writes in personal correspondence that he was “more or less responsible for omitting this species [1. e., in Wooton and Standley’s Flora of New Mexico], as the species has since been found in the very region traversed by the writer [Wooton] when making sets for distribution in 1897; it was either not seen then or was confused with something else and not collected.” He reports the following collections from New Mexico as deposited in the U. S. National Herbarium: “Arroyo ranch (Hagerman property), June, 1914, Wooton; Carlsbad Cavern, 1924, Vernon.Bailey; dry plains east of Carls- bad, 1924, Standley ; Carlsbad, 1924, W.T. Lee ; east of Carlsbad, 1937, W. B. Lang.’ He also lists collections of the species from the “Santa Catalina Mountains, near Dudleyville, and Santa Rita Range Reserve” in Arizona. Phacelia integrifolia Torr, was collected in the Big Hatchet Mountains in Hidalgo County (Hershey-Leyendecker No. 2659). These plants were growing gregariously on a fresh earthen em- bankment in a small canyon on the east slope. The plants were very robust but not numerous. Verbascum Blattaria L. was collected on one of the trips to the San Luis Mountains, Hidalgo County (Hershey-Leyendecker No. 2660). These plants were very common in a deserted ranch yard on the west slope a few miles east of Cloverdale. Among recent collections have appeared several plants of Composite which the authors would like to record as occurring in New Mexico. These include such common weeds as Trago- pogon pratensis L., Tragopogon dubius Scop. (Hershey No. 2661), Centaurea Picris Pall. (Hershey No. 2368 and 2662), Centaurea melitensis L. (Hershey No. 2663), and Cirsium arvense (L.) Scop. Pinaropappus parvus Blake was found in abundance on the rocky cliffs about permanent pools in the Guadalupe Mountains (Hershey-Leyendecker No. 2641). On the low ridges to the APRIL, 1944] A NEW PHACELIA 25 east of the mountains there was a great abundance of Pinaro- pappus roseus Less. (Hershey-Schaffner No. 2664), while Evax prolifera Nutt. occurred in great abundance on the low mesas which had been flooded during the early spring of 1941 (Hershey- Schaffner No. 2665). Agoseris heterophylla (Nutt.) Greene was recently collected at Mangas Springs, Grant County, where O. B. Metcalfe col- lected many new species. This composite (Hershey-Smith No. 2667) shows great variation in size, from a few inches in height at the edge of low hills to a foot or more in height along the edges of the small bogs still present in this locality. Oxytenia acerosa Nutt. grows quite rank on cretaceous for- mations near Waterflow, San Juan County. This tall, nearly shrub-like composite has come into prominence lately as being poisonous to livestock in New Mexico and Utah. The extent of its distribution within New Mexico has not been definitely determined. Xanthisma texanum DC. covers large areas of the plains of eastern New Mexico (Hershey No. 2666). It is surprising that this species has been omitted from the state floras. Zexmenia podocephala (Gray) Gray was found rather com- monly under the oaks and low-growing shrubs of the San Luis Mountains ( Hershey-Leyendecker No.2668). This plant is a per- ennial with tuberous-thickened roots and showy yellow flowers. Previous reports have given the distribution of this species as southeastern Arizona and northern Mexico. A NEW PHACELIA FROM SOUTHERN OREGON BY JOHN THOMAS HOWELL Phacelia Peckii J. T. Howell. Herba annua gracilis erecta 5—15 cm. alta; caulibus sepe purpurascenti-tinctis, hispido-hirsutulis vel hirsutulis, capitato-glandulosis, seepe elongatis infra folia prima foliacea, ramis paucis, plerumque substricte adscendentibus; foliis sparsis, anguste oblongis vel lanceolatis, 1—2.5 cm. longis, 0.2—0.5 cm. latis, substrigillosis supra, glandu- loso-hirsutulis infra, subacutis, basi attenuatis in petiolum anguste margi- natum sepe 0.5—1.5 cm. longum; racemis substrictis non flexuosis, 2—4 cm. longis, floribus approximatis, pedicellis 1—3 (vel 5) mm. longis; seg- mentis calycis ad anthesin 2 mm. longis, 0.25—0.5 mm. latis, in fructu valde inzqualibus, 3—6 mm. longis, 0.25—1 mm. latis, linearibus vel oblanceo- latis, apice rectis et planis, hirsutulis glandulosisque; corolla violacea, late 26 LEAFLETS OF WESTERN BOTANY [ VOL. IV, NO. 2 © rotato-campanulata, 5—6 mm. longa, tubo circa 2 mm. longo, lobis sub- integris; squamis 1 mm. longis, oblongis, quadratis, marginibus connatis ; staminibus 5 mm. longis, filamentis barbatis processus pilescentes elongatos planiusculos subretrorsos numerosos ferentibus, antheris 0.66 mm. longis; stylo et ramis 4—5 mm. longis, stylo et basibus ramorum piliferis, ramis 3—4 mm. longis; ovario 1 mm. longo, ovulis 8—11; capsula plerumque elliptico-ovata et 4 mm. longa, subacuta infra apiculum, sparse longe et breviter hirsutula, paulum glandulosa, valvis longitudinaliter sulcatis dorso; seminibus subovatis, acute angulatis, 1—1.5 mm. longis, fuligineis, tenuiter et minute reticulato-foveolatis. Type: Herb. Calif. Acad. Sci. No. 301079, moist ground near the top of Grizzly Peak, Jackson County, Oregon, collected by Prof. Morton E. Peck, July 17, 1913, No. 3291. There is a duplicate in the Herbarium of Willamette University. Phacelia Pecku is not a new plant even in botanical literature, for it has been known by numerous collections from Josephine and Jackson counties in southern Oregon under two different names. Usually it has been referred to P. Pringlei Gray, which is a rare and highly localized species confined to the Scott Mts. of northern California, but Macbride (Contrib. Gray Herb. n. ser. 49 :40,—1917) mistook P. Peckii for P. verna Howell, another rare plant which is restricted to ridges of the Umpqua Valley near Roseburg, Oregon. Although P. Pecku is closely related to P. Pringlei, it is quite distinct in a number of small characters and the habit of the two is quite different. In P. Pringlet the lavender corollas are 3—4 mm. long, the filaments bear short retrorse processes, the ovules are 4—7 in each capsule, and the coarsely pitted seeds are 1.5—1.66 mm. long. In P. Peckw the violet corollas are 5—6 mm. long, the processes on the filaments are elongate and hair-like, the ovules number 8—11, and the finely reticulate-pitted seeds are 1—1.5 mm. long. Both species are slender, but in P. Pringlei the subumbellately clustered branches tend to spread while in P. Peckw the branches are usu- ally fewer and substrictly erect. Phacelia Peckii may also be related to P. Greenei J. T. Howell, a little-known plant of Siski- you County, California, in which there are only 4 ovules to each capsule (cf. Amer. Midl. Nat. 30:17). It is an honor to name this Oregon endemic for Prof. M. E. | Peck, enthusiastic student of the Oregon flora and distinguished author of “A Manual of the Higher Plants of Oregon.” APRIL, 1944 | NEW VARIETY OF ECHINOCEREUS 27 A NEW VARIETY OF ECHINOCEREUS BY GRAHAM ,HEID Hollywood, California Echinocereus Salm-Dyckianus Scheer var. noctiflorus Heid, var. nov. Czspitosus; caulibus longis, tenuibus, procumbentibus vel pendentibus; aculeis tenuibus, teretibus, centrali porrecto; floribus elongatis, segmentis perianthii multis, roseis purpureis vel coccineis, semper nocturnis; fructu viridescente, pyriformi, tuberculato. Plant branching from the base; stems up to 1.75 inches in diameter, and to 24 inches long, procumbent or hanging from cliffs. Ribs 7 to 10; areoles small, round; spines yellow or gray-white, acicular, radials about 8, variable in length, central one porrect, up to 0.5 inch long; flowers pro- duced high or low on stem, purplish-pink to scarlet, opening at night; ovary and tube up to 3.5 inches long, axils of scales with weak, white, brown- tipped spines and long cobwebby wool; perianth-segments many, spreading or often recurved; fruit dull green, pear-shaped, tuberculate. Type: Herb. Calif. Acad. Sci. No. 305550, collected by Dr. R. T. Craig near Cerocahui, southwestern Chihuahua, Mexico. This plant is one of a complex involving FE. Salm-Dyckianus Scheer and FE. Scheeri Lem. The habit of night-opening flowers is not complete inasmuch as the stigma is exserted in the daytime, regardless of the weather, in many of the plants. Unlike flower- color, however, this character seems to be a bionomic rather than a mutational one, and therefore seems worth varietal recognition. Dr. Craig reports the plant as common in the mountains of southwestern Chihuahua up to 9000 feet, growing from rocky cliffs and among pines. MARIN COUNTY MISCELLANY—II BY JOHN THOMAS HOWELL In some respects, Point Reyes Peninsula on the ocean side of Marin County, California, is no more a part of the mainland than are the islands off the coast of southern California. From the “mainland” of Marin County the peninsula is cut off by the San Andreas Fault which has formed a rift valley between Tomales Bay and Bolinas Lagoon, and the differential movement of land masses on either side of the fault line is quickly apparent when one perceives that granite is common on the peninsula but is entirely lacking to the eastward. Although the geological 28 LEAFLETS OF WESTERN BOTANY _[VOL. IV, NO. 2 features are obvious enough to those trained to see them, to many the occurrence of certain plants on Point Reyes is an even more obvious indication of the insular character of the region. The presence of a grove of Bishop Pine (Pinus muricata Don) early indicated to students of the California flora that the locality was different from the rest of Marin County and continued field work has disclosed many plants which tell as clearly as the geology that the areas on either side of the fault line have had different histo- ries. From the point of view of plant distribution, the Bishop Pine aligns Point Reyes Peninsula with regions to the north and south where the pine is also found, the regions to the north form- ing a somewhat interrupted coastal strip from the Russian River northward to Trinidad Head; Humboldt County, the regions to the south occurring either as widely separated plant-islands or as actual islands as far south as middle Lower California. Many plants that are found with the pine on Point Reyes bear out the floristic interrelation between the different regions and every botanical visit to that part of Marin County west of the San Andreas Fault is not only an opportunity to wonder at the plants that have been isolated there due to diastrophic disturbances in ages past, but each succeeding visit is also a stimulus to detect new botanical links between Point Reyes Peninsula and other Bishop Pine-lands. On July 4, 1943, Dr. Hans Leschke and I made a collecting trip to Point Reyes Peninsula and among the plants that we col- lected are several that have not before been reported from the region and that emphasize again its peculiar insular character. Because of war-time restrictions on travel by private automobile, we went by bus from San Francisco to Inverness and from there we walked as far along the Point Reyes road as our botanical activities would permit, going beyond Ledum Swamp almost to the head of Drakes Bay. Although we traversed a route that a hundred collectors have taken since the visit of John M. Bigelow in April, 1854, our exceptional success amazes me and I cannot but venture the thought that botanical expeditions afoot are not without compensation. Most of our collecting centered in Ledum Swamp where so many remarkable plants are associated with the Labrador Tea (Ledum glandulosum Nutt.) and where Mr. Stacey and I found so many interesting carices in 1936 (cf. Leafl. APRIL, 1944] MARIN COUNTY MISCELLANY 29 West. Bot. 1:231). At that time three carices new to Marin County were discovered and now at this time three more carices are to be reported as well as more than a half dozen other plants from the swamp or along the road. AGROSTIS LONGILIGULA Hitche. Ledum Swamp in wet soil or on drier ground adjacent, No. 18250. Hitchcock (Man. Grasses p. 341) gives the range for this grass as from Mendocino County, California, north to Tillamook County, Oregon. The plants that grow on Point Reyes are not typical in that the awn does not exceed the lemma or is reduced to a rudiment, but in all other characters they correspond closely to specimens from the north coast of California. In one plant the ligules are nearly 11 mm. long, 5 mm. longer than the maximum given by Hitchcock. CAREX PHYLLOoMANICA W. Boott. In wet part of Ledum Swamp, growing up among bushes of Labrador Tea, No. 18255. The distribution of this species has been definitely known along the Pacific coast from Alaska south to Mendocino County, Cali- fornia, although Mackenzie (Erythea 8:35) indicated that a col- lection made by Bigelow in “California” probably came from Santa Rosa, Sonoma County. With the present report of this species from the Point Reyes Peninsula it would appear likely that Bigelow made his collection not in Sonoma County but on his visit to “Punta de los Reyes” 90 years ago. CAREX MONTEREYENSIS Mkze. Slope under the pines just north of Ledum Swamp, No. 18266. Most of the plants associ- ated with the pines on Point Reyes are species that occur with the pine along the northern California coast in Mendocino and Humboldt counties, but here is an outstanding record of a plant whose distribution is to be correlated with the occurrence of coastal pine-lands to the south. Although Mackenzie gave the distribution of C. montereyensis as “southwestern California, near the coast” in 1931 (N. Amer. FI. 18: 129) and as “near the coast, California’ in 1940 (N. Amer. Caricez, plate 133), I can find in the literature no collections cited by him except from. Pacific Grove, Monterey County (cf. Jepson Fl. Calif. 1:223). Hence it is of interest in citing the new station from Marin County to list the following collections which definitely outline the known distribution of this species. These specimens, which were col- lected by Eastwood and Howell unless otherwise noted, are in 30 LEAFLETS OF WESTERN BOTANY _ [VOL. IV, NO. 2 Herb. Calif. Acad. Sci. and have been determined by the late J. W. Stacey: near Pescadero, San Mateo Co., No. 6056; Afio Nuevo, San Mateo Co., No. 6049; 12 miles north of Salinas, - Monterey Co., No. 2171; Point Lobos, Monterey Co., Mason & Lee No. 9143, Wheeler No. 4285, Eastwood & Howell No. 6029; Cambria, San Luis Obispo Co., No. 5955; Huasna district, San Luis Obispo Co., No. 4143 ; Chalk Hill between Edna and Arroyo Grande, San Luis Obispo Co., No. 4139. CAREX LEPTALEA Wahl. Wet part of Ledum Swamp, No. 18254. This carex, which is so widely distributed in North America, is one of the rarest species in California. In 1922, Mackenzie reported it in California from only a single collection made by J. P. Tracy at Patricks Point, Humboldt County (Ery- thea 8:50). Aside from that record and a second Humboldt County specimen collected by Mr. Tracy at Humboldt Hill (Wo. 5086, Herb. Calif. Acad. Sci.), the plant may be reported from two collections made in Trinity County: Scott Ranch, 3 miles south of Trinity Center, Mrs. H. C. Cantelow in 1936, and Trinity Center, J. T. Howell No. 12703. The discovery of C. leptalea on Point Reyes Peninsula was one of the biggest botanical thrills I have experienced in many a day! Juncus BoLanpeErI Engelm. Wet soil of Ledum Swamp, No. 18259. Although I have known the Bolander Rush from several localities in Marin County for a number of years, I cannot find that it has been reported from there. In 1936, Mr. Stacey and I collected it along the road to Point Reyes (No. 12650) and more recently I have made collections of it on Mt. Tamalpais at Rock Spring and in Potrero Meadows (No. 14853, 17025). The species occurs northward chiefly in the Coast Ranges to middle-western Oregon.* STELLARIA BOREALIS Bigel. var. BoNGARDIANA Fern. Wet soil among shrubs, Ledum Swamp, No. 18261. A northern plant ranging from southern Alaska to Mendocino County, California, and occurring locally in eastern North America, this chickweed may now be reported from Marin County. Farther west towards the Point Reyes Lighthouse, S. littoralis is found and, although * Since writing the ahover I find that Dr. H. L. Mason reported Juncus Bolanderi from Ledum Swamp (which he called Inverness Biss Marsh) in 1934 (Carnegie Institution of Washington Publ. No. 415, p. 119). APRIL, 1944] MARIN COUNTY MISCELLANY 31 some puzzling intermediates between the two species occur, the plant from Ledum Swamp is definitely the variety named. Ceanothus gloriosus J. T. Howell var. porrectus J. T. Howell, var. noy. Frutex porrectus, 0.3—0.5 m. altus, caulibus pauciramosis, ramulis gracilibus patentibus decumbentibus vel adsurgentibus; foliis oblongo- ellipticis vel late obovatis, sepe 1—2 cm. longis, 0.5—1 cm. latis, margine 4-8 dentibus spinescentibus utrinque instructis, rarissime subintegris, stipu- lis 2—4 mm. longis; ceteris speciei. Type: Herb. Calif. Acad. Sci. No. 314109, collected on Inver- ness Ridge in the forest of Pinus muricata on the road to Point - Reyes, Marin County, California, June 4, 1939, J. T. Howell No. 14710; again collected under pines near Ledum Swamp, July 4, 1943, No. 18265, and on March 26, 1944, No. 19348. With its low sprawling habit and smaller leaves, the present plant appears quite different from either typical C. gloriosus which is prostrate or var. exaltatus which is erect, both of which have usually larger leaves. Except in habit, it resembles C. Ma- soni McMinn, but since both morphologic and field evidence indicates to me a hybrid origin for that species, I would regard it wrong to propose as a variety of C. Masonii our present plant, which appears to be a relatively stable woodland type that had nothing to do with the suspected hybrid origin of C. Masonu. Our plant is apparently the same as the “Mount Vision Ceano- thus” which is described from Mount Vision near Inverness by Van Rensselaer (Ceanothus, p. 55) as “showing an apparent relationship to C. Masonii, C. ramulosus, and C. gloriosus.” Ido not believe that it is closely related to C. ramulosus (Greene) McMinn except as that species may have been one of the parents of C. Masonii. Plants of the “Mount Vision Ceanothus” have been cultivated in the Strybing Arboretum of Golden Gate Park for several years and there they have maintained the sprawling habit and distinctive appearance of var. porrectus. HYDROCOTYLE VERTICILLATA Thunb. Wet ground, Ledum Swamp, No. 18253. This lowly pennywort, so widely ranging across the United States, has apparently not before been reported from Marin County. The nearest stations to Point Reyes given for the species or its var. cuneata by Jepson (FI. Calif. 2:615) are the Suisun Marshes and Crystal Springs Lake, both in the San Francisco Bay region. An earlier collection of H. verticillata from Point Reyes is one I made in 1933 (No. 11712) when I 32 LEAFLETS OF WESTERN BOTANY [ VOL. IV, NO. 2 found it on the west side of Drakes Bay growing with Lileopsis occidentalis C. & R. ARCTOSTAPHYLOS MEDIA Greene. Along the road west of Ledum Swamp, No. 18270, Only leafy assurgent stems from procumbent bushes were obtained but their identity is quite cer- tain. The plant cannot be 4A. Uva-ursi which I have collected on the rocky ridge near the Point Reyes Lighthouse because the stems are not prostrate, the apex of the leaves is acute, and their upper surface is slightly stomatiferous. Since the type of A. media which was collected in Washington possibly represents a hybrid between A. Uva-ursi and A. columbiana Piper, it may not be regarded as entirely scientific to apply the name to plants in California, where botanical opinions vary widely as to an acceptable segregation of entities in the columbiana-virgata com- plex and where, as a result, one of the suspected parents of “A. media’ may be any one of several closely related entities. However, the name Arctostaphylos media is very useful and, by restricting its application to the procumbent plants whose as- surgent branches are without bristly pubescence, it will embrace a rather common “entity” whose morphological stability is greater than that of many species recognized in the genus. Until the hybrid origin of A. media is fully demonstrated, the name may properly and usefully serve to designate a distinctive Pacific Coast plant whose range is now known to extend from Vancouver Island southward to Marin County, California. CrYPTANTHA HISPIDISSIMA Greene. West side of pine ridge beyond Inverness, No. 18271. Like Carex montereyensis, this plant is another which links the Point Reyes region with districts to the south. According to Johnston (Contrib. Gray Herb. n. ser. 74:93), the species occurs in the South Coast Ranges from San Francisco to Los Angeles County. More recently Jepson (FI. Calif. 3:348) has included C. hispidissima in C. Clevelandu Greene, but that species is also restricted to the southern moun- tains of the state. It is of special interest to note how markedly separate both habitally and ecologically are C. hispidissima and C. leiocarpa (F. & M.) Greene on Point Reyes Peninsula, although Jepson (ibid., p. 349) remarks that “on morphological grounds C. leiocarpa is weakly and inconstantly distinguishable from C. clevelandii” (i. e., including C. hispidissima). SLEAE UE ES of WESTERN BOTANY Available A limited number of copies of volumes I, II, and III at $5.00 each. Single copies of most numbers in these volumes are available at $.40 each. LINDLEY’s GLOSSOLOGY An excerpt from this most useful work, reissued with an introduction by Alice Eastwood. $.50. ALICE EastTwoop’s Key to the Common Families of Flowering Plants in California and Guide for the Analysis and Description of Flowering Plants. $.25. Address: Joun THomas Howe. California Academy of Sciences, Golden Gate Park San Francisco, California Vot. IV LEAFLETS of WESTERN BOTANY CONTENTS Some New World Cyperi Hucu O’NEILL AND BRoTHER Baas i ay Three New Species From the State of Washington . Rosert F. Hoover A Group of Lupines Related to Lupinus onustus A ice Eastwoop A New Lupine From Mt. Rainier Atice Eastwoop Specific and Varietal Transfers in Cyperacee, Tribe Scirpeze ar te, A. A. BEETLE A Note on Eriogonum C, V. Morton The Type Localities of Certain Californian Blackberries JoHN THomas Howe. SAN FRANCISCO, CALIFORNIA Jury 7, 1944 PAGE 33 38 41 43 47 48 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. URL IRA OR putt INCHES ee eT ee OO Uae blo Owned and published by A.ice Eastwoop and Joun THomas Howe. JULY, 1944] SOME NEW WORLD CYPERI g3 SOME NEW WORLD CYPERI BY HUGH 0)NEILL 4 AND BROTHER BENEDICT AYERS, F.S.C. Langlois Herbarium, Catholic University of America 1. THREE New SPECIES FROM MExIcO Cyperus Arsenii O’Neill & Benedict, spec. nov. (Non Chlorocyperus Arsenti Palla, Osterr. Bot. Zeitschr. 63: 403,—1913.) Perennis. Radices fibrose 0.1—0.3 mm. crasse. Rhizoma gracile, repens, circa 1—3 cm. longum, squamis lanceolatis 5—8 mm. longis indutum. Culmi 8—15 cm. alti, omnino 0.2—0.3 mm. crassi, apice minute scabridi, trigoni, filiformes, sulcati, non septato-nodulosi, basi tuberascentes fibrosique. Foliz duz in culmis singulis, culmo breviores, 5—10 cm. longe, 0.3—0.8 mm. late, at- tenuatz, planz vel conduplicate, rect, membranacez, non septato-nodu- lose, apice antrorse scabride alibi glabre ; vagine purpureo-fusce, in fibris dissolute. Bractez 2, valde inequales, longiores antheiam valde superans, 0.5—5 cm. long, 0.5—1 mm. late, marginibus (costaque interdum) an- trorse minuteque scabride, alibi foliis similes. Spicule 5—7 in capitulum hemisphericum 5—7 mm. diam. contracte, 4—5 mm. longe, 1.5—2 mm. late, oblongze, obtuse, subcompresse, 8—12-flore. Rhachilla 0.3—0.4 mm. lata, recta, straminea, exalata. Glumz inzquales, 1.3—1.6 mm. longe, 0.8—1 mm. late, acute, late ovate, membranacez, vix mucronulate, indistincte trinerves, flavo-fusce, carina viridi, marginibus non hyalinis. Stamen 1, raro 2, achenio non persistentes; filamenta 1.2 mm. longa, 0.1 mm. lata, linearia, rubra; antherz 0.2 mm. long, 1 mm. late, ellipsoidez, flave con- nectivum non productum. Stylus circa 0.5 mm. longus, trifidus. Achenium 1 mm. longum, 0.5—0.6 mm. latum, trigonum, purpureo-fuscum, obovoideum vel ovoideum, minute puncticulatum, vix stipitatum, lateribus planis, non falcatum. The type specimen is Arséne No. 7411, collected near San Miguel del Monte, Michoacan, No. 843010 in Herb. Missouri Bot. Gard. Cotypes are: Pringle No. 7182, Valley of Mexico (G*) ; Gentry No. 6431, Ocurahui, Sinaloa (CU). Cyperus Arsenii belongs to section Dichostylis subgenus Eucyperus. It can be distinguished from the two nearest related species, C. sesleroides and C. Tweedei, by its rough culm, the other species having perfectly smooth culms. The black leaf- sheaths and the white or pale yellow mucronate glumes of C. ses- * The following symbols are used in this paper to indicate institutions from which material is cited: CAS, California Academy of Sciences; CU, Catholic University; F, Field Museum; G, Gray Herbarium; M, Missouri Botanical Garden; NY, New York Botanical Garden; UC, University of Cali- fornia; UM, University of Michigan; US, United States National Herbarium. Leaflets of Western Botany, Vol. IV, pp. 33-48, July 7, 1944. BO 34 LEAFLETS OF WESTERN BOTANY [ VOL. IV, NO. 3 leroides are easily differentiated from the purplish-brown leaf- sheaths and the golden-brown obtuse glumes of C. Arsenit. Cyperus Tweediei, known only from Tucuman, Argentina, has dense heads containing 15—30, 12—24-flowered spikelets with their glumes broadly hyaline-margined. Cyperus Arsen, known only from Mexico, has lax heads containing 5—7, 8—12-flowered spikelets with the glumes not hyaline-margined. Cyperus Pennellii O’Neill & Benedict, spec. nov. Perennis. Radices fibrosee, circa 0.1—0.5 mm. crassz, pubescentes. Rhizoma abbreviatum, lignosum. Culmi 0.5—2 mm. alti, 1—1.2 mm. crassi, basi tuberascentes fibrosique, trigoni, rigidi, multistriati, leves, non septato-nodulosi. Foliz 4 vel 5, interdum a bracteis vix distinct, culmo valde superantes, 2—5 cm. longze, 1—5 mm. late, plane, acuminate, rect, subcoriacez, leves, non septato-nodulose; vagine fusce, in fibris dissolute. Bractee 3—6, an- thelam valde superantes, 2.5—11 cm. longe, 1—5 mm. late, marginibus antrorse scabridz alibi foliis similes. Spicula numerose in caput hemi- sphericum 15 mm. altum 20—30 mm. latum congeste, 7—9 mm. longe, ~ 2—2.5 mm. late vix compresse, divergenti-ascendentes, pleiostiche, 6- vel 7-flore. Rhachilla 0.5 mm. lata, geniculata, viridis, exalata. Glumze 2.8—3 mm. longe, 1.6-2 mm. late demum divergentes, non imbricate, obtuse, non mucronulate, membranaceze, 7—9-nervie, flavo-fuscz, carina viridi leve, marginibus hyalinis. Stamina 3; filamenta 3—4 mm. longa, 0.1 mm. lata; anthere circa 0.6 mm. longz, 0.15 mm. late, flave, lineares, con- nectivum productum. Stylus circa 1 mm. longus. Achenium 2 mm. longum, 0.8—1 mm. latum, ellipsoideum, purpureo-fuscum, minute puncticulatum, vix apiculatum, nec stipitatum nec falcatum, lateribus planis. The type specimen was collected by F. W. Pennell at Remudo, Durango, Mexico, No. 18470. It is in Herb. Acad. Nat. Sci. Philadelphia. A cotype, Pennell No. 18472 (US), was also collected at the same place. Cyperus Pennellii is most closely related to C. manime from which it may be distinguished by the extremely dwarfed culms which are only 0.5—2 cm. tall, whereas C. manime has culms 18—80 cm. tall. Another difference is that the glumes of C. Pen- nellit are 2.8—3 mm. long, while the glumes of C. manime@ are 1.6—2.5 mm. long. Apparently this species is endemic in the State of Durango, Mexico. So far it is known only from the type locality, Remudo, between El Salto and Cueva, on andesitic soil. . In Kiikenthal’s classification, C. Pennelliit falls in the sub- genus Mariscus section Laxiglume. It is readily distinguished from all the other species of this section by its extremely dwarfed habit. JULY, 1944] SOME NEW WORLD CYPERI 35 Cyperus Howellii O’ Neill & Benedict, spec. nov. Perennis. Radices pubescentes, 0.5—1.5 mm. crasse. Rhizoma abbreviatum. Culmi subjuncti, 25—70 cm. alti, apice 1.5—2 mm. basi 2—2.5 mm. crassi, obtuse trigoni, rigidi, papillosi, haud scabridi, multistriati, non septato-nodulosi. Foliz 8—10 in culmis singulis, plerumque culmo breviores, 10—50 cm. longe, 3—5 cm. late, attenuate, plane vel conduplicate, coriacez, arcuate, inter- dum obsolete septato-nodulose, marginibus antrorse spinulose, glauce; vagine purpureo-brunnez in fibris dissolute. Bractez 3—5, inequales, an- thelam valde superantes, 3—40 cm. longe, 1—3 mm. late, marginibus spinu- lose plerumque papillosz, alibi foliis similes. Spice 8—15 mm. longe, 8—10 mm. late, plures in caput 30—40 mm. diam. contractz vel explicate radiis 5—10, 1—4 cm. longis. Spicule 9—11 mm. longe, 2—4 mm. latz, oblongo-lanceolate, acute, turgide divaricate, perdense pleiostiche, 10— 16-florz. Rhachilla 0.5 mm. lata, recta, decidua vel diu persistens, exalata. Glumz cito decidue, 2—3 mm. longe, 2 mm. late, late ovate, acute, chartaceze, non mucronulatz, obscure 11—13-nervose, rubro-fusce, carina fusca, marginibus stramincis non hyalinis. Stamina 3; filamenta 2—2.5 mm. longa; anthere 1.2 mm. longe, 0.2 mm. late, lineares, rubre, connecti- vum rubrum productum. - Stylus circa 1 mm. longus. Achenium 1.2—1.5 mm. longum, 0.7—1 mm. latum, obovoideum vel ovoideum, atrosanguineum, minute puncticulatum, lucidum, apiculatum, substipitatum, non falcatum, lateribus planis. As type we designate J. T. Howell No. 10560 (Herb. Calif. Acad. Sci. No. 216724) collected at Mazatlan, Sinaloa, Mexico; an isotype is in the Catholic University Herbarium. As cotypes we cite: J. T. Howell No. 8415 (CAS), Socorro Island, Revilla- gigedo Islands; J. T. Howell No. 10528 (CAS), Isabel Island, Tres Marias Islands; J. T. Howell No. 10477 (CAS), San Juanito, Tres Marias Islands. Cyperus Howellit is apparently endemic to the Revillagigedo Islands, Tres Marias Islands, and the adjacent coast of Mexico. I. M. Johnston (Proc. Calif. Acad. Sci., ser. 4, 20:55,—1931) mentions this plant as “aff. brunneus Sw.” From this it is clearly distinguished by its wingless rhachilla and its red, not black, achene. It has also been suggested that this plant might be C. glaucus Steud. According to Ktikenthal (Fedde Repert. 23: 189,—1926), C. glaucus is synonymous with C. brunneus Sw. Cyperus Howellii, according to Kiikenthal’s classification is to be placed in the subgenus Mariscus section Turgidult. It is most closely related to C. ligularis L. and C. javantcus Houtt. (C. pennatus Lam.). In common with these two species, the culm is papillose ; in distinction to these two species, the rhachilla_ is wingless. 36 LEAFLETS OF WESTERN BOTANY _[VOL. IV, NO. 3 2. THe Status oF Cyperus BotTtTert, Cyperus Mourtt, AND CHLOROCYPERUS ARSENII Cyperus Bottert Boeck. (Allg. Bot. Zeitschr. 2:2,—1896) is based upon Botteri’s No. 33, Orizaba, Mexico. One sheet of this number is in the New York Botanical Garden. Britton recognized the plant as a distinct species which he named C. Mohri, evidently being unaware that Boeckeler had already used the same specimen as the type of C. Botteri, possibly be- cause the label on this specimen was misread as Baker No. 33 (Britton ex Clarke, Kew Bull. Add. Ser. 8:10,—1908). Un- fortunately this misreading has been repeated by Kiukenthal (Pflanzenr. IV. 20 (Heit 101): 461,—1936), who treats C. Mohrit as a valid species with the comment “nicht gesehen.” Kukenthal (loc. cit.), however, places C. Botteri in the synonymy of C. ischnos Schlectend. and cites Botteri’s specimen, No. 33, from Orizaba. This plant is certainly not the same as C. tschnos. While it has glumes with tridentate apex, it differs from C. tschnos in having a conspicuously broadly winged rhachis and a scabrid culm, while C. ischnos has a wingless rhachis and a smooth culm. Chlorocyperus Arsenii Palla (Osterr. Bot. Zeitschr. 63 :-403,— 1913) is based upon Arséne No. 5318 and 5580, both from Morelia, Michoacan. One sheet of each of these numbers is in the New York Botanical Garden and in the Missouri Botanical Garden. The isotype, No. 5318, has been photographed by the Catholic University. This plant as treated by Kiikenthal is C. ischnos var. Arsenti (Palla) Kiikenth. All these specimens of Arséne, however, have a well-developed broad wing on the rhachis and scabrid culms and in other respects are indistinguish- able from C. Bottert. Accordingly C. Botterit seems best treated as follows: Cyperus Botterr Boeck., Allg. Bot. Zeitschr. 2:2 (1896). C. Mohri Britton ex Clarke, Kew Bull. Add. Ser. 8:10 (1908). C.ischnos Schlechtend. var. Arsenii (Palla) Kiikenth., Pflanzenr. IV. 20 (Heft 101) : 462 (1936). Chlorocyperus Arsenti Palla, Osterr. Bot. Zeitschr. 63:403 (1913). 7 3. Two NEw NAMES IN CYPERUS Cyperus manime H. B. K. var. phzeocephalus (Griseb.) O’Neill & Benedict, stat. nov. C. pheocephalus Griseb., Abh. —_ oO". - JULY, 1944] SOME NEW WORLD CYPERI 37 Goett. Ges. Wiss. 19:216 (1874). Mariscus pheocephalus (Griseb.) Clarke, Kew Bull. Add. Ser. 8:16 (1908). Grisebach assigned the name Cyperus pheocephalus to a plant with dark purple glumes. Clarke, although he placed this plant in the genus Mariscus, was of the opinion that the plant could not be maintained separately from C. manime (notes ap- pended to several of his determinations, e. g., Forrer No. 9, Mazatlan, Durango, Mexico, in the New York Botanical Garden). A thorough study of specimens of this plant from Mexico and South America by the authors has led to a similar conclusion. The only character which separates this plant from C. manime is its dark purple glumes. (Cyperus manime@ has light brown glumes.) Coloration of glumes is usually of little significance. Forms intermediate in color between the dark purple of C. pheo- cephalus and the red to brown glumes of C. manime can be found, e. g., Pringle No. 8921, F. Miiller No. 1988, Pringle No. 3160; Fisher No. 129, all of which were collected in Mexico. On this account it seems best to treat Grisebach’s species as a variety of C. manime. The range of this Cordilleran plant appears to be Mexico to Argentina. Hitherto it has been reported only from South America, where it is apparently common in Peru and Bolivia. The present publication extends its range to Mexico. ARGENTINA: Sierra de Cordoba, C. Galander (NY); Prov. de Jujuy, Venturi No. 4889 (CU) ; Prov. de Tucuman, L. R. Parodi No. 10686 (CU). Borivia: O. Kuntze No. 7 (NY); near La Paz, Rusby No. 100(NY); Sonata, Rusby No. 158 (NY), No. 174 (UM), G. Mandon No. 1394 and No. 1434 (NY); La Paz Bang No. 95, No. 95a (NY); Depto. Cocha- bamba, Eyerdam No. 24655, No. 25024 (UC). CotompBia: Mutis No. 2903 (NY). Ecuapor: A. Rimbach No. 200 (NY); Prov. Tunguraqua, Hitchcock No. 21899 (NY); Spruce No. 5904 (NY). ; Peru: C. Vargas No. 685 (F) ; Depto. Cuzco, Vargas No. 9737 (UC). Mexico: Pochuca, Hidalgo, Rose et al. No. 8883 (US); Somoriel, Hidalgo, Rose et al. No. 9214 (NY, US) ; Trinidad, Hidalgo, Pringle No. 8291 (US); El Salto, Durango, Pennell No. 18370 (CU); Amecameca, Federal District, Fisher No. 134 (M, US); Ajusco, Federal District, Orcutt No. 3710 (M, US), No. 3711 (US) ; Mt. San Rafael, Federal Dis- trict, Lyonnet No. 721 (US); Lomagrande, Vera Cruz, Balls No. 5384 (US); Mt. Orizaba, Vera Cruz, Seaton No. 283 (G, US); La Parada, Oaxaca, Nelson No. 1005 (US) ; Esperanza, Puebla, O. Purpus No. 2510 (US). 38 LEAFLETS OF WESTERN BOTANY _ [VOL. IV, NO. 4 Cyperus nudiceps (Clarke ex Standl.) O’Neill., comb. nov. Kyllinga nudiceps Clarke ex Standl., Field Mus. Bot. Ser..4: 199 (1929). This species belongs to the subgenus Kyllinga section Eukyl- linga. It is most nearly related to the pantropical C. peruvianus (Lam.) N. F. Williams as shown by the reduction of leaves and bracts, the quadrangular culms, the puberulent roots, the smooth- keeled, wingless glumes, etc. Apparently it is confined to Cocos Island. The following key shows some of the distinguishing characters : Culms 0.5 mm. thick at apex and at base; bracts perfectly smooth, 2—4 mm. long; head 5—6 mm. in diameter containing about 30 ovate to ovate- lanceolate spikelets, spikelets 2.2—2.5 mm. long; achene 1 mm. long, 5.1 ties AY Pian een PE Ak Mune IN Kena ARS ea ie, oneal Se C. nudiceps Culms 1—2 mm. thick at the apex, 3 mm. thick at the base; bracts an- trorsely scabrid on the margins, 4—30 mm. long; head 6—10 mm. in diameter, containing 100—200 oblong spikelets, spikelets 3.5 mm. long; achene 1.3 mm. long, oblong-obovate......................--...------ C. peruvianus THREE NEW SPECIES FROM THE STATE OF WASHINGTON BY ROBERT F. HOOVER ‘ Lithophragma Thompsonii Hoover, spec. nov. Pubescens ; rhizomati- bus plerumque purpurascentibus ; caulibus haud bulbiferis, 15—35 cm. altis; foliis radicalibus non nullis, tenuibus, vel trilobatis sinibus extendentibus circa ad medium vel interdum breviter quinquelobatis, lobis latis integris vel breviter 2—4-lobulatis; foliis caulinis sepissime 2, late separatis, partitis ad basin vel fere ad basin in 3 segmenta linearia vel oblanceolata integra paucidentata vel trilobata lobis linearibus divergentibusque; pedicellis flori- feris divergentibus, 1—2 mm. longis, fructiferis suberectis, 4—7 mm. longis ; calyce cyathiformi basi acuto, ad anthesin 3.5—4 mm. longo, in fructu 4—6 mm. longo, lobis late triangulis, 1 mm. longis vel brevioribus; petalis albis, 5—6 mm. longis, laciniate 7—9-lobatis, lobis basalibus brevibus angustissi- misque ; carpellis junctis ad calycem tantum basi, in fructu calyce paululum longioribus. Okanogan County: Omak, Fiker No. 8. Grant County: near Coulee City, Thompson No. 6162. Yakima County: East Selah Hills, Hoover No. 5661; south rim of Cowiche Canyon, Hoover - No. 5694; south fork of Wide Hollow Creek, Hoover No. 5745 (type). The existence in Washington of any species of Lithophragma having simple basal leaves has not previously been recorded. The JULY, 1944] THREE NEW SPECIES 39 two other species known to occur in the state, L. parviflora (Hook.) Nutt. and L. bulbifera Rydb., both have most of the basal leaves parted into three distinct leaflets. It is a pleasure to name this Lithophragma for Mr. J. W. Thompson of Seattle, an authority on the flora of Washington and one of the few persons who have collected the species. When I was in the army and stationed in Seattle, Mr. Thompson gen- erously invited me to make full use of his privately owned herba- rium located in his home. Most of the work involved in describ- ing these three new species was done in Mr. Thompson’s herba- rium. Lomatium tuberosum Hoover, spec. nov. Radix oblonga vel raro sub- globosa tuberformi, 5—10 cm. longa, 2—4 cm. diametro, axi longo vel hori-’ zontali vel verticali; petiolis foliorum radicalium 7—12 cm. longis, laminis glabris glaucis cuneatis, 4—8 cm. longis, ternatis denique 3—4—plo pinnatis, segmentis ultimis linearibus, 3—6 mm. longis; caulibus gracilibus, 12—23 cm. altis, subnudis, plerumque ferentibus prope basin bracteam unam scario- sam interdum terminatam lamina reducta; umbella ex 5—8 radiis divergen- tibus composita, radiis in fructu 2—8 cm. longis, 8—12-floris, tantum paucis floribus fructiferis; bracteolis paucis inconspicuisque, anguste linearibus ; pedicellis ad anthesin brevissimis, in fructu divaricatis, 7—14 mm. longis; petalis purpurascentibus; antheris flavis; fructu sine alis 4—5 mm. lato, 9—11 mm. longo, alis aliquantum crassis, latitudine minore quam 1 mm. Yakima County: hill above Selah Creek, Hoover No. 5603; hills south of White Swan, Hoover No. 5726 (type). The plants were found only on gentle slopes among loose rocks. At both localities the number of plants in flower in 1942 was very small. This is one of the species which formerly would have been re- ferred to Leptotenia, which is not tenable as a genus, as Mathias and Constance have shown (Bull. Torr. Club 69: 244,—1942). Specimens of the two collections cited were identified by Mathias and Constance, but with an expression of doubt, as Lomatium cuspidatum M. & C. (Leptotenia Watsonii C. & R.), an endemic of the Wenatchee Mountains. Examination of the specimens of L. cuspidatum reveals several differences, the most conspicuous of which is in the root. These differences are here summarized. Plant with a long stout vertical cylindrical tap-root ; old leaf-bases form- ing a tuft of coarse fibers on root-crown; stems usually comparatively stout, usually bearing 1 to 3 well-developed leaves above the base; petioles about equaling or mostly shorter than blades; ultimate leaf- divisions lanceolate to ovate, 2 to 4 mm. long; petals and anthers (except for pollen) bright purple... tcecseeeeseeeceeeeeeeeeees L. cuspidatum 40 LEAFLETS OF WESTERN BOTANY _ |VOL. IV, NO. 3. Plant with an oblong tuberous root ; old leaf-bases membranous, not per- sistent ; stems always slender, bearing at most only one leaf-sheath (sometimes with small blade) above the base; petioles longer than blades; ultimate leaf-divisions linear, 3 to 6 mm. long; petals dull purple ; anthers yellow even without pollen.................2.....---.--- L. tuberosum In my opinion, L. tuberosum is most closely related to L. minus (Rose) M. & C., a very rare and poorly known species of eastern Oregon which is likewise tuberous-rooted. The one available specimen of L. minus was supplemented by an authentic illustra- tion published by Coulter and Rose (Contrib. U.S. Nat. Herb. 7: pl. 8) and the accompanying description. The following differ- ences are evident. Stem, rays, and pedicels stout; leaves primarily pinnate; ultimate leaf- divisions 1.5 to 3 mm. long; fruit 13 to 16 mm. long, the wings fully ASG shor eke (0 | aaa eae Sn ea ME et Me ane. Kee SET elma ee Eo L. minus Stem, rays, and pedicels slender ; leaves primarily ternate ; ultimate leaf- divisions 3 to 6 mm. long; fruit 9 to 11 mm. long, the wings less than pruners Wales CoM 5 Sh eee a ae eee L. tuberosum Erigeron basalticus Hoover, spec. nov. Perennis ex radice gracili; cau- libus multis patentibus vel pendentibus super rupes, ramosis, 7—25 cm. longis, dense hirsutis, fragilibus, radicantibus prope basin; foliis plurimis, saturo-viridibus, 15—40 mm. longis, laminis sensim angustatis in petiolem, oblanceolatis vel late obovatis, trilobatis, lobis integris vel sepe 1- vel 2-dentatis, foliis inferioribus mox marcescentibus, apice tridentatis vel ali- quot subintegris, foliis supremis sepe multo reductis et bracteatis, integris vel tantummodo dentatis ; capitulis ramulos breves terminantibus in apicibus caulium; involucro ad anthesin 10—12 mm. diametro, bracteis plerumque subzequalibus, circa 30—35, herbaceis, margine scariosis, pubescentibus, 5—6 mm. longis; radiis 25—30, lilacinis vel albis, 6—7.5 mm. longis; acheniis 1.5 mm. longis; setis pappi circa 10, albescentibus, fragilibus, 2.5—3 mm. longis. Yakima County : west base of Selah Butte, Hoover No. 6021; Selah Creek, Hoover No. 5983 (type). This species grows on dry vertical basalt cliffs and apparently is restricted to the canyon of the Yakima River, together with its small tributary canyons, be- tween Yakima and Ellensburg. It is in flower through the summer from May to October. So far as I have been able to ascertain, only three other known species of Erigeron, at least on this continent, have lobed leaves. Of these, E. compositus Pursh is distinctive in having the leaves all basal. Erigeron allocotus Blake, which is known only from the type locality in Wyoming, has lobed cauline leaves similar to those of E. basalticus and might be considered its closest relative. The JULY, 1944} A GROUP OF LUPINES AI two are very different in habit, however, F. allocotus having erect stems with few leaves on the upper part. As to other differences, the stems of E. allocotus are less hairy, its leaves or their lobes are mostly narrower, and its flower-heads are noticeably smaller than those of E. basalticus. Erigeron vagus Payson, a Colorado species, apparently resembles E. basalticus more closely in habit but is a much smaller plant with small leaves crowded in tufts at the ends of branching rootstocks and with flower-heads on comparatively long leafless peduncles. AGROUP OF LUPINES RELATED ‘TO LUPINUS ONUSTUS BY ALICE EASTWOOD In this group three Ses have been described: Lupinus onustus Watson (Proc. Amer. Acad. 11: 127,—1876), type col- lected in California by Mrs. M. E. Pulsifer Ames at Indian Valley, Plumas County, also by J. G. Lemmon, Sierra County ; L. mucro- nulatus Howell (Erythea 1: 109,—1893), type collected by Thomas Howell in serpentine near Waldo, Josephine County, Oregon; and L. violaceus Heller (Muhl. 2: 65,—1905), type col- lected by Heller, No. 8037, under pine trees “on the first ridge west of Sisson,’ Siskiyou County, California .. . growing in thick mats, the stems not rising much above the ground.” All of these plants are alike in the decumbent habit, similar leaves with long- spreading petioles and broadly oblanceolate to narrowly spatu- late open spreading leaflets, and with many leaves from the base of the plant; the flowers are on the violet or purple shades and roundish in outline; the banner is broad with the middle groove deep, apex not notched and on the back finely pubescent near the base, also with some hairs on the middle or near the upper edges ; the wings are broad, covering the keel ; the keel is curved, broad at the middle, and more or less ciliate; the ripe legume is broad and thick with a few large seeds; the upper calyx-lip is bidentate and the lower is longer and entire. Lupinus onustus Watson. Several specimens from Plumas County, California, agree so well with the original description that to see the type was not necessary for identification: Mrs. R. M. 1 Now Mt. Shasta City. 42 LEAFLETS OF WESTERN BOTANY _ [VOL. IV, NO. 3 Austin No. 41; Mrs. Mary Strong Clemens, Greenville; Mrs. H. C. Cantelow, 2 miles north of Quincy, Highway 89. These are all in flower and similar. At Forest Lodge, near Greenville, the author collected good specimens in flower and fruit, No. 14461. LuPINUS MUCRONULATUS Howell. Through the kindness of Dr. H. L. Mason a specimen of L. mucronulatus from the Bran- degee Herbarium of the University of California was loaned to the author. This was collected at Waldo by Thomas Howell and is probably part of the type collection. Two specimens are on the sheet, one with leaflets narrowly spatulate, obtuse, and mucronu- late at apex ; the other with leaflets oblanceolate, pointed at apex and base. The same leaflet variation is seen in other members of the group. Specimens from the mountains of Trinity County, California, appear to represent L. mucronulatus: Eastwood & Howell No. 4919, on the road from Minersville to Trinity Center, a complete plant with flowers and immature fruit; Mrs. E. C. Van Dyke, specimen from Trinity Center, without flowers or root but with fruit almost ripe, 1 cm. broad, 3 cm. long; Lester Rowntree, Big Flat Road, several flowering branches only. All of these have the spreading decumbent habit, long petioles, narrowly spatulate leaf- lets verging to oblanceolate, upper and lower surface with scat- tered appressed hairs, the flowers similar in shape and size to those of L. mucronulatus, the keel slightly ciliate. LuPpINus vIoLAcEus Heller. Besides the type collection of L. violaceus (Heller No. 8037), several specimens from Shasta Springs and Mt. Shasta City represent the species in the Herba- rium of the California Academy of Sciences. The best specimen is Eastwood No. 11864 collected under pines at Shasta Springs. It is a complete plant with roots, flowers, and fruit. The heavy fruit is a marked feature of the group. The smaller flowers distin- guish this from the two preceding. Lupinus violaceus Heller var. shastensis Eastwood, var. nov.” Specimens collected near Baird, Shasta County, California, Eastwood & Howell No. 1819, have the small flowers of L. viola- 2 Lupinus violaceus Heller var. shastensis Eastwood, var. nov. A specie differt: foliolis 2/3 minoribus, infra densissime argenteo-sericeis, pedunculis et racemis longioribus. ‘Type: Herb, Calif. Acad. Sci. No. 216343, collected at Pollock’s Store near te Shasta County, California, April 21, 1934, Hastwood & Howell No. JULY, 1944] A NEW LUPINE FROM MT. RAINIER 43 ceus but the leaflets are not half the size, the upper surface is glabrous but the lower more densely pubescent with long silvery appressed hairs. Two specimens are on the sheet, one with flowers a trifle larger than the other ; the peduncles and racemes are much longer than in any of the specimens of L. violaceus and the flowers are whorled. Lupinus violaceus Heller var. delnortensis Eastwood, var. nov.? From Del Norte County, California, we have two collections similar to L. violaceus in habit, leaves, and small flowers: French Hill, near Adams Station, Eastwood No. 69, collected August, 1907 ; on the grade from Patrick Creek to Shelley Creek, East- wood & Howell No. 3666. This latter is a complete specimen in flower and immature fruit which, however, is advanced sufficiently to show its character. In these specimens the upper calyx-lip is more deeply bidentate and the lower lip longer, ovate-lanceolate in shape ; the keel appears to be glabrous ; the upper surface of the leaflets is glabrous, the lower lightly appressed-pubescent; the shape of the flower is that of the group. KEy TO THE SPECIES AND VARIETIES Flowers about 10 mm. long. Keel densely ciliate, upper leaf-surface glabrous, lower appressed- CAL" SCPSCT 591 ail NAG RS ne ty Py ERR PU CUO ACO L. onustus Keel lightly ciliate, upper and lower leaf-surface lightly pubescent.... lat UR) RSE OU NR OR PEA ASR a A L. mucronulatus Flowers about 8 mm. long. Keel ciliate. Longest leaflets 4 em. long,-1 cm. broad. ..o..: oi ele L. violaceus Longest leaflets 10 mm. long, 4 mm. broad..L. violaceus var. shastensis BtcelFolaDrousi set tke es CEM a enh ih oot L. violaceus var. delnortensis A NEW LUPINE FROM MT. RAINIER BY ALICE EASTWOOD Lupinus rainierensis Eee dtl anne. nov. Ramosus basi et supra ex radice simplici et lignea; caulibus gracilibus, circa 2—3 dm. altis, glabris vel parce pubescentibus, ramis patentibus; petiolis inferioribus foliolis longioribus, petiolis superioribus foliolis subzequalibus, foliolis 6 vel 7, circa 4 cm. longis, 5—10 mm. latis, oblanceolatis, obtusis cum acumine, supra 3 Lupinus violaceus Heller var. delnortensis Eastwood, var. nov. A specie differt: carina glabra; calycis labio superiore bifido profundius, labio in- feriore longiore ovato-lanceolato. Type: Herb. Calif. Acad. Sci. No. 235467, collected on the old road on the grade between Patrick Creek and Shelley Creek, Del Norte County, Califor- nia, July 2, 1936, Hastwood & Howell No. 3666. 44 LEAFLETS OF WESTERN BOTANY _ [VOL. IV, NO. 3 fere glabris, infra leviter adpresse pubescentibus; floribus circa 10 mm. longis, violaceis, verticillatis in racemis ramulos terminantibus; pedun- culis plerumque folia superantibus; pedicellis gracilibus, circa zquilongis calycibus, adpresse pubescentibus; tubo calycis brevissimo, basi gibboso, labio superiore circa 2 mm. longo, basi 2 mm. lato, apice abrupte bidentato, tecto vexillo, labio inferiore 7 mm. longo, lanceolato-attenuato; vexillo glabro, ex basi reflexo, violaceo, croceo-maculato, 7 mm. longo, 8 mm. lato, ungue lato, basi gibboso, croceo, margine crispo; alis 8 mm. longis, 6 mm. latis, ungue 1 mm. longo; carina exserta, inferiore parte flavescenti, su- periore parte violacea, margine leviter ciliata prope medium. Type: Herb. Calif. Acad. Sci. No. 235204, collected June 20, 1936, on White River at boundary of Mt. Rainier National Park, King County, Washington, by Eastwood and Howell, No. 2922. The following specimens in Herb. Calif. Acad. Sci., though taller and with flowers a little larger, seem to belong to this species: Eastwood No. 9655, Mountaineer’s Camp, Kitsap County, Washington; L. S. Rose in 1916, South Fork Sno- qualmie River, elev. 1500 ft., King County, Washington; H. M. Hall No. 10945, north base of Mt. Hood, Oregon. This species was collected in August, 1943, on Mt. Rainier by Dr. Hans Leschke, who collected also and readily identified L. subalpinus Piper & Robinson and L. volcanicus Greene, but this was neither and evidently new to Mt. Rainier. In studying Dr. Leschke’s plant, I came across the specimen that I have taken as the type and the others listed above. Lupinus rainierensis may be dis- tinguished from related species by the following tentative key: leeStems widely; spreading above....:)....02.0. ee. 2 Hee Stemsvsiinpley sek moti is ee ye es a ee ee 3 2: weaves pale om lower surfaces. .-.40 8.8 ne L. columbianus Heller AwJteaves oreen on both ’suriaces. x! a 8 ee L. rainierensis Eastw. See Pubescence shag eyo. oer nes eee ek, L. volcanicus Greene See UDeESCENCe: cP PLESSeG 4.2 2.2 eee L. subalpinus Piper & Robinson SPECIFIC AND VARIETAL TRANSFERS IN CYPERACEA, TRIBE SCIRPE BY A. A. BEETLE Division of Agronomy, University of California, Davis The description of the unnatural genus Isolepis (Cyperacee ) by R. Brown (Prod. Fl. Nov. Holl..—1810) has considerably delayed an appreciation of generic limits within the tribe Scirpee and has been a major factor in the nomenclatorial unrest within JULY, 1944] SPECIFIC AND VARIETAL TRANSFERS 45 the group. An appreciation of the names described under Isolepis has proved to be a prerequisite to a monograph of many sections in the genus Scirpus, a study which has been under way for some time and which has resulted in the accumulation of some routine transfers. The following are presented at this time: Bulbostylis conifera (Kunth) Beetle, comb. nov. Isolepis conifera Kunth Enum. Pl. 2:206 (1837). Isolepis ciliifolia Steud. Syn. Pl. Glum. 2:99 (1855). Scirpus Reichenbachii Boeck., Linnea 36:737 (1870). Fimbristylis conifera Reich. in Boeck., 1. c. Bulbostylis conostachya (Boeck.) Beetle, comb. nov. Isolepis conostachya Boeck., Flora 61:597 (1858). Scirpus conostachys Boeck., Linnea 36:756 (1870). Bulbostylis densa (Wall.) Beetle, comb. nov. Scirpus densus Wall. in Roxb. Fl. Ind. 1:231 (1820). Isolepis densa R. & S. Mant. Syst. Veg. 2:71 (1824) ; Wight, Contrib. Bot. Ind. 109 (1834). Isolepis tenuissima D. Don, Prod. Fl. Nepal. 40 (1825). Bulbostylis Gardneriana (Steud.) Beetle, comb. nov. Isolepis Gardneriana Steud. Syn. Pl. Glum. 2:99 (1855). Bulbostylis Labillardieri (Steud.) Beetle, comb. nov. Isolepis Labillardieri Steud. Syn. Pl. Glum. 2:96 (1855). Bulbostylis pilosa (Steud.) Beetle, comb. nov. Isolepis pilosa Steud. Syn. Pl. Glum. 2:97 (1855). Bulbostylis Pringlei (Britton) Beetle, comb. nov. Scirpus Pringlei Britton, Bull. Torr. Bot. Club 15:103, pl. 80 (1888). Bulbostylis Schraderiana (Steud.) Beetle, comb. nov. Isolepis Schraderiana Steud. Syn. Pl. Glum. 2:103 (1855). Bulbostylis subdisticha (Boeck.) Beetle, comb. nov. Scirpus subdistichus Boeck., Linnza 36:490 (1870). Bulbostylis trichoides (Schrad.) Beetle, comb. nov. Isolepis trichoides Schrad. in R. & S. Mant. Syst. Veg. 2:64 (1824). Bulbostylis trichokolea (Steud.) Beetle, comb. nov. Isolepis trichokolea Steud. Syn. Pl. Glum. 2:96 (1855). Bulbostylis Wallichiana (Schult.) Beetle, comb. nov. Isolepis Wallichiana Schult. in R. & S. Mant. Syst. Veg. 2:533 (1824). Isolepis capillaris D. Don, Prod. Fl. Nepal. 39 (1825). Fimbristylis armerioides (Miq.) Beetle, comb. nov. Isolepis armerioides Mig. FI. Ind. Bot. 3:310 (1859). Scirpus africanus Beetle, nom. nov. Scirpus tenuis Spreng. in Flora Beil. p. 12 (1829). Not Willd. Enum. Pl. 1:76 (1809). Isolepis tenuis Schrad. Anal. Fl. Cap. 15 (1832). Not Presl, Delic. Prag. 144 (1822). Isolepis atropurpurea Nees, Linnea 7 :495 (1832). Not R. & S. Syst. Veg. 2:106 (1817). Nor Scirpus atropurpureus Retz. Obs. 5:14 (1789). 46 LEAFLETS OF WESTERN BOTANY _ [VOL. IV, NO. 3 Isolepis pusilla Kunth Enum. Pl. 2:190 (1837). Not Scirpus pusillus Vahl Enum. Pl. 2:246 (1806). Scirpus amazonicus Beetle, nom. nov. Scirpus fluitans L. var. emergens Pfeiff., Rep. Spec. Nov. 33 :202-204 (1933). Not Scirpus emergens (Norman) Fernald, Rhodora 45: 281 (1943). This plant and Scirpus confervoides Poir. are the only American representatives of the Section Eleogiton (Link) Pax. This section, principally African, does not seem to have received taxonomic study in more than a century since Nees (1834 to 1836 in Linnea) treated the section as the pos Eleogiton. Two of his names have been overlooked : Scirpus rubicundus (Nees) Beetle, comb. nov. Eleogiton rubicunda Nees, Linnea 10:164 (1836). Isolepis rubicunda Kunth Enum. PI. 2:188 (1837). Scirpus striatus (Nees) Beetle, comb. nov. Eleogiton striata Nees, Linnea 9:29 (1834). Isolepis striata Kunth Enum. Pl. 2:189 (1837). Scirpus mindorensis Beetle, nom. nov. Isolepis Cumingii Steud. Syn. Pl. Glum. 2:101 (1855). Not Scirpus Cumingti Boeck., Linnzea 36 :491 (1870). Fimbristylis Cumingii F.-Vill. in Naves & F.-Vill. Noviss. App. Fl. Philipp. 308 (1882). wr uA 7 The species of the section Reigera (Opiz Seznam 83,—1852) were treated under the sectional name Bolboschenus (Palla, 1904) by Beetle (Amer. Jour. Bot. 29: 82-88,—1942) at which time the earlier sectional name was overlooked. Since that time a study containing some valuable points (Fernald, Rhodora 45 :279-296,—1943) has added to our knowledge of the section. Fernald accepts the eastern North American plants included by Beetle in his composite Scirpus maritimus L. as Scirpus mari- timus, but points out that the American material differs from the European in sufficient degree to form a geographical variety. This in no way changes the status as a variety of the local but very characteristic Scirpus maritimus L. var. Fernaldi (Bicknell) Beetle. Except for this variety, then, the American material of Scirpus maritimus becomes: Scirpus maritimus L. var. agonus (Fernald) Beetle, comb. nov. Scirpus maritimus L. var. Fernaldi (Bicknell) Beetle forma agonus Fernald, Rhodora 45 :288 (1943). 7 7 > A Further study of the Hawaiian material of Scirpus paludosus A. Nels. has revealed that its unusually large, pale and long-rayed JULY, 1944] A NOTE ON ERIOGONUM 47 spikelets amply justified its description as a variety by Hillebrand. On the other hand its affinities, as shown by its lenticular achene, are with Scirpus paludosus (achene lenticular) and not Scirpus maritimus (achene trigonous) which was recognized as cosmo- politan at the time of Hillebrand’s study. Scirpus paludosus A. Nels. var. digynus (Hillebrand) Beetle, comb. nov. Scirpus maritimus L. var. digynus Hillebrand, Fl. Hawaiian Isl. 475 (1888). 7; 7 5 A ERIOPHORUM CRINIGERUM (Gray) Beetle. The proper list- ing of the above combination in the Gray Herbarium Index calls attention to the incorrect form used at the time the combination was originally made (Beetle, Leafl. West. Bot. 3: 165,—1943). A NOTE ON ERIOGONUM BY C. V. MORTON = ; - OF ‘ Smithsonian Institution, Washington, D. C. In 1935 I described? a new species, Eriogonum panamintense, from the Death Valley region of California. At that time only a single collection was known, but shortly afterward two ad- ditional specimens were received from Mr. M. French Gilman, viz., Titus Canyon, Grapevine Mountains, Inyo County, Cali- fornia, June 23, 1935, Gilman No. 1831, and Wildrose Canyon, Panamint Mountains, Inyo County, California, Aug. 12, 1935, Guman No. 2009. The purpose of this note is to point out.a new synonym, Eriogonum reliquum Stokes (Leafl. West. Bot. 2:52,—1937), based on specimens collected in Charleston Park, Charleston Mountains, Clark County, Nevada, by I. W. Clokey (No. 7491), distributed as No. 706 of the Plante Exsiccate Grayane. The flowers are somewhat smaller, but otherwise the Clokey speci- mens are identical with the type of E. panamintense, the publi- cation of which two years previously was undoubtedly overlooked by Miss Stokes. The occurrence of E. panamintense in Nevada was to be expected, for the Charleston Mountains form a natural phytogeographic unit with the Panamint and Grapevine moun- tains of California. 1 Journ. Washington Acad. Sci. 25:308 (1935). 48 LEAFLETS OF WESTERN BOTANY [ VOL. IV; NOS THE TYPE LOCALITIES .OF CERTAIN CALIFORNIAN BLACKBERRIES BY JOHN THOMAS HOWELL ¢? In a recent genetic and taxonomic study of our Californian blackberries by Spencer W. Brown (Amer. Journ. Bot. 30: 686— 697,—1943), a new species and two new varieties were named and described but an inadequate statement was given for the local- ities where the types were collected. Although such data may not be of much importance to a student primarily concerned with the genetics of a group of plants, they are of critical interest to a sys- tematist, and because we have been particularly interested in the systematics of western blackberries at the California Academy of Sciences, I obtained for our records through the cooperation of Miss Annetta Carter the type collection data from the Herbarium of the University of California where the types are deposited. In order that the published record of our blackberries may be system- atically complete, these data are given here: RuBus LEMURUM Brown (Amer. Journ. Bot. 30: 695). Type (pistillate) : Herb. Univ. Calif. No. 666159, collected in ravine 15 miles north of Fort Ross, Mendocino County, California, Brown No. 42015. Rusus ursinus C. & S. var. Mepus# Brown (ibid., p. 696). Type (pistillate) : Herb. Univ. Calif. No. 666171, collected in bog along State Highway No. 20 at Cold Spring in the upper Clear Lake region, Lake County, California, Brown No. 41053. Rusus ursinus C. & S. var. PENTAPHYLLUS Brown (l.c.). Type (pistillate) : Herb. Univ. Calif. No. 666162, collected on U. S. Highway No. 199, 4.6 miles south of Redding, Shasta County, California, Brown No. 41131. SMALL GERANIUM IN CALIFoRNIA. The occurrence of Gera- nium pusillum Burm. f. was to be expected in California since it is commonly established in Oregon and Washington, but one would have anticipated its appearance in the western coastal parts of the state. Instead the first Californian record of this Eurasian species that I] have seen comes from Independence, Inyo County, where specimens were collected by Mark Kerr in 1941.—J. T. Howell. LEAFLETS of WESTERN BOTANY Available A limited number of copies of volumes I, IT, and III at $5.00 each. Single copies of most numbers in these volumes are available at $.40 each. LINDLEyY’s GLOSSOLOGY An excerpt from this most useful work, reissued with an introduction by Alice Eastwood. $.50. ALicE Eastwoop’s Key to the Common Families of Flowering Plants in California and Guide for the Analysis and Description of Flowering Plants. $.25. Address: Joun THomas Howe. California Academy of Sciences, Golden Gate Park San Francisco, California Ne wth a ot ey Oot ; canes } Vou. IV No. 4 LEAFLETS of WESTERN BOTANY CONTENTS PAGE See ist raralorMir hE os ag ves ak yess bok AD R. C. BARNEBY New Western Plants—V ee ated rie TRO es aaa el Setdate Ot aoe ae ALIcE Eastwoop SAN FRANCISCO, CALIFORNIA NoveMBeER 30, 1944 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. PA UO OU ee pevunvenggenapovaggannyocagyserea agg svt ayes gece pe ey Owned and published by A.ice Eastwoop and JoHN THomMaAs HowEti ior teteatce Se} = Bh eae ba a Sd net oe Mae Se, es eer NOVEMBER, 1944] PUGILLUS ASTRAGALORUM 49 PUGILLUS ASTRAGALORUM III BY R. C. BARNEBY The following paper, the third of a series of miscellaneous notes on Astragalus, is based mainly on material gathered by Mr. H. D. Ripley and the writer in the southwestern States. There are also included, however, a few changes in nomenclature which have been found necessary in the course of more extended studies. During the preparation of the manuscript it has been possible to make critical comparisons at the New York Botanical Garden (NY), Pomona College (PO), and the University of Notre Dame (ND), and I wish to acknowledge with gratitude the facilities offered by the curators of these herbaria. I am also especially indebted to Mr. F. W. Peirson and to the California Academy of Sciences for the gift of valuable specimens. Types of proposed entities, as well as most of the other col- lections cited, are deposited in the herbarium of the California Academy of Sciences. Astragalus canovirens (Rydb.) Barneby, comb. nov. Homalobus canovirens Rydb., Fl. Rocky Mts. & Adj. Pl., ed. 2, 1126 (1922). Astragalus Coltoni var. foliosus Jones ex Eastw., Zoe 4:115 (1893), nom. nud. A. Coltoni var. moabensis Jones, Contr. W. Bot. 8:11 (1898). CoLtorapo: Yellowjacket, west of Cortez, Montezuma Co., Ripley & Barneby No. 5362. Utan: north of Blanding, San _ Juan Co., No. 5390. I agree with Rydberg in regarding A. canovirens as a species distinct from, though closely allied to, A. Coltoni Jones, from which it differs not only in the regularly pinnate leaves, but also in the coarser, more heavily pubescent stems, denser racemes of slightly longer flowers, and wider, more coriaceous pods. The stems of A. Coltoni rise from an intricately branched aerial caudex, those of A. canovirens from the summit of a multicipital taproot. Astragalus iodanthus Wats. var. typicus Barneby, nom. nov. A. iodanthus Wats., Bot. King Expl. 70 (1871), sensu stricto. The plant described by Watson is the uncommon form of this variable species characterized by a comparatively long calyx and Leaflets of Western Botany, Vol. IV, pp. 49-64, November 30, 1944. 50 LEAFLETS OF WESTERN BOTANY [ VOL. IV, NO. 4 large violet-blue flowers (fig. 19). I have seen it only in the mountains of northern Nevada between 5000 and 6000 feet, as at Unionville in the West Humboldt Mountains or on Emigrant Pass in Eureka County. At similar altitudes near Virginia City, whence came part of Watson’s material, and throughout the range of the species, the flowers are commonly smaller and either ochro- leucous or but faintly lilac-tinged. In these two montane variants of var. typicus the relative proportions of the petals remain nearly constant : in desert valleys of northwestern Nevada and adjacent California, however, at altitudes between 4000 and 4600 feet, the species is represented by a race in which the lamina of the keel- petals is much enlarged at the expense of the claws, and the keel itself is exserted beyond the wings and sometimes nearly equals the banner. Since this variation is readily recognizable and occu- pies a distinct altitudinal zone, it may be worthy of separation under the name Astragalus iodanthus Wats. var. diaphanoides Barneby, var. nov., a var. typico nob. presertim carine majuscule alas superantis vel saltem zquantis lamina 6 mm. usque longa unguiculis suis manifeste longiori diversa. Przeterea legumine ob suturam dorsalem plus minusve introversam septiferamque inferne imperfecte biloculari a var. typico sepissime ulterius recedit, sed hzc nota inconstans videtur. NevApDA: between Reno and Dewey, Washoe Co., alt. 4200 ft., Ripley & Barneby No. 5659. Type in Herb. Calif. Acad. Sci., No. 313543. Ibid., No. 4494, flor. Washoe Valley, Dr. Stretch (NY). Quinn River Crossing, Humboldt Co., alt. 4100 ft., Ripley & Barneby No. 4559. Trinity Mts., northwest of Love- lock, Pershing Co., alt. 4500 ft., No. 5637 (forma pube crispula magis copiosa transitum ad A. pseudiodanthum Barneby pre- bens). CaLtForNIA: Smoke Creek, northeast of Viewland, Lassen Co., alt. 4600 ft., No. 5768. Fig. 20—23. Apart from the peculiar flower, the var. diaphanoides is also notable for the structure of the pod. In all but one of the cited collections the dorsal suture is more or less broadly introverted along the lower half of the legume, thereby forming a distinct though narrow partial septum, 0.5—1.5 mm. high, which, on account of the strong dorsiventral flattening of the fruit, divides the cavity into two almost separate locules. As this feature ap- pears to be not entirely constant, it cannot be used as diagnostic. It should be remarked, however, that Rydberg, noticing the sep- NOVEMBER, 1944] PUGILLUS ASTRAGALORUM 51 tum in the legume of Dr. Stretch’s specimen from Washoe Valley, was sufficiently impressed to refer that collection (Bull. Torr. Bot. Club 54:20,—1927) to Hamosa drepanoloba (Gray) Rydb. (A. diaphanus Dougl. ex Hook.), a species with papery-trans- lucent pods endemic to the valleys of the Columbia and John Day rivers in Oregon and Washington, and to this misdetermination - can be traced the Nevada record of H. drepanoloba in the North American Flora. That it was possible for Rydberg to refer two specimens of what is here considered the same species, sensu lato, to two sepa- _ rate genera (Xylophacos and Hamosa) is sufficient indication of the variability and somewhat anomalous character of A. iodan- thus. While the large-flowered extreme of var. typicus is some- times difficult to separate from A. cibarius Sheld., a species naturally placed in X ylophacos Rydb., the more pubescent phase of var. diaphanoides exhibited by Ripley & Barneby No. 5637 presages a definite trend toward A. pseudiodanthus Barneby, a species not known to Rydberg, but one which by his own criteria would be classed with Batidophaca Rydb. Moreover both Jones (Rev. Astrag. 203) and Rydberg himself (Bull. Torr. Bot. Club 52: 144,—1925) have remarked on the extraordinary re- semblance between A. iodanthus and forms of A. lentiginosus Dougl. (i. e., Cystium Rydb.). Thus A. iodanthus, which under various aspects combines salient characters of Cystium, Hamosa, and X ylophacos, and is at the same time related through A. pseu- diodanthus to Batidophaca, furnishes yet further proof (if any be needed) of the unreality of Rydberg’s astragaline genera. ASTRAGALUS NATURITENSIS Pays., Bot. Gaz. 60:377 (1915). A. desperatus Jones, Rev. Astrag. 203 (1923), pro parte, non Zoe 2:243 (1891). Batidophaca desperata Rydb., N. Amer. FI. 24 :319 (1929), quoad syn., non descr. Cotorapo: sandy ledges of cliffs and on detrital slopes be- neath the Cliff Palace, south end of the Mesa Verde, Montezuma Co., alt. 6600—6700 ft., Ripley & Barneby No. 5359. Fig. 16, 17. Authentic material of A. naturitensis, a species which seems to have been reported only from the type-locality in Montrose Co., Colorado, about seventy miles north of the present station, has not been available for comparison, but our gathering, in all but 52 LEAFLETS OF WESTERN BOTANY _ [VOL. IV, NO. 4 the ochroleucous coloring of the petals, answers so perfectly to Payson’s description that one can scarcely doubt the correctness of the determination. It is not known whether Rydberg was familiar with the original A. naturitensis, but it seems unlikely, as the name appears in the North American Flora merely as a doubtful synonym of Batidophaca desperata (Jones) Rydb. In this disposition he followed Jones, who had already suppressed the species in favor of his own A. desperatus, in my opinion quite unjustifiably. It is true that the two are closely allied and similar in general facies, stipule, leaf, and flower, but the pods are entirely different. As will be seen from the accompanying illustration, the legume of A. naturitensis (fig. 17) is narrowly lanceolate in outline, gently falcate up to the gradually acuminate tip, at least four times as long as its greatest diameter and minutely strigu- lose with short, filiform, appressed hairs, in contrast to that of A. desperatus (fig. 18) which, though variable in size, is much more strongly inflated, proportionately twice as wide, abruptly arched towards the apex into a characteristic porrect triangular beak, and villous with long spreading hairs seated on a pustular base. In addition the peduncles of A. naturitensis are more slender and flexuous, inclined to be arcuate-prostrate in fruit, the racemes remain very short at maturity and do not elongate as in A. desperatus, while the pod is held either erect or horizontal to the axis and is never strongly declined. Astragalus desperatus, it is true, is somewhat variable in the density of vesture of the pod, but the hairs are always long and spreading: the corolla, as also, apparently, in A. naturitensis, may be purple, wholly ochro- leucous, or with just the wings and keel purple-tipped. Both the flower and the legume of A. desperatus vary in size within rather unexpectedly wide limits, the smallest extreme having been recognized as var. petrophilus Jones (Batidophaca petrophila (Jones) Rydb.), but this seems to be an insignificant entity. Inthe summer of 1943, one of the driest seasons on record in the Southwest, several collections of A. desperatus were made ~ in southeast Utah and near Grand Junction, Colorado, the type locality. In nearly every case the dimensions of our specimens either agree more closely with var. petrophilus as shown in its type (San Rafael Swell, Utah, Jones, PO) than with the typical form, or are exactly intermediate, and it is only an occasional NOVEMBER, 1944] PUGILLUS ASTRAGALORUM 53 individual, from a locality especially favored by moisture or shaded exposure, that agrees precisely with well-developed A. desperatus. From these observations I conclude that var. petrophilus is merely a depauperate form, and Batidophaca petro- phila should therefore be reduced to synonymy with A. desperatus. Astragalus humistratus Gray var. crispulus Barneby var. nov. ab A. humistrati formis omnibus hucusque descriptis imprimis caulibus foliis calycibusque pube patula crispa undique villosulis, foliis brevissimis 1—3 cm. tantum longis, necnon leguminis minoris vix 7 mm. longi pauciovulati valvulis tenuiter membranaceis manifeste diversa. _ ARIZONA: roadside bank in pinewoods above Nutrioso, Apache Co., alt. 8150 ft., Ripley & Barneby No. 5074. Type in Herb. Calif. Acad. Sci., No. 313637. Fig. 24—26. The group of Astragali immediately allied to A. humistratus Gray has been variously treated by Jones (Rev. Astrag. 81, sequ.) as a single polymorphous species with three named varieties, and by Rydberg (N. Amer. FI. 24:315, sequ.) as a group of about half a dozen species in his genus Batidophaca. The recognized entities are all very similar in habit and structure, and differ chiefly in the length and compression of the pod and in the quantity and distribution of the pubescence. These characters, however, are subject to some fluctuation and Rydberg’s species, even disregarding such obviously minor variations as he recog- nized under the names Batidophaca stipulacea and B. humivagans, are not very clearly defined either morphologically or geographi- cally. Therefore, although the var. crispulus is at least as dis- tinct from the typical phase of A. humistratus as A. Hosackie Greene is from the latter or from A. sonore Gray, I am disposed for the present to follow Jones’ classification and to regard the plant described above merely as another marked variety within the compass of a complex and widely distributed species. In its slender habit, short subsessile leaves, and few-flowered racemes the var. crispulus is likely to recall A. humistratus var. tenerrimus Jones (A. sonore var. tenerrimus (Jones) Kearn. & Peeb.), a form which is known only from the high parks of the Kaibab Plateau, but the resemblance goes no further. The var. crispulus is unique among its close relatives both in pubescence and in the texture of the pod. In all described varieties and segre- gates of 4. hwmistratus, the pubescence is composed of straight, basally or sometimes medially attached, strigose hairs more or 54 LEAFLETS OF WESTERN BOTANY [VOL. IV, NO. 4 less closely appressed to the surface of the leaflets, stems, and pods. Throughout the herbage of var. crispulus, on the other hand, the hairs are of a quite different character, being not at all appressed or even evidently ascending, but rather spreading, curly and tangled, and extremely fine: the vesture of the pod is similar but shorter and mixed with a few longer, straight hairs. The pod itself is smaller than in any of the other varieties, with a maximum length of 7 mm., while it is more strongly curved into almost half a circle and has thinly membranous, at length translucent valves. The corolla becomes ochroleucous immediately after fertilization, but at first anthesis the banner at least is veined and suffused with pale and sordid rose. The var. crispulus was seen in only one locality along Nutrioso Creek in the yellow pine forest of the White Mountains, but it was locally plentiful. The stems were strictly prostrate and rather freely branched, and formed wide circular mats on an open gravel bank. Both typical A. humistratus and its variety sonore (Gray) Jones were also collected in the neighborhood, the former in the shade of pines a few miles to the south towards Alpine, the latter at the upper edge of the pinyon-zone above Eagar, and the differ- ences were very apparent even in the field. Astragalus flexuosus Dougl. var. Diehlii (Jones) Barneby, comb. nov. A. Diehlit Jones, Rev. Astrag. 194 (1923). Ptso- phaca Diehlii (Jones) Rydb., N. Amer. Fl. 24: 324 (1929). Urtan: Buckhorn Truck Trail east of Castle Dale, Emery Co., Ripley & Barneby No. 4732. East of Cisco, Grand Co., No. 5423. CoLorApbo: west of Mack, Mesa Co., No. 5424. Jones himself doubted the specific distinctness of this race, acknowledging that it might be “only an extreme form of A. flexuosus.” In fact, in the Index to the Revision of Astragalus the plant is listed as a variety of A. fleruosus, but this combi- nation, which lacks any reference to the description in the text, must be regarded as a nomen provisorium and is therefore invalid. Astragalus flexuosus Dougl. var. sierree-blance (Rydb.) Barneby, comb. nov. Pisophaca sierre-blance Rydb., N. Amer. Fl. 24 :323 (1929). The var. sierre-blance, reduced by Tidestrom and Kittell (Fl. Ariz. & N. Mex. 207,—1941) to A. flexuosus, is indeed very close to the species, differing mainly in the glabrous legume. NOVEMBER, 1944] PUGILLUS ASTRAGALORUM 55 Astragalus niveus (Rydb.) Barneby, comb. nov. Phaca nivea Rydb., N. Amer. Fl. 24:348 (1929). A. Peirsoni Munz & McB., Bull. S. Calif. Acad. Sci. 31:67 (1932). ? A. Crotalarie var. piscinus Jeps., Fl. Calif. 2:350 (1936), quoad pl. descr., non A. piscinus Jones. CattrorNniA: Algodones Sand-dunes west of Yuma, Imperial Co., Ripley & Barneby No. 2783. The type of Phaca nivea Rydb. (MacDougal in 1904, NY) was collected near the head of the Gulf of California, that of A. Peirsonii (Munz & Hitchcock No. 12132, PO) about a hun- dred miles to the north in southeast Imperial Co., California: it is apparent that they represent the same species. I have not seen Dr. Jepson’s specimens from near Yuma (where A. niveus is plentiful), but from the description it seems probable that the same plant was intended. Astragalus piscinus Jones is a glabres- cent plant which has otherwise been recorded only from the Pacific coast of the peninsula of Baja California. An earlier Astragalus niveus is recorded by the Index Kewen- sis: it appeared, however, merely as a synonym of A. pelliger Fenzl, ascribed by Bunge (Gen. Astrag. Geront. 2: 189,—1869) to de Monbret, and was never validly published. ASTRAGALUS SABULONUM Gray. New Mexico: sandy valley 17 miles south of Shiprock, San Juan Co., Ripley & Barneby No. 5285. Also observed near Farmington, San Juan Co. A new record for the flora of New Mexico, but scarcely sur- prising, since the species is found on dunes along the San Juan River in adjacent Utah, whence it follows the Colorado River southward into Sonora. Astragalus iochrous Barneby, nom. nov. A. violaceus St. John, Research Stud., State Coll. Wash., 1:98 (1929), non A. violaceus Basil., Not. Syst. Herb. Hort. Petrop. 3:115 (1922). Astragalus monoensis Barneby, spec. nov. egregia sed affinitatis adhuc dubiz, hinc stipulis connatis Chetodontibus Gray inter quos ob legumen obcompressum calycem multoties superans anomala, vel Ocreatis Gray (Batidophace ser. Humistratis Rydb.) a quibus legumine subbiloculari graviter discrepat, illinc A. inyoensi Sheld. (Titi sp. Rydb.) qui ovario longe stipitato stipulisque disjunctis a nostra differt accedens, nec ulla cum specie (nisi cum A. striatifloro Jones, habitu notulisque compluribus haud absimili sed quoad stylum exsertum carinamque acuminatam omnino singulari) arcte comparanda. 56 LEAFLETS OF WESTERN BOTANY _ [VOL. IV, NO. 4 Herba perennis humilis pulchella cryptophyta indumento e pilis patulis adscendentibusque brevibus plus minusve crispatis constituto preter petalos undique molliter villoso- vel subsericeo-cinerascens: caulibus numerosis e radicis verticalis graciliusculi pluricipitis collo suhterraneo emissis, 1—2 dm. longis, prostratis flexilibus sed haud abrupte flexuosis exangulatis hinc inde divaricatim ramosis, inferne filiformibus nudis, superne magis robustis et confertim foliatis, tota longitudine pube laxiuscule adscendenti indutis: stipulis difformibus, imis scariosis erectis vix 1 mm. longis obtusis sub- perfecte adnatis connatisque (ut caulem ocrea cupuliformi utrinque emargi- nata laxe vaginent) in medias summasque herbaceas ovato-acuminatas 2—3 mm. longas reflexas per tertiam partem tantum connatas breviusque adnatas - gradatim accrescentibus, omnibus extus pubescentibus intus glabris, imis nonnumquam demum glabratis: foliis praesertim patulis, petiolo gracili 3—6 (vel 12) mm. longo incluso 1—3 cm. longis: foliolis (4 vel) 5—7-jugis, approximatis vel foliorum summorum juniorumque valde congestis, ob- longis obovatisve obtusis vel leviter retusis, 2—6 mm. longis, seepissime con- duplicatis, manifeste falcatis, utrinque canescenti-villosulis, sursum secus rachin paulo decrescentibus: pedunculis adscendentibus rectis, folium suf- fulcrans saltem breviter superantibus vel sepius eo dimidio longioribus, in racemum 6—10-florum primo congestum subcapituliformem demum laxi- orem et ineunte fructu circa 1 cm. longum abeuntibus: bracteis subulatis herbaceis anguste scarioso-marginatis 1.5—3 mm. longis pedicellum gra- cilem superantibus: floribus patentibus : calycis membranacei extus sericeo- villosi tubo campanulato vel late tubuloso-campanulato 3—4 mm. longo, FIGURES OF ASTRAGALUS Fig. 1—5. A. platycarpus var. montezume Bby. 1, raceme and leaf x 4; 2, ventral view of pod x 1; 3, lateral view of pod x 1; 4, longitudinal section of pod x 1; 5, cross-section of pod x 2% (all from type-collection). Fig. 6 and 27. A. platycarpus var. typicus Bby. 6, cross-section of pod x 2%; 27, pair of leaflets x 1 (both from type). Fig. 7—15. A. monoensis Bby. 7, habit of young plant x 4% (Ripley & Barneby No. 5833) ; 8 and 9, lateral and apical views of pod x 1% (type) ; 10, cross-section of pod x 2 (type) ; 11, 12, and 13, flower, wing, and keel- petals x 2 (Ripley & Barneby No. 5833); 14 and 15, lower and upper stipules x 4 (Ripley & Barneby No. 5833). Fig. 16 and 17. A. naturitensis Pays. 16, leaves and raceme x %; 17, pod x 1% ( both from Ripley & Barneby No. 5359). Fig. 18. A. desperatus Jones. 18, pod x 1%. Fig. 19. A. iodanthus var. typicus Bby. 19, flower x 2 (type-collection, NY). Fig. 20—23. A. iodanthus var. diaphanoides Bby. 20, flower x 2 (Ripley & Barneby No. 4498) ; 21, base of calyx with bracteole x 4 (Ripley & Barneby No. 4498) ; 22 and 23, ssid gt sections of pods x 1 (Ripley & Barneby No. 5768, 4559). Fig. 24—26. A. humistratus var. crispulus Bby. 24, raceme mb leaf x 1; 25, pod x 3; 26, flower x 3 (all from type). NOVEMBER, 1944] PUGILLUS ASTRAGALORUM 57 ad basin obliquam minutissime bibracteolato (sed bracteole spe in vestigia exigua zgre cernenda reducte sunt vel omnino deficiunt), ovario turge- scenti mox rupto, dentibus subulatis acutiusculis circa 2 mm. longis, ciliatis, sinu lato obtusiusculo inter se separatis: petalis ex albo lutescentibus, vel vexillo dilute purpureo-striatulo carinaque apice maculata : vexillo oblanceo- lato-obovato profunde emarginato, explanato circa 12 mm. longo, in ungui- culum latum sensim angustato, medium versus ad angulum fere rectum retroarcuato, marginibus patentibus vel leviter reflexis: alis subrectis, 10—11 mm. longis, lamina oblanceolata obtusa auriculo parvo reflexo in- cluso 6 mm. longa: carine (7 vel) 8—9 mm. longz petalis secus margines tota fere longitudine connatis, unguiculis rectissimis circa 4.5 mm. longis, laminis oblique semiorbicularibus, inferne per annulum dimidium in apicem obtusum zequaliter arcuatis, superne levissime concavis subrectisve: legu- mine’ sessili (vel ob pericarpium carnosulum in sicco contractum quasi- substipitato) chartaceo valde turgido, ambitu de visu ventrali ovoideo vel ovoideo-acuminato, de visu laterali lanceolato-lunato, basi abrupte obtuso vel truncato, apice acuto, valde falcato et preeter rostrum acutum vel acumi- - natum vacuum compressum obcompresso, 1.5—2 cm. longo, suturis ventrali per totam longitudinem acuta dorsali prominula sed depressa et infra medium legumen introversa ut septum presertim incompletum 1—2.5 mm. altum suturam oppositam vix attingentem efformet, sectione compresse ob- cordata, 6—9 mm. lato, 3—5 mm. alto, valvulis pulchre reticulatis stramineo- purpurascentibus pube brevissima strigoso-villosula indutis : seminibus valde compressis, 2—2.5 mm. longis, circa 2 mm. latis, ad hilum emarginatis, atrocastaneis. CALIFORNIA: near Crestview, Mono Co., 10 Aug. 1938, J. T. Howell No. 14500. Type in Herb. Calif. Acad. Sci., No. 313539. Hills north of Mammoth, sandy plain, alt. 7900 ft., Peirson No. 6093. Valley north of Mammoth and east of Inyo Crater Lake, Mono Co., alt. 7500 ft., Ripley & Barneby No. 5833. Fig. 7—15. Astragalus monoensis was first collected nearly twenty years ago by Mr. F. W. Peirson, to whom the writer is indebted for a portion of his original material. Since then it has been collected twice, always in the region immediately north of Mammoth, the three stations lying but a few miles apart. Mr. J. T. Howell, who kindly communicated the fine specimens selected as the type, alone has obtained the mature fruit. The species is a remarkable one in that it exhibits a combi- nation of characters not previously met with in North America. If the pod were considered the sole criterion of affinity in the genus—a course which Rydberg followed to unfortunate lengths—A. monoensis would be included quite naturally in Tium sensu Rydb., the falcate, obcompressed and imperfectly bilocular * 58 LEAFLETS OF WESTERN BOTANY _ [VOL. IV, NO. 4 legume being not unlike that of A. salmonis Jones (though with the ventral, not the dorsal, suture concave) or that of A. inyoensis (though sessile). By the same system it might equally well be classified with certain varieties of A. lentiginosus Dougl. (Cystium Rydb.), though differing in the campanulate calyx. It departs, however, from all these species and their close allies in the connate stipules, a character which has proved of fundamental import almost throughout the genus, both in Asia as well as in the New World. Yet an attempt to attach the species to some natural group characterized by connate stipules is scarcely more successful. In general aspect A. monoensis is likely to recall A. sonore Gray (sect. Ocreati Gray), but in that species, as well as its relatives, the stipules are scarious throughout, and the pod has no vestige of dissepiment. Perhaps more closely allied are the Chetodontes (A. Austine Gray etc.), but these are of rather different aspect, the long-villous, compressed pod is but little exserted from the calyx, and the keel of different shape. The species most closely resembling A. monoensis in the form of the stipules, in pubescence and in general habit is the singular A. striatiflorus Jones (sect. Cystiella Barneby), confined to a small area about Zion Park in southern Utah: but this species, with its attenuate keel, exserted style and membranous legume, seems otherwise remote. Yet again, in the general structure of the pod as well as in the obscurely bibracteolate calyx, A. mono- ensis may be linked with the very different A. todanthus var. diaphanoides Barneby. It is probable that the species will eventu- ally constitute a section apart. In a narrow valley surrounded by pinewoods on the east slope of the Sierra Nevada north of Mammoth, A. monoensis is found in flat open places where the soil is composed of fine white gravel mixed with sand. Here, in a xerophytic association of low herbs which includes Hulsea vestita Gray, Genothera xylocarpa Cov., and Eriogonum esmeraldense Wats., the species occurs locally but in some quantity, forming mats of silvery herbage closely pressed to the ground, or more often climbing up through the twigs of a dwarf Artemisia, when the whole plant becomes more lax and green. The numerous stems, filiform below but branched and stouter above, arise from the multicipital, subterranean crown of a slender taproot, rather in the manner of A. lentiginosus var. 2. a ' NOVEMBER, 1944] PUGILLUS ASTRAGALORUM 59 ineptus (Gray) Jones which grows nearby. The short leaves with their congested leaflets vested in villous hairs, the subcapi- tate racemes of proportionately large ochroleucous flowers, and the peculiar stipules are characteristic. Astragalus remotus (Jones) Barneby, comb. nov. A. ar- rectus var. remotus Jones, Rev. Astrag. 162 (1923); Clokey, Madrofio 6:217 (1942). Tium remotum (Jones) Rydb., N. Amer. Fl. 24:391 (1929). In the original description Jones characterized his variety as “the extreme variation [of A. arrectus] caused by aridity and hot climate,” and stated that “all these forms intergrade from one to the other as you go south.’ There is, however, no evidence whatever that A. remotus intergrades with A. arrectus or its allies. Since its discovery by Jones at Good Springs, Nevada, the species has been collected on numerous occasions in the can- yons of the Charleston Range by Clokey (cf. loc. cit.), Maguire, and the writer (No. 2897, 3371), and from the copious material now at hand it is evident that it there maintains its identity. It has been found nowhere else. Astragalus eremiticus var. typicus Barneby (A.’ arrectus var. eremiticus (Sheld.) Jones), the “form” with which, from its southerly occurrence, one might expect A. remotus to intergrade, is amply distinguished by its larger calyx and flower, as well as by the long stipe and narrow partition of the pod. Astragalus platycarpus (Rydb.) Barneby var. typicus Barneby, comb. et var. nov. Tium platycarpum Rydb., N. Amer. Fl. 24:387 (1929), Bull. Torr. Bot. Club 57:398 (1930). Fig. fie 7. This species, perfectly distinct from its nearest allies A. Drum- mondiu Hook., A. scopulorum Porter, and A. racemosus Pursh, is inadequately known from the type only, a single branch bearing one raceme in advanced flower and one in ripe fruit (NY). It was collected by Brandegee in southern Colorado in 1873, but the precise locality, recorded on the label as Oak Creek, has not been identified. A collection from the Mesa Verde, which I had at first supposed to represent the long-lost A. platycarpus, proves on comparison to differ in a number of details and may be de- scribed as Astragalus platycarpus (Rydb.) var. montezumz Barneby, var. nov. 60 LEAFLETS OF WESTERN BOTANY ~[VOL. IV, NO. 4 -florum leguminisque structura cum var. typico sat bene congruens, sed ab eo foliolis anguste linearibus 0.3—3.5 mm. tantum latis utrinque cinereis, nec ellipticis 6—9 mm. latis superne glabratis, racemis laxioribus axin pri- mariam multo superantibus, necnon legumine striguloso (nec glabro) magis obcompresso graviuscule absimilis. CoLoravo: plateau at the south end of the Mesa Verde, Montezuma Co., alt. 6900 ft., Ripley & Barneby No. 5357. Type in Herb. Calif. Acad. Sci., No. 313546. Fig. 1—5. The final status of 4. platycarpus var. montezume@ must in- evitably await further collections of Brandegee’s plant, when it may prove to be merely an extreme form of the species. From the material now available, however, I judge it to be a distinct race characterized by its narrow, strigose-cinereous leaflets, very long and lax racemes, and by the strigulose, more strongly ob- compressed legume: in the structure of the flower the two varieties are nearly identical. The var. montezume, a rather coarse, erect, perennial herb with stout fistular stems 5—6 dm. tall, ochroleucous petals, and pendent, long-stipitate, linear-oblanceolate, retroarcuate pods, is locally common along the rim of the Mesa Verde overlooking the canyon of the Mancos River, where it occurs on flat sandy ground among the junipers and pinyons. Although abundant in its zone, in the parts of the Mesa which rise above 7000 feet it is abruptly replaced by the superficially similar A. scopulorum Porter. Astragalus Paysonii(Rydb.) Barneby, comb. nov. Hamosa Paysonii Rydb., Bull. Torr. Bot. Club 54:22 (1927) ; N. Amer. FI. 24:425 (1929). Apparently a well-marked species, differing from A. umbra- ticus Sheld. in the pubescence and smaller flowers. It is known to me only from the type (Payson & Payson No. 2748, NY) which came from western Sublette Co., Wyoming, at a point some 600 miles east of the nearest known station for A. umbraticus. Astragalus Bigelovii Gray var. typicus Barneby, var. nov. A. Bigelovii Gray, Pl. Wright. 2:42 (1853), sensu stricto. Astragalus Bigelovii Gray var. mogollonicus (Greene) Barneby, comb. nov. A. mogollonicus Greene, Bull. Torr. Bot. Club 8:97 (1881). A. Bigelovii fma. MacDougal Gdgr., Bull. Soc. Bot. Fr. 48: xiv (1902), non A. MacDougali Sheld. NOVEMBER, 1944] PUGILLUS ASTRAGALORUM 61 A. Bigelovii auct. pro parte. A. Thompsone Kearn. & Peeb.. Fl. Pl. Ariz. 489 (1942), pro parte, non Wats. Arizona: Colorado Plateau south of the Grand Canyon, L. F. Ward in 1901 (NY). Southwest of Frazier’s Wells, Coco- nino Co., among pines, alt. 6500 ft., Ripley & Barneby No. 5220. Near Flagstaff, Coconino Co., MacDougal No. 128 (NY, type- collection of fma. MacDougalii Gdgr.). East of Flagstaff, stony plateau among pines, alt. 6600 ft., Ripley & Barneby No. 3103. New Mexico: bleak grassy summits of the Mogollon Mts., Greene in 1851 (ND, type of A. mogollonicus Greene). Kearney and Peebles (loc. cit.) have recently drawn attention to the existénce in northern Arizona, far out of the natural range of A. Bigelovii, of a plant combining the pod of that species with other characters of A. Thompsone Wats., to which, in fact, they refer their aberrant material. Likewise for some years I have possessed a collection of an Astragalus from Flagstaff, Arizona, which could not be satisfactorily placed either with A. Thomp- sone or with A. Bigelovii, and which I had tentatively identified as Gandoger’s fma. MacDougali. This race, which occurs from the Grand Canyon southeast along the Mogollon Rim into south- western New Mexico, is the plant described long ago by Greene as A. mogollonicus, and, although the name was reduced by Jones (Rev. Astrag. Index) and more recently by Tidestrom and Kittell (FI. Ariz. & N. Mex. 220,—1941) to a synonym of A. Bigelovit, I believe that it is at least sufficiently distinct to deserve yarietal _ recognition. Astragalus Bigelovii var. mogollonicus differs most evidently from A. Thompsone in the smaller, narrower and nearly straight pod with a perfect septum (i. e. not unilocular at the apex), but it is also a lower, more congested plant, with denser racemes and shorter peduncles. The habital characters also serve to dis- tinguish it from A. Bigeloviit var. typicus, which moreover has somewhat larger flowers, pods and leaflets, and a pubescence of shorter, more closely appressed and matted hairs. In stature and pubescence var. mogollonicus most closely resembles A. Bigelovi var. Matthewsii (Wats.) Jones, which I have recently collected near the type-locality, Fort Wingate, New Mexico (No. 5253, 5258), but in that variety the pod is conspicuously inflated. More impressive than the morphological differences, often difficult to 62 LEAFLETS OF WESTERN BOTANY _ [VOL. IV, NO. 4 make out in herbarium specimens, is the perfect geographic segre- gation of A. Thompsone and each variety of A. Bigelovi. Thus A. Thompsone, which, on the structure of the legume, may be treated specifically, is endemic to the Sonoran Zone of the Navajo Basin, its range being approximately limited by the Virgin and Little Colorado rivers to the south and west, the Uintahs to the north, and the foothills of the Rocky Mts. to the east: A. Bigelovu var. typicus is widely distributed in, but confined to, the desert grasslands from southern Arizona east to Texas and south to Durango: while vars. mogollonicus and Matthews occupy dis- tinct territories in the wide band of mountains which divide the two major areas, occurring in open pinewoods of the Transition Zone. ASTRAGALUS CRASSICARPUS Nutt. ARIZONA: stony mesa 4 miles southeast of Eagar, Apache Co., alt. 7750 ft., Ripley & Barneby No. 5049. : Some doubt seems still to exist in the literature as to the occurrence of a member of the section Sarcocarpi Gray west of the Continental Divide. In his monograph of Geoprumnon (Bull. Torr. Bot. Club 53: 164,—1926) Rydberg recorded a specimen from Arizona, without precise locality, which he referred doubt- fully to G. pachycarpum (T. & G.) Rydb., remarking that the legume was atypical for the species: this record, however, was dropped from his later account of the genus in the North Ameri- can Flora. More recently Tidestrom and Kittell (Fl. Ariz. & N. Mex. 221,—1941) reported A. crassicarpus (as A. caryo- carpus Ker) from eastern Arizona, but it is not mentioned in Kearney and Peebles’ Flora of the state (1942). Our specimens, collected in early September, bear fruit only: while the pods exactly resemble those of typical A. crassicarpus in shape and size, the valves seem to be unusually thin-walled. It is hoped that botanists visiting the region in the spring will secure flowering © material for further study. AstraGALus Parryanus Tidestr. Oxytropis Parryt Gray. NevapDA: in calcareous scree, denuded ridge of the Egan Range on the northern slope of Ward Mt., White Pine Gotan 8900 ft., Ripley & Barneby No. 4014. Top of ridge about 2 miles north of Toyabe Dome, Nye Co., alt. 10500 ft., Hitchcock & Martin No. 5610 (NY). NOVEMBER, 1944] | NEW WESTERN PLANTS 63 Not previously recorded from Nevada. The species was long supposed to be confined to the main range of the Rocky Mts., where it is known to extend from Lincoln and Sublette counties, Wyoming, south into the high peaks of northern New Mexico: evidently it is much more widely distributed. Utah specimens have been seen from the La Sal Mts. (Purpus No. 6574, PO) and Abajo Peak (Maguire & Redd No. 1943, PO; Goodman & ’ Hitchcock No. 1413, NY) and from the Wasatch Plateau at Scofield (Jones in 1904, NY). The material from Nevada differs slightly from the typical Rocky Mountain form in the extremely small size of the flowers, the banner being scarcely 8 mm. in length, and the keel, including its apiculate beak, between 5 and 6 mm. One collection from Abajo Peak, however, has equally short corollas. NEW WESTERN PLANTS—V BY ALICE EASTWOOD Arenaria stenomeres Eastwood, spec. nov. Caules dense czspitosi ex radice perenne, 1—2 dm. alti, glabri, graciles, foliosi; foliis nodis inferiori- bus 3-verticillatis, filiformibus, erectis, 2—2.5 cm. longis, rigide acuminatis, basi dilatatis, glabris ; floribus cymoso-paniculatis, bracteis subulatis, rigide attenuatis, ramulis filiformibus erectis; sepalis et pedicellis florium medi- orum et superiorum stipitato-glandulosis ; calyce basi truncato, sepalo ovato- acuminato, superante capsulam ; petalis albis, anguste linearibus, 1 mm. latis, sepalos superantibus 2—3 mm.; staminibus 10, filamentis filiformibus, circa 3 mm. longis ; capsulis ovatis, obtusis, circa 4 mm. longis; seminibus orbicu- laribus, anguste alatis, circa 1 mm. diametro. Type: Herb. Calif. Acad. Sci. No. 320030, collected May 19, 1944, on limestone cliffs in a canyon at the south end of Meadow Valley Range, Lincoln Co., Nevada, by H. D. Ripley and R. C. Barneby, No. 3469. This is related to A. aculeata Watson, differ- ing in the long, narrow, linear petals and capsules surpassed by the sepals. Dodecatheon spilantherum Eastwood, spec. nov. Glabrum, 2—5 dm. altum ; foliis integris, pallido-viridibus, laminis late oblanceolatis, 4—6 cm. longis, 8—15 mm. latis, apice obtusis, basi angustatis ad petiolum margi- natum zquilongum laminis; ramis umbelle circa 6 vel 12 inzequalibus in umbellis; floribus nutantibus, violaceis; calyce 4 mm. longo, segmentis acuminatis longioribus tubo ; segmentis corollz violaceis basi albis, oblongis, circa 1 cm. longis et 2 mm. latis; basi staminum alba tuberculata in lineis circa 1 mm. longa, antheris circa 4 mm. longis, albis, basi maculatis, macula atro-purpurea, triangulari; capsulis erectis apice dehiscentibus in 5 valvis. 64 LEAFLETS OF WESTERN BOTANY [VOL. IV, NO. 4 Type: Herb. Calif. Acad. Sci. No. 312345, collected May 25, 1941, near Sunnyside, northeastern Nye County, Nevada, by H. D. Ripley and R. E. Barneby, No. 3618. It grew in heavy moist soil along the White River, at an altitude of 4900 ft. In June, 1938, Miss Anita Noldeke sent a single specimen of this species which she collected on Whiskey Creek, Mono County, California. I had named it as a new species with this name, but hesitated to publish it until better specimens could be obtained. These that have been selected are much larger specimens, repre- sented by four plants, three with single scapes and one with two. The underground petioles of the leaves are papery-membranous and show no tendency to form tubers. The roots are slender, fleshy, fascicled. The chief distinguishing feature is the tri- angular pointed purple spot on the anthers to which the specific name refers. A New NAME For A CALIFORNIAN GROUNDSEL. While at- tending to some bibliographic details arising from a report on my Sierra Nevada collections of 1944, I noted that the name of the High Sierran endemic, Senecio Muir Greenm. (Ann. Missouri Bot. Gard. 5:56,—1918), is antedated by the South African S. Muirii L. Bolus (Ann. Bolus Herb. 1: 192,—1915). The small Sierran alpine may be properly called Senecio speculicola J. T. Howell, nom. nov.—J. T. Howell. PEREGRINATOR RUBIACEARUM. Another Old World weed is to be added to our California weed flora, Crucianella angustifolia L. Specimens were collected in scrub oak thickets on dry hills near Igo, Tehama County, on May 22, 1944, by Dr. Alan A. Beetle, who reports that it is abundant in the area. This accession to the Madder Family in California is a slender erect annual with vege- tative parts as in Galiwm but with conspicuous white-margined floral bracts that mostly conceal the inconspicuous flowers of the elongate spike. The plant is indigenous to southern Europe and northern Africa.—J. T. Howell. LEAFLETS of | WESTERN BOTANY Available A limited number of copies of volumes I, II, and ITI at $5.00 each. Single copies of most numbers in these volumes are available at $.40 each. LINDLEY’s GLOSSOLOGY An excerpt from this most useful work, reissued with an introduction by Alice Eastwood. $.50. ALICE Eastwoop’s Key to the Common Families of Flowering Plants in California and Guide for the Analysis and Description of Flowering Plants. $.25. Address: Joun Tomas Howe. California Academy of Sciences, Golden Gate Park San Francisco, California Vot. IV No. 5 LEAFLETS of WESTERN BOTANY CONTENTS PAGE Pugillus Astragalorum IV: The Section Diplocystium . . 65 R. C. BARNEBY mame vy ester Platite—-VEiu 2 ee ee a 148 ALIcE Eastwoop meena Pr racclia——lV 8. ae en a. - 50 JoHN THomas Howe. SAN FRANCISCO, CALIFORNIA Marcu 26, 1945 LEAFLETS : a of = WESTERN BOTANY E. 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. WEstT. Bort. ROA MAILN PL INCHES UOLUOOUUTTOOON UMP OD CLLO COLO LLY CELT TEL LOL QOL DOr eo METRIC Owned and published by A.ice Eastwoop and JoHn Tuomas Howe. - MARCH, 1945 ] PUGILLUS ASTRAGALORUM 65 PUGILLUS ASTRAGALORUM IV: THE SECTION DIPLOCYSTIUM BY R. C. BARNEBY INTRODUCTION The study of Astragalus § Diplocystium, the results of which are presented in this paper, was begun in 1940 when it was found impossible by use of the existing monographs of the genus to identify with any degree of confidence or accuracy more than half the forms encountered by the writer in the deserts of Nevada and California. By reference to M. E. Jones’ Revision of Astragalus it was generally feasible to arrive at some sort of determination, but it soon became evident that many of his names covered un- reasonable extremes of variation, or that his descriptions and indications of range were either inaccurate or actually misleading. Recourse to Rydberg’s monograph of the genus Cystiwm in the New World, a much more critical and ambitious treatment than that of Jones, was only rarely more fruitful of results, while both the nomenclature and the specific concept adopted by him in the North American Flora appeared altogether unsatisfactory. The imperfect nature of the current manuals and monographs was made even more manifest by the confusion found to exist in her- baria among plants of this alliance, and the necessity of a critical reevaluation of this difficult and extensive group became doubly apparent. Although a large amount of material has been examined and some clarification of the literature and taxonomy has been achieved, much still remains unknown about the section, and several of the races accepted here are at present ill-defined and perhaps chimerical. Further exploration, particularly in the still obscure mountains of Nevada where § Diplocystiwm is very highly developed, will be necessary before the picture can be anything like complete, while the difficult question of specific limits, here almost completely evaded by the reduction of all the known rela- tives of A. lentiginosus to varieties of that species, can probably be solved only by cultural and genetic studies. The present account is offered rather as a basis for further investigation than as an attempt at a definitive monograph: it is hoped, nevertheless, Leaflets of Western Botany, Vol. IV, pp. 65-152, March 26, 1945. 66 LEAFLETS OF WESTERN BOTANY _ [VOL. IV, NO. 5 that by use of the key, descriptions, and figures it will at least be possible to determine with greater certainty than heretofore the majority of the specimens to be met with in the herbarium and the field. ACKNOWLEDGMENTS In the course of the preliminary studies the writer spent several days at the New York Botanical Garden where most of Rydberg’s types are preserved together with a large quantity of material determined by him. This visit was of special value in giving an insight into the scope of Rydberg’s specific concepts which often (in Cystium at least) do not at all correspond with his published descriptions, prepared, it seems, chiefly from a single sheet. For permission to study at the Garden, I am spe- cially grateful to Dr. H. A. Gleason. I am also much indebted to Dr. P. A. Munz for the loan of the copious material at Pomona College: it is largely from specimens in this collection that the figures were prepared. To Mr. Ira W. Clokey and Mr. Frank W. Peirson I am grateful for gifts of pertinent material. Mr. N. Y. Sandwith of Kew Gardens supplied valuable information about the types of Astragalus lentiginosus and A. diaphanus, thus clearing up a long-disputed nomenclatural point. Finally to the persons in charge of the herbaria listed below, in particular to Miss Alice Eastwood, Prof. M. L. Fernald, Mr. J. T. Howell, Dr. Theodor Just, Dr. T. H. Kearney, Dr. H. L. Mason, Mr. R. H. Peebles, Dr. H. K. Svenson, and Dr. P. C. Standley I am indebted for help and many courtesies. HeErBARIA IN WuicH CITED MATERIAL HAs BEEN SEEN (Symbois in parentheses have been used in the customary manner to indicate the herbarium in which cited material was examined ) Brooklyn Botanic Garden, Brooklyn.) 4.00202.) ee (Br) California Academy of Sciences, San Francisco.........2.2.0.2...::2::0-0-0 (CAS) Chicago Natural History Museum (Field Museum)...........2.02...2-..02.--- (F) Gray Herbarium) Harvard: University=+....2.-2.-0).00-0.2. 2. i ee (G) Herbarium ot) Hl. 1D:|Ripley and) Ry Co. Barneby22)...3. 2 eee (RB) New York Botanical-Garden, New York..........-.....-...::000:--ascsectst-oueeee (NY) Pomona Colleses Claremont ce tcc ee ee Aly a eee (PO) Santa. Barbara betanie Garden, Santa Barbara. 20000. ee (SB) Southern California Academy of Sciences, Los Angeles.................0---- (LA) Was. PieldStation: SacatomArizonats 5 te tek) pu Tl ee eee (SAC) Whriyersity,or Calitormia berkeley. on suse ek ea eee (UC) WniversityohNeyadal Reno. 30 ie ee A ol See (UN) University of, Notre Dame, South Bend... 020 ee (ND) MARCH, 1945] PUGILLUS ASTRAGALORUM 67 History OF THE SECTION DIPLOCYSTIUM The segregation of the species of Astragalus native to the North American continent into species-groups, series, sections, and subgenera has exercised the ingenuity of all the authorities who have dealt with the genus. For our purpose it is not neces- sary to go back beyond Gray’s classic memoir of 1863 (Proc. Amer. Acad. vol. 6) in which he first united the Linnean Phaca, hitherto accepted by Bentham, Nuttall, Hooker, and Torrey, with Astragalus proper, and arranged the species in groups accord- ing to the structure of the pod. Although abandoned as a genus, Phaca was here retained as a major division of Astragalus under the title of ‘‘series” to contain the species with unilocular pods, while those with partially or perfectly bilocular pods were placed in Astragalus series Astragalus. These two series, which were about equivalent to the subgenus of modern taxonomists, were again divided into sections (§) with names formed adjectivally in the nominative plural (the “series” of today). In Gray’s memoir three of the five species in § Diplocystium which had then been published, A. diphysus, A. lentiginosus, and A. Fremonti, were placed together for the first time to form the section Diphyst in ser. Astragalus : of the two others known at that period A. Coul- teri had not been seen by Gray himself, while A. Arthu-Schottu was known only from flowering material, and both were included in another section. By 1871 more material had accumulated from exploration of the West and when Watson published a revised synopsis of the genus in the Appendix to the King Report he was able to make a number of additions and corrections. The Diphysi here include four species, A. diphysus, the newly described A. platytropis, A, lentiginosus, and A. Coulteri. Astragalus Fremontiu and A. Arthu-Schottii were reduced respectively to the two last. The next systematic treatment was Sheldon’s Preliminary List (1894) in which Gray’s section was remodeled and enlarged to contain ten species, and renamed Astragalus ser. Eu-astragalus § Lentigi- nosus. Here we find correctly placed, in addition to the species admitted by Watson, two newly described forms, A. araneosus and 4. MacDougali, together with A. salinus Howell and A. latus Jones. The section, however, is spoiled by the inclusion of A. bajaensis Sheld. with unilocular pods, and A. diaphanus 68 ' LEAFLETS OF WESTERN BOTANY __ [VOL. IV, NO. 5 Dougl., at that time altogether obscure and in any case not closely allied to the rest, while A. Coulteri (with its congener A. eremicus Sheld.) is again, and inexcusably, referred to quite another section. It was partly the unsatisfactory nature of Sheldon’s List which prompted Jones to prepare his own taxonomic arrange- ment, and he brought to the subject not only a profound field- knowledge of the plants themselves, but also an attitude of mind totally dissimilar to that of his predecessors. For a number of years he attempted to perfect a classification based primarily on characters of the flower, stipules, and pubescence, similar, one may imagine, to that applied by Alexander von Bunge to the Astragali of the Old World, and although this proved in the main impracticable for our species, the Revision of Astragalus as it finally appeared in 1923 contained a number of radical inno- vations. Not least among these was the reduction of the Diphysi to a single species with some fifteen varieties which he attached to his section /nflati, a large and somewhat heterogeneous group in which A. lentiginosus was anomalous in having a bilocular fruit. Astragalus platytropis, however, was not included in this general reduction. Meanwhile Rydberg had started that generic segregation of Astragalus species which was to culminate in the North American Flora in the recognition of no less than twenty-seven genera. In 1906 he revived for two of the Diphysi in the Flora of Colorado the genus Cystium, originally proposed by Steven for three spe- cies from the Levant but long ignored by nearly all systematists. This was followed in 1913 by a new batch of sixteen combinations in the same genus and, finally, by raising nearly all Jones’ varie- ties of A. lentiginosus to specific rank and by the description of twelve new species he raised the number of Cystia to thirty-three. The small group of species, similar in aspect to the Diphysi but with legumes seldom inflated, for which Rydberg created his series Palantia of Tium, is here included in the account of § Diplo- cystium. They have had a less complex taxonomic history than the Diphysi proper, for the first two species, A. mokiacensis and A. ursinus, were not described until 1878. Gray considered these to be allied to A. todanthus Wats. Sheldon placed them in his § Sericophyllus, an odd mixture of unrelated and dissimilar plants, MARCH, 1945 | PUGILLUS ASTRAGALORUM 69 while he disposed of the closely allied A. palans in his § Orobot- deus. Jones treated all of these, as well as A. Wilson Greene and A. amplexus Pays., as varieties of A. lentiginosus, a course which has been followed in the present paper. My own views on the relationship of these species have been expressed in a previous note on Astragalus (Proc. Calif. Acad. Sci. IV, 25:163—4, passim,—1944). In the classification proposed below, an elaboration of that in Jones’ Revision of Astragalus, the section Diplocystium is restricted to a single species, A. lentiginosus, with forty varieties. DEFINITION OF THE SECTION The section Diplocystium may be separated from related groups in North America which also bear sessile, bilocular, terete or didymous (but never tricarinate) pods by various combi- nations of characters. The section Sarcocarpi Gray (Geoprum- non Rydb.), which replaces § Diplocystium on the prairies east of the Continental Divide, differs in the fleshy texture of the in- dehiscent or tardily dehiscent legume. From § Uliginosi Gray it is readily distinguished by the pubescence of basifixed, never medially attached hairs: from the caulescent species of § Mollis- simi Gray by the more shortly cylindric calyx-tube, smaller flowers, and strongly arcuate banner: and from §§ Orocystium and Cystiella by the stipules, petiolar or obscurely cauline but ever connate. PHYLOGENY AND CLASSIFICATION The varieties of A. lentiginosus fall readily into three groups distinguished by the size of the flowers and the length of the racemes, and it has been suggested by the writer (Proc. Calif. Acad. Sci. IV, 25:161,—1944) that these might be specifically separated. On purely morphological grounds this arrangement is certainly both convenient and possible, but it would fall far short of a true expression of the numerous cross-relationships manifestly existing between members of each group. Thus var. diphysus and var. tehatchapiensis, judged by structural criteria alone, would be placed in the same species, although their very close relatives var. australis and var. sierre would in each case belong in another, and a similar situation would arise with such obvious 7O LEAFLETS OF WESTERN BOTANY [ VOL. IV, NO. 5 twins as vars. sesquametralis and albifolius or vars. platyphyl- lidius and salinus. ‘ In spite of numerous attempts at breaking up the unwieldy A. lentiginosus into more than one species or into subspecific units, I have failed to find adequate criteria in the plants them- selves to support the desired result. Field-knowledge of more than half of the recognized races has merely deepened my con- viction that Jones was correct in his estimate of the specific limits of the species. The evolution of the section seems to have pro- gressed not as a lineal sequence but as a number of radiating (and sometimes anastomosing) lines which have given rise to forms morphologically similar but of different immediate origins. For example, starting from the complex made up of the closely related vars. salinus, macrolobus, and Fremontti, a chain of races, each link of which is cemented by the existence of structurally (and often geographically) intermediate individuals, can be traced southwestward from central Nevada through the vars. flori- bundus, ineptus, sierre, tehatchapiensis, cesariatus, and nigri- calycis to var. variabilis, and another southeastward through vars. Kennedyi, araneosus, diphysus, and yuccanus to var. australis. When these progressions are laid side by side it becomes apparent that starting from a given point the original stock has produced, as it has been affected by similar climatic conditions, to a series of parallel forms. In attempting a classification which will recon- cile the demonstrable affinities with the visible aspect of the forms of A. lentiginosus, only two consistently satisfactory arrang ments are possible: either each race must be regarded as a dis- tinct species, or all must be united as subsidiary categories of a polymorphic whole. The varieties of A. lentiginosus, as known at present, are not of equal stature: some, indeed, are doubtfully distinct, while others appear to be isolated and might, in another group of the genus, pass as species of the first rank. It is noticeable, however, that every example of the latter type is comparatively little known, whereas all those represented by extensive collections are found to intergrade at some point witha related variety. For this reason it has been thought prudent not to employ a complicated hier- archy of subspecies, varieties, and forms (as will eventually be needed ) to indicate their relative values, though some discussion of the facts will usually be found in the text. MARCH, 1945] PUGILLUS ASTRAGALORUM 71 CRITERIA Legume. In § Diplocystium the pod is exceedingly variable in both size and shape, though in some of the varieties either one or the other may be nearly constant. Thus in vars. typicus, vitreus, and australis the pod can be either broadly ovoid with the ventral suture convex at the middle of the valves or narrowly ovoid-lunate to lanceolate-falcate with both sutures curved up- ward throughout their entire length. In vars. tehatchapiensis, Fremontii, ineptus, variabilis, etc., on the other hand, the outline remains fairly constant, but the length and degree of inflation fluctuate within rather wide limits. Equally unreliable is the character of the pubescence, often used by Rydberg as a criterion, for whereas in a few cases glabreity (e. g. vars. salinus, scorpionis, and Kennedyi) or more often the presence of at least some hairs on the valves (e. g. vars. typicus, variabilis, ineptus, albifolius, stramineus, etc.) appears from the available material to be diag- nostic, in the majority of cases both states occur with almost equal frequency, though seldom in the same population. The true septum in A. lentiginosus (i. e. the partition formed by dorsal introversion of the valves) is typically, and indeed nearly always, complete at the center of the pod, but the degree to which it is produced into the beak of the pod and its absolute width are equally subject to variation. Ina few cases, notably var. mokia- censis, the septum traverses merely half the width of the cavity, and in several it may be complete or fail (on the same plant) to reach the ventral suture. Much more trustworthy as a criterion is the texture of the pod which may be membranous (very thin and translucent), chartaceous (papery and shining but scarcely diaphanous), coriaceous (leathery), or even woody, and con- siderable reliance has been placed on this character in the key and in the definition of races. It should be noticed that the word “chartaceous” has been used by Jones to denote a state which is here described as leathery (cf. var. chartaceus). Calyx. Both the absolute and the relative dimensions of the calyx and calyx-teeth are of the greatest value as criteria in § Diplocystium, though the former especially is to some degree subject to variation coincident with the age and luxuriance of the individual plant. As a rule, however, the length of the teeth relative to the tube is constant, in spite of an occasional specimen 72 LEAFLETS OF WESTERN BOTANY _ [VOL. IV, NO. 5 in which the teeth, stipules, and bracts are all abnormally elongate or abnormally abbreviate, this usually correlated with a similar alteration in the legume. It is important to bear in mind that the measurements of the calyx-tube have been taken along the abaxial (dorsal) side of the calyx: owing to the oblique orifice of the tube this is often longer than that taken along the axial side. Corolla. In most systematic treatments of Astragalus, the greatest length of the corolla, i. e. of the banner, has been used diagnostically, but I have found here that the length of the keel is much more constant and has other advantages. Not only does the banner, being arcuate-reflexed at anthesis, present greater difficulties in precise measurement, but the keel, which exactly enfolds the andreecium and develops with it, has a narrower lati- tude of variation such as one might expect to find correlated with a reproductive organ. Ina few of the varieties of A. lentiginosus there occur two sorts of flowers, one with the normal keel sur- passed by the wings and another with a much elongated keel exserted from between the wings and sometimes equaling the banner: the latter condition, which I have called “‘carinate,” has been observed in vars. carinatus, variabilis, and australis, but the reason for its occurrence is obscure. It should be noted that the length of the keel-petals as given in this paper includes the claws and is not restricted to the blades as in Jones’ Revision. The shape of the keel and other petals is often characteristic, but the differences are so minute and difficult to describe that these points have not been found of great value as criteria. The color of the petals is commonly distinctive, although the varieties with purple corolla may occasionally throw off an albino sport, while the white-flowered varieties are often tinged with pink or bear a purple spot at the apex of the keel. Pubescence. The pubescence of A. lentiginosus is inclined to be variable in density, but the quality of the hairs (short or long, appressed or spreading, straight or curly) is a valuable character. In nearly all the varieties some black hairs may occur on the calyx, pedicels, and axis of the raceme, sometimes in such abundance as to give a distinctive appearance to the whole plant. This character is most unreliable, however, and must be used with great caution: Rydberg described, both in this group and else- where, a number of fallacious species based primarily on the as- sumption that a nigrescent calyx is constant. MARCH, 1945 | PUGILLUS ASTRAGALORUM 73 Habit. The various races of A. lentiginosus differ widely in appearance, depending on stature, disposition and size of the leaves, stems, internodes, and peduncles, and on the density and length of the racemes both in flower and fruit. Since these _ features have proved in most cases to be constant within reason- ably narrow limits, they have been largely used as a basis of classification. Habitat and Distribution. In spite of the fact that the inflated pods of most of the varieties of A. lentiginosus are perfectly adapted to wide distribution by wind, or perhaps for that very reason, each presents an almost uniformly continuous and often mutually exclusive area of distribution limited abruptly, it seems, by factors of soil and climate. Throughout the Basin regions of the western States, from the most arid Sonoran deserts to the bare screes of mountain peaks, the species is repeatedly met with under different guises, yet each race is typically developed only - under certain more or less fixed conditions of temperature, alti- tude, and humidity. The entire complex is no doubt genetically unstable and responds readily to changes in environment, a fact especially evident in those varieties which are widely distributed. From personal observations of var. Fremontu and var. araneosus in the field I am led to believe that each, as defined in this paper, in reality embraces a large number of subsidiary races morpho- logically constant in the restricted area to which each is confined, and actually separable by minute characters of stature, pubes- cence, or other details. So constant are these minor strains, or biotypes, that with experience it is not infrequently possible to guess the approximate provenance of a given specimen without aid from the label, but to recognize these taxonomically is clearly undesirable and of no practical:value. On the other hand, in the systematic treatment presented below, two closely related and similar varieties, when found to occupy distinct and disjunct terri- tories, have been treated as separate units. PoIsoNous PROPERTIES Information concerning the poisonous properties of A. lenti- ginosus has been available from several sources, and there seems to be no doubt that some, if not all, the races of the species are more or less harmful to stock. Jones (Rev. Astrag. 57) states that “various forms of A. lentiginosus have a bad reputation in 74 LEAFLETS OF WESTERN BOTANY _ [VOL. IV, NO. 5 southern Utah.” Paul-Louis Monnet (Explorations Botaniques 141, undated ), writing of var. Fremontii in southwestern Nevada, records that it was there considered “‘un poison violent pour les bestiaux qui vivent dans le désert.”” Kearney and Peebles (FI. Pl. Ariz. 487,—1942) state that A. diphysus is “known to cause loco disease.” A letter from Mr. G. H. Vansell addressed to Miss Eastwood and filed in the herbarium of the California Academy with a specimen of var. araneosus from near Austin, Nevada, declares that this form was observed to be “definitely killing bees” and according to reports had caused ‘“‘severe trouble among horses and cattle.” Indications of damage to stock at- tributed to A. lentiginosus presumably all refer to the true loco disease, caused by the ingestion of poisonous alkaloids present in all parts of the plant. So far as is known, A. lentiginosus does not secrete large quantities of selenium (though sometimes occur- ring where free selenium is available in the soil) and lacks the characteristic odor, described by Jones as “snakelike” and by Tre- lease as “garlicky,”’ which is commonly associated with the sele- niferous species. In fact, experiments with cultures of A. palans conducted by Trelease showed (Scientific Monthly 54:21,— 1942) that that plant is itself poisoned by selenium. SYSTEMATIC TREATMENT ASTRAGALUS § DIPLOCYSTIUM Barneby, Proc. Calif. Acad. Sci. IV, 25: 166 (1944) Astragalus ser. Astragalus § Diphysi Gray, Proc. Amer. Acad. 6: 192 (1863). Astragalus ser. Eu-astragalus § Lentiginosus Sheld., Minn. Bot. Stud. 1: 168 (1894), pro max. parte. Cystium Rydb., Bull. Torr. Bot. Cl. 32: 659 (1905) ; N. Amer. FI. 24: 405, sequ. (1929), exclus. ser. Platytropiis; Amer. Journ. Bot. 17: 232 (1930). Non Cystium Stev., nec Astragalus subgen. Cercidothrix § Cystium Bee. Astragalus § Inflati Jones, Rev. Astrag. 123, sequ. (1923), quoad sp. ult., non aliorum. Astragalus subgen. Diphysus Tidestr., Contrib. U. S. Nat. Herb. 25: 315 (1925). Annual or commonly perennial caulescent herbs of diverse aspect, pros- trate and densely cespitose or erect and open, pubescent with simple, basi- fixed hairs or glabrate: stipules briefly adnate (exceptionally through half MARCH; 1945] PUGILLUS ASTRAGALORUM 75 their length) to the petiole or nearly free, the lower often reflexed, her- baceous or membranous: calyx tubular or tubular-campanulate, usually oblique at the orifice and more or less gibbous dorsally at base, ruptured by the turgescent ovary, ebracteolate: banner arcuate-reflexed at anthesis : keel-petals obtuse but not truncate, the blades mostly lunate, surpassed by or sometimes surpassing the wings: /egume very variable in shape, size, texture and pubescence, bladdery-inflated or merely turgid, globose to lanceolate -in outline, terete or didymous, slightly compressed or obcom- pressed, commonly sulcate along either or both sutures, the body usually sharply differentiated from the laterally flattened, triangular, rarely terete and filiform-tubular, sterile beak, completely (rarely incompletely) bi- locular by dorsal introversion of the valves as a double partition or septum produced across the entire width of the cavity (rarely halfway) at the center of the pod, the beak usually unilocular, the ventral suture filiform or more often produced inwards as two narrow wings which meet and partially enfold the upper edge of the septum proper : ovules few or numerous, seldom all developing : seeds compressed, notched at the hilum, smooth. A single species: A. Jentiginosus Dougl. ex Hook., sensu ampliatissimo Jones, 1. c. 1923. Following the conclusions outlined in the introduction, A. lentiginosus is here accepted as a polymorphic aggregate of forty varieties, which may be keyed out as follows. ARTIFICIAL KEY TO THE VARIETIES OF A. lentiginosus 1. Flowers small, the keel not exceeding 8 mm. in length. 2. Pods distinctly woody or coriaceous in texture. 3. Legume strigose, the valves dull. 4. Ventral suture of the pod sulcate: pod ovoid to lanceolate, broadest below the middle where abruptly incurved.............. cee aioe RM) SRC Wt AO VLG oe RORY On SEY SOS 1. var. typicus 4. Ventral suture acute: pod obliquely ellispsoid, broadest at the middle, regularly falcate.......0.................... 2. var. carinatus 3. Legume glabrous, the valves shining. 5. Pods 7—15 mm. long: plants from over 7000 feet in the moun- tains of Nevada. 6. Stems 5—15 cm. long: leaflets somewhat conduplicate, AO min: lonpk koe ii ee eS 3. var. scorpionts 6. Stems 2—3 dm. long: leaflets flat and thin, 7—19 mm. jf} of. DIAS oP Meaencerae She. PcPar I ee CAEN LY 8 4. var. tremuletorum 5. Pods 15—25 mm. long: plants of alkaline plains, widely dis- GILL 5 ¢ afe RA AOE Rl 1)eSo SRRTRNRED oy Dae 5. var. salinus 2. Pods papery or membranous, usually diaphanous. 7. Beak of the pod, when differentiated from the body, triangular or deltoid, not tubular-filiform. 76 LEAFLETS OF WESTERN BOTANY _ [VOL. IV, NO. 5 8. Beak of the pod straight or ascending. 9. Leaves 6 cm. long or more, if less, then some of the leaflets well separated or over 12 mm. long. 10. Leaflets all flat. 11. Calyx-teeth less than half as long as the tube. 12. Flowers white or ochroleucous, the wings sometimes tinged with pink and the keel purple-tipped. 13. Leaflets subglabrous, at least above. 14. Fruiting raceme dense, cylindric or globose, about half as long as the leaf: stems pros- trate, 25—40 cm. long, often branched: Wo Nevada. So ee 8. var. floribundus 14. Fruiting raceme rather loose, about equaling the leaf: stems ascending, 5—30 cm. long, simple: widely distributed............ 5. var. salinus 13. Leaflets pubescent on either side: racemes loose RE Ars. te. veenn st Aes MORSE s Ren mee 6. var. macrolobus 12. Flowers purple. 15. Pods glabrous or if pubescent then fine-strigillose with short straight hairs and the racemes over 5 cm. long : widespread in S. Nevada and adja- cent Calitorniaznce sc eee 7. var. Fremontts 15. Pods pubescent with longer incumbent hairs: ra- ceme 1—3 cm. long: local in the San Gabriel MtsofsSeCalitormia. =. et 13. var. antonius 11. Calyx-teeth half as long as the tube: racemes commonly loose and surpassing the leaves.................- 6. var. macrolobus 10. Leaflets conduplicate. 16. Stems 2.5—4 dm. long : racemes very dense in fruit, many- MOW CREM Ee Seton Kem ee eke 8. var. floribundus 16. Stems 7—15 cm. long: racemes few-flowered and loose in si G6 LN eee Ee cs Os 0) ARB ahi Soe AC BC ieee a ey 10. var. semotus 9. Leaves 2—5 cm. long, the leaflets conduplicate and crowded, 5—10 mm. long. 17. Calyx-teeth 1.5 mm. long or more...............2..-..-------- 11. var. ineptus 17. Calyx-teeth 1 mm. long or less. 18. Leaflets glabrate above, ciliate...............-2.-:---- 12. var. sierre 18. Leaflets densely sericeous on either face.........- 13. var. antonius 8, Beak of the pod’ declined’... ee ee ee ee 9. var. albifolius . MARCH, 1945] PUGILLUS ASTRAGALORUM 77 7. Beak of the pod filiform-tubular, appearing like a persistent style. 19. Pods ovoid, the body 9—12 mm. long: leaflets 7—10 pairs PE A ee en ener os Oe! eg. ak aN oe eee 14. var. charlestonensts 19. Pods globose, the body 6—8 mm. long : leaflets 4—7 pairs POS fee TESS | BOSS tes Se ee te, oo Peek Ne aE 15. var. kernensis 1. Flowers larger, the keel 8.5 mm. long or more. 20. Racemes short and dense, not exceeding 4 cm. in fruit. 21. Keel exserted, exceeding the wings...1.................. 2. var. carinatus 21. Keel equaling or commonly shorter than the wings. 22. Pods strongly inflated, more than 6 mm. in greatest diam- eter, or, if less, then sub-globose and not exceeding 13 mm. in length. 23. Flowers whitish or ochroleucous, rarely pink-tinged or the keel purple-tipped, but never purple throughout. 24. Pods coriaceous or very firmly chartaceous. 25. Legume not exceeding 13 mm. in length, Slabrousn is 4) ee eo 3. var. scorpionis 25. Legume over 2 cm. long, or, if shorter, then strigose. 26. Calyx-teeth 1 mm. long, one-fourth the length of the tube..19. var. tehatchapiensis 26. Calyx-teeth 2 mm. or more long, about half the length of the tube... ete en a GPS 16. var. platyphyllidius 2A.:'Pods membranous....0../0000 0s 17. var. toyabensis 23. Flowers always purple. 27. Calyx loosely black-pubescent........ 32. var. ce@sariatus 27. Calyx merely strigose. 28. Leaves much exceeding the internodes : stems not exceeding 4 dm. 29. Pods pubescent even in age: calyx-teeth 0.75—1 mm. long. 30. Peduncles much exceeding the leaves: pods 15 mm. long or less................-. ion Mien earns ax ee 18. var. idriensis 30. Peduncles surpassed by the leaves: pods 2—2.5 cm. long............-.00----+ 4 RAEN SS 19. var. tehatchapiensis 78 LEAFLETS OF WESTERN BOTANY __ [VOL. IV, NO. 5 29. Pods glabrous, at least in age, or if pubescent then the calyx-teeth over 1.5 mm. long. 31. Racemes very shortly pedunculate, much surpassed by leaves : stems cespitose, prostrate, 5—15 cm. long. 32. Pods subglobose and abruptly short-beaked : sep- tum produced to the apex.................. 20. var. latus 32. Pods obliquely ovoid, long-acuminate, the beak dniloctlat. an aae eee ee 21. var. chartaceus 31. Racemes about equaling the leaves: stems 1.5—4 dm. long. 33. Pods chartaceous: peduncles widely spreading: local in the Grand Canyon, Arizona................ RUE GSE EHS RN) See als rs MO 37. var. oropedit 33. Pods leathery : peduncles strict. 34. Legume broadly ovoid, broadest toward the middle, shortly and abruptly beaked, the sutures little curved: calyx white- strigose: herbage subglabrous: Arizona to New Mexico and Colorado south of the Colorado River..............23. var. diphysus 34. Legume ovoid-acuminate, broadest toward the base, or, if as above, then with an elongate ascending beak, the sutures com- monly arcuate throughout: herbage can- escent, at least in youth, rarely glabrous: Nevada and Utah................22. var. araneosus 28. Leaves about equaling the internodes: stems prostrate, 5—7 Fi 7 |. | ees ce se OY, a 24. var. sesquimetralis 22. Pods little inflated, lanceolate in outline, not exceeding 6 mm. at the greatest diameter, or three times longer than wide. 35. Petals white, pinkish or ochroleucous : pods ascending. 36. Keel about 13 mm. long: petals white or pinkish : racemes 7—12-flowered: plants of the Mogollon Plateau, RTT ZONA eee eta woh: Sen ago Sa ec 39. var. Wilsonii 36. Keel 11—11.5 mm. long: petals ochroleucous: racemes (usually lax) 15—20-flowered: plants of the Salt River GWanyon, Arizonay. ver bie 2704 23) 40. var. maricope 35.) Petals purple: pods spreading. au.06 2 ee 36. var. palans 20. Racemes loose and open, or, even if rather dense in fruit, the axis exceeding 4 cm. ee ee MARCH, 1945] PUGILLUS ASTRAGALORUM 79 37. Pods strongly inflated, ovoid to subglobose, exceeding 7 mm. at the greatest diameter. 38. Legume coriaceous to firmly chartaceous, glabrous. 39. Peduncles shorter than the leaves: racemes not exceeding 6 cm. in length: flowers drying purple............ 23. var. diphysus 39. Peduncles very stout and about equaling the leaves : racemes 7 cm. long or more: flowers drying violet..................-....- se SEARO J) Ua aU NG Duct PAE TEE gt 38. var. mokiacensis 38. Legume membranous or chartaceous, if firmer then strongly pubescent. 40. Whole plant variably pubescent, but the pod neither silvery- sericeous nor white-tomentulose. 41. Petals purple, purplish-pink, or lilac-tinged, but neither pure white nor ochroleucous. 42. Pods membranous to chartaceous, glabrous or mi- nutely ‘strigulose. 43. Leaves strigose beneath or glabrate: calyx stri- gose, if loosely so, then the corolla very pale, 44. Leaves more than 8 cm. long: leaflets strigose beneath, 1—2 cm. long. 45. Keel 12—15 mm. long: petals purple........ Pear DAIL at SS ADS a 25. var. australis: 45. Keel 10—11 mm. long: petals very pale.... Ban NeLy ene s 2.0 CLE Zoe VATE INC COS 44. Leaves 6—8 cm. long: leaflets glabrous, 6—12 mitt.) Tora. 2.8 Se 2 ae oa Raa | ORY PIES 43. Leaves villosulous beneath: calyx loosely pubes- cent: petals bright purple.......... 28. var. Kennedyi 42. Pods chartaceous to subcoriaceous, pubescent with incumbent or curly hairs: flowers always bright purple. 46. Calyx-tube not exceeding 5 mm. in length: stems canescent, erect or ascending. 47. Teeth of the calyx half as long as the tube: extra-Californian.............. 29. var. stramineus 47. Teeth of the calyx one-third as long as the tube: Mohave Desert of southern Cali- HOUTA ee ey eed ML seh i 30. var. variabilis 44, Calyx-tube 6 mm. long: stems glabrate, sinuous ANC PFOStLA Ler scene 32. var. c@sariatus 80 LEAFLETS OF WESTERN BOTANY __ [VOL. IV, NO. 5 41. Petals pale yellow, ochroleucous or pure white without trace of purple. 48. Pod loosely pubescent : herbage villosulous beneath: cis- montane California. (2340s ex. 31. var. nigricalycis 48. Pod glabrous or minutely strigulose: herbage strigose beneath or glabrate : western Arizona....26. var. yuccanus 40. Whole plant, including the pod, silvery-sericeous or white- tomentulose. 49. Calyx-teeth not exceeding 1.75 mm. in length, or the keel OU rrr gess ae oe RAE nn NR ee oe Eee 30. var. variabilis 49. Calyx-teeth 2—3 mm. long, and the keel at least 10 mm. long SCOR 8, Ay dace Pca EN OI, Ober ARAN. SEI TT We 33. var. Coulteri 37. Pods not strongly inflated, nor exceeding 6 mm. in react diameter. 50. Flerhage silvery-semecous....02.. 8 34. var. borreganus 50. Herbage green, strigose or glabrous. 51. Pods decidedly coriaceous. 52. Petals purple or violet. 53. Calyx short, the tube less than 4 mm. long, the teeth lessithan!] mmflong:2 35. var. ursinus 53. Calyx longer, the tube more than 4 mm. long, the teeth more than 1 mm. long. 2. 54. Pods strongly ascending in very open racemes: petals drying violet................ 38. var. mokiacensis 54. Pods spreading, in rather compact racemes: pet- als drying purple or whitish...._....... 36. var. palans 52. Petals white, pinkish, or ochroleucous: pods strongly ascending. 55. Keel about 13 mm. long: petals white or pinkish: racemes 7—12-flowered: plants of the Mogollon PlateatArizonace eee ee 39. var. Wilsonii 55. Keel 11—11.5 mm. long: petals ochroleucous: ra- cemes (usually lax) 15—20-flowered: plants of the Salt River Canyon, Arizona......40. var. maricope $1. Pods thin-chartaceous to membranous. 56. Keel 10—11 mm. long: leaves 6—8 cm. long: S. Utah ATCT IN VARTA Z Older sece ear ate ere ac een aes 27. var. vitreus 56. Keel 12 mm. long or more: leaves more than 8 cm. long: south of the Mogollon Plateau................ 25. var. australis MARCH, 1945] PUGILLUS ASTRAGALORUM 81 DESCRIPTIONS 1. var. typrcus Barneby Astragalus lentiginosus Dougl. var. typicus Barneby, nom. nov. A. lentiginosus Dougl. ex Hook., Fl. Bor.-Amer. 1: 151 (1831), sensu stricto; Don, Gen. Syst. 2: 257 (1832) ; T. & G., Fl. N. Amer. 1: 333 (1838) ; Walp., Repert. 1: 700 (1842) ; Gray, Proc. Amer. Acad. 6: 194 (1863) ; Gray, Bot. Calif. 1: 147 (1880), exclus. syn.; Wats., Bot. King Explor., Append. 435 (1871), pro parte, exclus. syn., non op. cit. 65; Wats., Bibl. Ind. 195 (1878), exclus. syn.; Macoun, Catal. Canad. Pl. 110, 507 (1883) ; Sheld., Minn. Bot. Stud. 1: 170 (1894) ; Jeps., Fl. Calif. 2: 356 (1936), exclus. syn.; Peck, Man. Pl. Oreg. 449 (1941) ; H. T. Rogers, Research Stud. State Coll. Wash. 10: 124 (1942) ; Abrams Ill. Fl. Pac. St. 2: 599 (1944), pro max. parte. A. lentiginosus “typical” Jones, Rev. Astrag. 123 (1923). Tragacantha lentiginosa (Dougl.) O. Kze., Rev. Gen. 2: 946 (1891). Phaca lentiginosa (Dougl.) Piper, Contrib. U. S. Nat. Herb. 11: 368 (1906). Cystium lentiginosum (Dougl.) Rydb., Bull. Torr. Bot. Cl. 40: 50 (1913) ; Fl. Rocky Mts. 491 (1917), pro parte; N. Amer. Fl. 24: 412 (1929). C. salinum Rydb., Fl. Rocky Mts. 492, quoad pl. descript., non Astragalus salinus Howell. Astragalus lentiginosus var. scorpionis Jones, Rey. Astrag. 124 (1923), pro parte. ? Pods ovoid and inflated or sometimes lanceolate and merely turgid, abruptly incurved at the middle or a little below, broadest toward the base, leathery or very firmly chartaceous in texture, 1—2 cm. long, 3—9 mm. in greatest diameter, somewhat sulcate dorsally and ventrally, straw-colored at maturity and occasionally mottled, strigose with short appressed hairs: calyx sparingly strigose with white or commonly some black hairs, the tube 3—4 mm., the teeth 1.75—2.75 mm. long: banner narrow, about 10 mm., the wings 8 mm., the keel 6.5—7 mm. long: leaves 5—10 cm. long, petiolate, with 5—8 pairs of flat, obovate, oblong or broadly elliptic to suborbicular obtuse or retuse leaflets 7—15 mm. long, subglabrous or with a few scattered hairs on the margins and midribs: stems rather coarse, procumbent or ascending, 2—3 dm. long. Illustrations. P\. I, figs. 1—5. Jones, Rev. Astrag., pl. 23. Abrams, Ill. F1., fig. 2878. Type-locality. Subalpine ranges of the Blue Mountains of Northwest America (NE. Oregon), the type collected by David Douglas. Distribution. Sterile plains and hillsides, commonly in sagebrush, from interior Washington and Oregon to Idaho and (according to Macoun and Sheldon) British Columbia. Exsiccata. British Cotumsia. Nicola Valley, Kamloops (fide Ma- coun, 1. c.). WAsHiInGcTonN. Washington Territory, N. Transcontinental Survey, No. 715. Okanogan Co.: Coulee City, J. W. Thompson No. 9100 (NY, PO), Piper No. 3885 (G) ; N. of Dry Falls, Rogers No. 448 (CAS). 82 LEAFLETS OF WESTERN BOTANY _ [VOL. IV, NO. 5 Adams Co. : Cow Creek, Columbia Valley, Lyall in 1860 (G) ; Washtucna, Cotton No. 975 (G). Benton Co.: Horse Heaven Hills S. of Prosser, G. N. Jones in 1930 (PO). Lincoln Co.: Sprague, Sandberg & Leiberg in 1893 (Br, UC). Yakima Co.: sine loc., Henderson No. 2353 (G), Brandegee in 1893 (UC). Kittitas Co.: Vantage, J. W. Thompson No. 13706 (F, G, NY, UC), No. 11440 (CAS, G); Ellensburg, Piper No. 2674 (G, NY). Orecon. Blue Mts., Douglas (G, a single pod ex herb. Hooker, isotype). John Day Valley, Howell No. 264 (G). Morrow Co.: Boardman, J. W. Thompson No. 4881 (G). Union Co.: sine loc., Cusick No. 112 (G). Sherman Co.: DeMoss Springs, Peck No. 9949 (F, NY). IpaHo. Sine loc., Leiberg No. 132 (PO). Kelly’s Hot Springs, Mulford in 1892 (NY). Canyon Co.: Emmett, Macbride No. 896 (F, G, NY, PO, UC). Bannock Co.: Pocatello, Mrs. Soth No. 399 (NY). Blaine Co.: Hyndman Creek, J. W. Thompson, No. 13616 (G). The var. typicus is most characteristically developed in sage- brush areas of the Columbia River Valley, where the common form is that with the strongly inflated, obliquely ovoid pod. In much of the mature material from Idaho, however, the legume is merely turgid, narrowly ovoid or lanceolate in outline, though the plants differ in no other particular. As the variety passes south- ward it is sometimes difficult to separate from var. carimatus and var. salinus: for example, in Thompson No. 13616 the“pods are strigose but approach that of var. salinus in texture, while speci- mens of var. carinatus from Klamath Co., Oregon, resemble var. typicus in the atypically inflated legume. The Macbride gathering from Emmett, Idaho, was distributed with a printed label under a binomial attributed to Gray, the de- termination having been verified by Aven Nelson. I cannot find that this combination was ever proposed by Gray, and it is per- haps a lapsus for A. lentiginosus var. floribundus Gray. This same specimen, as well as Brandegee’s collection from the Yakima region, was regarded by Jones as part of his var. scorpionis, but differs in its much greater stature, strigose pod, and northern, lowland habitat. 2. var. CARINATUS Jones Astragalus lentiginosus Dougl. var. carinatus Jones, Rev. Astrag. 125 (1923). Cystium lentiginosum Rydb., N. Amer. FI. 24: 412 (1929), pro parte, ex syn. C. cornutum Rydb., |. c., pro parte, saltem in herb. A. lentiginosus var. cuspidocarpus Jeps., Fl. Calif. 2: 356 (1936), pro parte, ex num. cit., non A. cuspidocarpus Sheld., sensu stricto, nec A. lentt- ginosus var. cuspidocarpus Jones, nec A. missouriensis var. cuspido- carpus Jones. MARCH, 1945 | PUGILLUS ASTRAGALORUM 83 Pods obliquely lanceolate-ovoid, falcate, coriaceous, strigose, often mottled, less inflated than in the preceding, 11—17 mm. long: racemes 10—18-flowered, 1—3 cm. long, surpassed by the subtending leaf: calyx strigose with white and black hairs, the tube 2.5—3 mm., the teeth 1—1.75 mm. long; flowers whitish or purple-tipped, of two sorts, either “carinate” as in the type, with banner 9—12 mm., wings 8—9 mm. long, and a broad obtuse keel exserted from the wings 9—10 mm. long, or more commonly of normal proportions, with banner 8—9 mm., wings 7—8 mm., and included keel 6—7 mm. long: leaves (2.5) 5—9 cm. long, petiolate, with 5—8 pairs of obovate, oblong or suborbicular flat leaflets (3) 5—10 mm. long, strigose on both surfaces or only beneath, commonly ciliate, occasionally nearly glabrous; stems 1—3 dm. long, prostrate or ascending. Illustrations. P1. I, figs. 6—12. Jones, Rev. Astrag., pl. 24. Type-locality. Baker City, Oregon, on flats, the type collected by Jones in 1902. Distribution. On sagebrush plains or in open forest of Pinus ponderosa, usually on volcanic substrata, from interior southern Oregon to adjacent California. Exsiccata. Orecon. Baker Co.: * Baker City, Jones in 1902 (PO, type). Malheur Co.: *Harper Ranch, Leiberg No. 2110 (G, NY, PO, UC). Harney Co.: base of Steens Mt., Howell No. 379 (G). Klamath Co.: north of Keno, Peck No. 9421 (G); Barclay Spring, Applegate No. 3553 (G). CALiForNIA. Siskiyou Co.: Montague, K. Brandegee in 1887 (UC); Yreka, Greene No. 848 (G); Grenada Station, Heller No. 8062 (F,G, NY, UC). Modoc Co.: sine loc., M.S. Baker in 1893 (UC) ; *Lake Shore, Austin & Bruce No. 2210 (UC); Egg Lake, M. S. Baker in 1893 (UC), Nutting (PO). Lassen Co.: Harvey Valley, J. T. Howell No. 12437 (CAS, PO) ; Chat, Jones, 18 June 1897 (PO) ; Pine Creek, Baker & Nutting (UC) ; W. of Susanville, Ripley & Barneby No. 5724 (CAS, RB). The var. carinatus is one of the most puzzling and ill-defined races and perhaps should be altogether united with var. typicus, from which it differs essentially only in its regularly falcate legume devoid of ventral sulcus. Much variation exists in the cited material, especially in the size and proportions of the flower. The type from Baker City, Oregon, has the relatively broad and long, curiously exserted keel described and figured by Jones, but this is decidedly unusual: only two other specimens, marked above with an asterisk, bear this same peculiarity and it cannot be considered diagnostic. Jones seems to have reached the same conclusion (in spite of his choice of varietal epithet), as in his description he cited a collection from Chat, California, yet neither of the two specimens from that locality at Pomona bears “‘cari- nate” flowers. Leiberg No. 2110, with abnormally large corollas, very broad banner, and exserted purple-tipped keel, is referred 84 LEAFLETS OF WESTERN BOTANY __ [VOL. IV, NO. 5 here only with doubt, in spite of the carinate condition of the flower : it is without fruit and could as well. belong to var. platy- phyllidius. Another troublesome collection is J. T. Howell No. 12437, distributed as ‘‘var. sierre ?” and in the short leaves and small leaflets indeed resembling that variety, but the legume is typical of var. carinatus. The literature dealing with this race is rather confusing. Ryd- berg, perhaps not having seen the type of var. carinatus, reduced the name to synonymy with Cystium lentiginosum: at the same time, however, he annoted such plants as Austin & Bruce No. 2210 and Heller No. 8062, to my mind inseparable from var. carinatus, as Cystium cornutum Rydb., although they by no means correspond with either the type or the description of that species. Jepson disposed of the name in the same manner, but the actual plants he referred to A. lentiginosus var. cuspidocarpus Jones, a name which, as shown by Rydberg (Bull. Torr. Bot. Club 52: 145,—1925) is wholly untenable in this context. Astragalus cus- pidocarpus Sheld., upon which Jones’ variety was founded, is, when restricted to its type, a species with unilocular pods allied to A. cibarius Sheld., and not at all related to A. lentiginosus. Jones was originally deceived by the distribution as A. cuspido- carpus of a specimen, Leiberg No. 171, which does belong to the A. lentiginosus complex and which afterwards became the type of Cystium platyphyllidium Rydb. Later, in his Revision of Astragalus, Jones partially rectified this error by reducing his var. cuspidocarpus (as to the plant understood by him, i. e. the Leiberg element) to A. lentiginosus var. diphysus, and by referring A. cuspidocarpus Sheld. sens. strict. to a variety of A. missouriensis Nutt. The combination A. lentiginosus var. cuspidocarpus, how- ever, having been founded on Sheldon’s species, must always re- main an exact synonym of that species, however erroneously it may have been applied. Apart from nomenclature, Jepson’s var. cuspidocar pus is a mixture, as may be inferred from the synonymy and cited material (cf. vars. idriensis and platyphyllidius). 3. var. SCORPIONIS Jones Astragalus lentiginosus Dougl. var. scorpionis Jones, Rev. Astrag. 124 (1928), sensu restricto. Cystium scorpionis (Jones) Rydb., N. Amer. FI. 24: 411 (1929). Pods ovoid with ascending beak, not much inflated, coriaceous, glabrous, castaneous or brightly mottled, 7—15 mm. long, 4+—6 mm. high, shallowly MARCH, 1945 ] PUGILLUS ASTRAGALORUM 85 sulcate ventrally : racemes 1—2.5 cm. long, on peduncles shorter than the leaves: calyx sparingly strigose, usually with some black hairs, the tube 3—4 mm., the teeth 1—1.25 mm. long: petals white or purple-tinged, the narrow banner over 9 mm., the wings 8—10 mm., the kee! 7—8.5 mm. long: leaves 4—9 cm. long, petioled, with 7 pairs of obovate to oblong, obtuse or acute, rather distant and often conduplicate leaflets 4—9 mm. ‘long, pubes- cent below with a few straight appressed hairs: stems ascending, rather stout, 5—15 cm. long. Illustrations. P\. I, figs. 18—24. Jones, Rev. Astrag., pl. 24. Type-locality. Morey Peak, Nevada, the type collected by C. A. Purpus. Distribution. Exposed gravel ridges and mountain peaks between 9000 and 11,300 feet altitude, in the mountains of west-central Nevada. Exsiccata. Nevapa. Nye Co.: Morey Peak, Purpus No. 6365 (PO, type) ; top of ridge, about 2 miles N. of Toyabe Dome, Hitchcock & Martin No. 5599 (NY, PO). Mineral Co.: Mt. Grant, Wassuk Range, Archer No. 7115 (NY, UN). I follow Rydberg in excluding from Jones’s var. scorpionis the northern plants with strigose pods cited after the type in the original description : these are believed to be inseparable from var. typicus, although resembling var. scorpionis in the little-inflated pod. The var. scorpionis is closely related to the next, differing chiefly in its low stature, small leaflets, and commonly smaller | pod and may prove to be merely an ecological form from high and exposed gravelly ridges. Good flowering material is still a desid- eratum. 4. var. TREMULETORUM Barneby Astragalus lentiginosus Doug]. var. tremuletorum Barneby, var. nov., precedenti verosimiliter arcte affinis et forsan nil nisi proles umbrosus vel e solo humidiori ortus, sed imprimis caulibus elongatis, foliolisque sub- duplo longioribus planis, adspectu valde absimilis. Pod obliquely ovoid to oblong-lunate, coriaceous and glabrous, about 15 mm. long: racemes rather laxly 7—10-flowered. 1—2.5 cm. long, on peduncles much shorter than the leaves : calyx sparingly strigose with mixed black and white hairs, the tube 3 mm., the teeth 1.75 mm. long: petals whitish, the banner about 10 mm., the wings 9 mm., the keel about 7 mm. long : leaves 7—10 cm. long, petiolate, crowded, with 6—8 pairs of oblanceo- late, obovate or elliptic, obtuse or often emarginate, flat leaflets 7—19 mm. long, glabrous or with a few scattered hairs on the margins and midribs: stems diffuse, very leafy, 2—3 dm. long. Illustration. P1. 1, figs. 13—17. Type. Meadow, 8500 ft., upper Lamoille Canyon, Ruby Mts., Elko Co., Nevada, P. A. Muns No. 16324 (Pomona College Herbarium No. 254608). Distribution. Mountains of central and northern Nevada, between 8000 and 10,000 feet. 86 LEAFLETS OF WESTERN BOTANY _ [VOL. IV, NO. 5 Exsiccata. Nervapa. Elko Co.: Lamoille Canyon, Ruby Mts., Munz No. 16324 (PO, type, CAS); S. side of Sherman Ridge, S. end of Ruby Mts., Hitchcock & Martin No. 5630 (NY); East Humboldt Mts., Jones in 1897 (PO). Nye Co.:7 miles N. of Toyabe Dome, Hitchcock & Martin No. 5619 (NY, PO). The var. tremuletorum is probably not more than a luxuriant shade form of var. scorpionis, differing in the much longer stems, broader and longer, bright green, flat and thin leaflets, and in the often more inflated pod. The type has falcate pods, but in other- wise similar material from the Ruby Mts. the pods are ovoid with upturned beak. In general appearance var. tremuletorum some- what resembles slender specimens of var. latus, to which Jones referred his own collection, but it can be immediately separated by the small size and pale coloring of the flowers. A field-note attached to Prof. Hitchcock’s specimen from the Toyabe Mts., to the effect that the plant was found growing among aspens, hints that the plant may be mesophytic, and has also suggested the varietal name. 5; 5. var. SALINUS (Howell) Barneby Astragalus lentiginosus Doug!. var. salinus (Howell) Barneby, comb. nov. A. salinus Howell, Erythea 1: 111 (1893) ; Sheld., Minn. Bot. Stud. 1: 169 (1894) ; Peck, Man. Pl. Oreg. 451 (1941) ; Abrams, Ill. Fl. Pac. St. 2:599 (1944). A. diphysus Porter in Hayden, Rep. U. S. Geol. Surv. 5: 481 (1872), quoad num. cit.; Parry, Amer. Nat. 9: 270 (1874), quoad num. cit., non Gray. — A. lentiginosus var. latus Jones, Proc. Calif. Acad. Sci. II, 5: 675 (1895), pro parte, non alibi. A. lentiginosus var. floribundus Jones, Contrib. W. Bot. 8:3 (1898), pro parte, non Gray. Phaca inepta Rydb., Mem. N. Y. Bot. Gard. 1: 246 (1900), quoad pl. cit., non A. ineptus Gray. ' Cystium ineptum Rydb., Bull. Torr. Bot. Club 32: 659 (1906), pro max. parte, exclus. syn. Astragalus ineptus Coult. & Nels., Man. 285 (1909), pro parte, non Gray. Cystium salinum (Howell) Rydb., Fle Rocky Mts. 492, 1063 (1917); N. Amer. FI. 24: 411 (1929), exclus. syn. “A. lentiginosus var. chartaceus.” C. heliophilum Rydb., Fl. Rocky Mts. 491, 1063 (1917). Astragalus lentiginosus var. neptus Jones, Rev. Astrag. 124 (1923), pro parte, i. e. “a form with broad leaflets. .. .” A. heliophilus (Rydb.) Tidestr., Contrib. U. S. Nat. Herb. 25: 325 (1925). Pods variable in shape and size, much inflated, obliquely ovoid to ob- pyriform or subglobose, but always with a well-defined straight or upturned unilocular lanceolate or triangular beak, the valves membranous-diaphanous MARCH, 1945] PUGILLUS ASTRAGALORUM 87 or somewhat firm in texture, glabrous, smooth and shining, usually not mottled, more or less sulcate on either suture, 1.5—2.5 cm. long, about half as broad as high: racemes rather loosely 10—25-flowered, about equaling the leaves: calyx sparingly strigose, the tube 2.5—3. mm., the teeth 1—1.5 mm. long: petals white, ochroleucous, or purple-tinged, the banner 9—12 mm., the wings 7.5—11 mm., the keel 5.5—7.75 mm. long: leaves as in var. typicus, the leaflets 5—8 pairs, glabrate with ciliate margins and midribs or sparingly strigose above, below, or on either face: stems numerous, ascending, 1.5—3 dm. high. Illustrations. Pl. 1, figs. 42—45. Abrams, Ill. Fl., fig. 2877. Type-locality. Saline soils of SE. Oregon, the type collected by Thomas Howell. Distribution. Alkaline flats and about depressions on sagebrush plains, from interior Oregon to southwestern Montana, southward to northeastern California, northern Nevada, and western Utah. Exsiccata. Orecon. Crook Co.: Bear Buttes, Leiberg No. 798 (G). Harney Co.: Harney Valley, Howell No. 799 (F, ND, UC, probably parts of the type-collection) ; Wagontire, J. W. Thompson No. 12167 (G); Narrows, J. W. Thompson No. 12061 (CAS); Burns, Jones No. 25440 (LA, PO). Malheur Co.: Harper’s Ranch, Leiberg No. 2256 (PO, UC) ; Malheur region, Cusick No. 1942 (F, in part, UC) ; summit of Buck Mt., Cusick No. 2618 (F, ND, NY, PO, UC) ; Beulah, Leiberg No. 2301 (UC). IpaHo. Blaine Co.: Ketchum and Guyer Hot Springs, Nelson & Macbride No. 1281 (F, G, NY, PO, UC). Twin Falls Co.: Castleford, Nelson & Macbride No. 1710 (NY). Lincoln Co.: between Owinza and Besslyn stations, Eggleston No. 22053 (G); Shoshone, Nelson & Macbride No. 1172 (G). Custer Co.: Mackay, Nelson & Macbride No. 1575 (F, G, PO, UC). Owyhee Co.: 23 miles SW. of Marsing, Ripley & Barneby No. 6147 (CAS, RB). Montana. Madison Co.: Stinking Water Creek, Hayden No. 5 in 1871 (G). Beaverhead Co.: sine loc., Tweedy No. 203 (NY); Lima, Shear No. 3430 (NY, type of Cystium heliophilum) , No. 4683 (NY), No. 558S (NY), Jones in 1908 (PO). Nevapa. Elko Co.: Mountain City, Nelson & Macbride No. 2199 (G, NY); Wells, Jones No. 25441 (CAS, LA, PO), L. S. Rose No. 32372 (CAS) ; Cobre, Ripley & Barneby No. 5519 (RB). Humboldt Co.: Quinn River Crossing, Griffiths & Morris No. 141 (NY). Co. undetermined: Humboldt Plains, A. Gray in 1872 (G). Lincoln Co.: Deer Lodge, Train No. 2524 (UN) ; Geyser Ranch, Ripley & Barneby No. 6326 (CAS, RB). Uran. Beaver Co.: Frisco, Jones No. 1813 (G, NY, PO). Iron Co.: Iron Mt., Ripley & Barneby No. 4374 (PO, RB) ; Cedar City, Parry No. 48 (G, NY, in part, mixed with A. Wardii Gray) ; Escalante Desert, Ripley & Barneby No. 4969 (PO, RB). Catt- FORNIA. Lassen Co.: Susanville, Eastwood & Howell No. 8335 (PO). Siskiyou Co.: Laird’s Landing to Bray, Peirson No. 3602 (RB). The var. salinus is one of the most widely distributed of all the forms of A. lentiginosus, and I judge it to be the most primitive of its immediate group, perhaps representing the original type from 88 LEAFLETS OF WESTERN BOTANY _ [VOL. IV, NO. 5. which the others are derived. Not only is it the central race which connects all the varieties with small, pale flowers, but also it is possible to trace from var. salinus a gradual progression or se- quence of variation through var. floribundus to the high altitude _ races of the south (var. ineptus and var. sierre) and thence, through var. tehatchapiensis to var. idriensis and var. cesariatus: through var. macrolobus to var. Fremontii and its relatives in the southwestern deserts: and through var. platyphyllidius to vars. latus, diphysus, and palans. As might be expected in a race covering so large an area, con- siderable variation exists in the material accepted as belonging to var. salinus, and it is probably an aggregate of several more local populations marked by trivial but nevertheless more or less con- stant characters. Thus in southwestern Montana and adjacent Idaho the pubescence of the leaflets is more copious (Cystium heliophilum sensu stricto): in northern California both pod and flower are inclined to be larger and the latter of thicker tex- ture: and in its isolated occurrence in southwest Utah the indi- vidual plants are more robust, the flowers are distinctly ochro- leucous with purple-tipped petals, and the pods are somewhat coriaceous and often mottled. All these, however, have in com- mon the characteristic shiny, glabrous pod with ventral suture either straight (except in the beak which is nearly always up- turned) or convex, and I can find no reliable basis for further segregation. The var. salinus has been variously disposed of in the litera- ture. First reported by Hayden in 1872 from Montana as A. di- physus Gray, it was described as a species in 1893 from Oregon material as A. salinus Howell. Jones (1895) promptly reduced this to his A. lentiginosus var. latus, three years later to var. flori- bundus Gray, and finally (1923), as a “form... intergrading . with the type,” to var. ineptus (Gray) Jones. Nelson, and Ryd- berg (as Phaca or Cystium), for long confused it with A. ineptus Gray, but discovering the error, Rydberg (1917) redescribed it under the name Cystium heliophilum, at the same time applying . the epithet salinus, combined in Cystium, to var. typicus nob., and it is only in his final monograph (1929) that C. salinum corre- sponds with the original species of Howell. I have also referred here certain plants such as Leiberg No. 2301 and Griffiths & MARCH, 1945] PUGILLUS ASTRAGALORUM 89 © Morris No. 141 which Rydberg annotated as Cystium platyphyl- lidium. 6. var. MACROLOBUS (Rydb.) Barneby Astragalus lentiginosus Doug]. var. macrolobus (Rydb.) Barneby, comb. nov. Cystium macrolobum Rydb., N. Amer. Fl. 24: 408 (1929), exclus. syn. C. vulpinum Rydb., op. cit. 409. Astragalus lentiginosus Wats., Bot. King Explor. 65 (1871), quoad num. cit., non Dougl. A. lentiginosus var. Fremontii Rothrock, Rep. Geogr. Surv. (Wheeler Exped.) Bot. 93 (1878), non aliorum. Pods membranous, ovoid to narrowly ovoid-acuminate, usually rather long-beaked, 1.5—2.5 cm. long, glabrous or strigillose: racemes 12—25- flowered, commonly rather lax in fruit, 3—7 cm. long, on peduncles of about the same length: calyx copiously pubescent with loosely ascending or incumbent white or black and white hairs, rarely glabrate, the tube 3—4 mm., the teeth 2—3 mm. long: petals ochroleucous with purple-tipped keel, rarely purple throughout, the banner 9—12 mm., the wings 7—9.5 mm., the keel 6—8 (8.5) mm. long : leaves ascending, 4—11 cm. long, with 4—9 pairs of obovate to oblong, obtuse, truncate or emarginate leaflets 0.5—1.5 (2) cm. long, silky- or curly-pubescent on either face, often less so above, or occasionally glabrate and ciliate: stems several, ascending, 1—2.5 dm. long, strigose, from a short-lived perennial tap-root. Illustration. P1. I, figs. 37—41. Type-locality. West Humboldt Mts., Nevada, the type collected by Sereno Watson. Distribution. Dunes, sandy valleys, and river-banks, along the Hum- boldt River and its tributaries from Elko County west to the region of the Carson Sink, Nevada. Exsiccata. Nevapa. Elko Co.: Wells, Jones in 1881 (PO); SE. of Elko, Ripley & Barneby No. 5561 (RB) ; between Elko and Wells, East- wood & Howell No. 278 (CAS, PO). Lander Co.: Grass Valley, Watson No. 254 in part (NY, type of Cystium vulpinum, G). Humboldt Co. : Hum- boldt Lake, Jones No. 3846 (NY, and two sheets in herb. PO labelled “East of Humboldt” and “Rose Creek’) ; Rose Creek, Jones in 1882 (PO, one plant with purple flowers) ; Winnemucca, Ripley & Barneby No. 5603 (CAS, RB). Pershing Co.: West Humboldt Mts., Watson No. 254 in part (NY, type of Cystium macrolobum) ; Humboldt House, Ripley & Barneby No. 4548 (CAS, PO, RB). Churchill Co.: Carson Desert, Wat- son No. 254 in part (NY). Washoe Co.: Truckee River gorge E. of Reno, Ripley & Barneby No. 4455 (RB); W. of Reno, Mason No. 4601 (UC). Co. undetermined: Elko to Winnemucca: Mrs. E. C. VanDyke in 1930 (CAS, flowers purple). Cysttum macrolobum and C. vulpinum seem to be trivial vari- ants of a single race, although in Rydberg’s treatment they are quite widely separated on the fallacious character of pubescence go LEAFLETS OF WESTERN BOTANY [ VOL. IV, NO. 5 in the legume. The type of C. macrolobum has exceptionally long calyx-teeth and the crisp but fine indumentum of C. vulpinum~ appears at first sight peculiar. Nonetheless among my own collec- tions (Ripley & Barneby No. 4548) from the Humboldt House, a locality at the base of the West Humboldt Mts. and type station of C. macrolobum, there are specimens which combine these two outstanding features, while the pods vary from pubescent to glabrous, and the calyces may bear all white or some black hairs. Rydberg himself could not distinguish them: for example, Jones No. 3846 (NY), referred by him to C. vulpinum, has the pod glabrous from the beginning and agrees perfectly with C. macro- lobum. Abrams & McGregor No. 204 (Frazier Borax Mine, in the mountains of Ventura Co., Calif.), on the basis of which Rydberg reported C. vulpinum from California, is in all probabil- ity a hybrid: in any case it does not belong here, in spite of a superficial resemblance (cf. var. tehatchapiensis). The synonym of C. macrolobum, “A. diphysus Wats., Bot. King . . . in part,” quoted by Rydberg in the original description, cannot be correct. The type of the species, Watson No. 254 in herb. Torrey., to which this reference apparently belongs, was distributed as A. lentiginosus, and is referred to under that name and number in Watson’s account of the expedition. Considérable variation has been noted in var. macrolobus. The material from Elko County is somewhat depauperate, with sub- capitate or very short racemes, but with typical calyx and pubes- cence, while the collections from Washoe County combine the loose racemes and calyx of the type with green and glabrate herb- age. Purple flowers have been seen only in part of Jones’s gather- ing from Rose Creek and in that of Mrs. VanDyke. This race is of peculiar interest in that it approaches var. Fremontu (to which C. vulpinum was reduced by Jepson, FI. Calif. 2: 357,—1936, in syn.), and forms a passage between the montane and northern races with densely racemose, small, white flowers and the purple- flowered forms with elongate inflorescence of the deserts to the south. 7. var. FREMONTII (Gray) Wats. Astragalus lentiginosus Dougl. var. Fremontii (Gray) Wats., Bot. King Explor. 66 (1871), quoad nomen, exclus. num. cit.; Gray, Bot. Calif. 1: 147 (1880) ; Jones, Contrib. W. Bot. 8: 4 (1898) ; Rev. Astrag. 127 MARCH, 1945 | PUGILLUS ASTRAGALORUM gI (1923), pro parte; Munz, Man. S. Calif. 269 (1936), pro parte; Jeps., Fl. Calif. 2: 356 (1936), pro parte, exclus. descr. & num. cit. plur. A. Fremontii Gray ap. Torr., Pac. R. R. Rep. 4:80 (1857), exclus. var. ; Proc. Amer. Acad. 6:194 (1863); Sheld., Minn. Bot. Stud. 1: 169 (1894), exclus. syn.; Tidestr., Contrib. U. S. Nat. Herb. 25:325 (1925) ; Tidestr. & Kitt., Fl. Ariz. & New Mex. 216 (1941). A. eremicus Sheld. ap. Cov., Contrib. U. S. Nat. Herb. 4: 86 (1893) ; Minn. Bot. Stud. 1: 140 (1894). A. Coulteri var. Fremontii (Gray) Jones, Proc. Calif. Acad. Sci. II, 5: 670 (1895). A. lentiginosus var. Coulteri Jones, Rev. Astrag. 127 (1923), quoad syn. cit., vix ulterius, non A. Coulteri Bth. A. Coulteri Jeps., Man. Calif. 568 (1925), pro parte, non Bth. Cystium Fremontit (Gray) Rydb., N. Amer. Fl. 24: 407 (1929). C. eremicum (Sheld.) Rydb., op. cit. 409. C. griseolum Rydb., op. cit. 411. Astragalus Fremontti subsp. eremicus (Sheld.) Abrams, Ill. Fl. Pac. St. 2:597 (1944). Pods ovoid with a short triangular beak, much inflated, papery-trans- lucent, shining, glabrous or sparsely strigillose, 1.5—2.5 (3) cm. long: racemes 15—30-flowered, lax in flower, dense or much elongate in fruit, 5—10 cm. long, commonly surpassing the leaves: calyx rather copiously strigose or silky with black or white, mostly appressed or occasionally spreading hairs, rarely glabrate, the tube 2.5—3.5 mm., the subulate teeth 0.75—1.75 (2) mm. long: petals purple, very unequal, with wings commonly much exceeding the keel, the banner 8—12 mm., the wings 6.5—9 mm., the keel 6—7 (exceptionally 7.5) mm. long: leaves 5—8 cm. long, petiolate, with 5—8 pairs of obovate to oblanceolate, obtuse or retuse leaflets 5—17 mm. long, very variable in pubescence, with either both faces glabrous, both faces more or less silky-strigose or subhirsutulous, or with the upper glabrous and the lower sparingly strigose : stems erect or ascending, simple, canescent to glabrate, (0.6) 1—3 dm. high. Illustrations. Pl. I, figs. 46—50. Jones, Rev. Astrag., pl. 25, exclus. fig. a. Abrams, Ill. Fl. fig. 2871. Type-locality. Banks of the Rio Virgin (Nevada), the type collected by Frémont in 1844. Distribution. Most abundant on open sandy deserts associated with Larrea, but occurring also on alkaline clay lake-beds, in rocky canyons, and bare stony slopes of the pinyon-belt, from southern Nevada westward to Inyo, Mono, and San Bernardino counties, California. Exsiccata. Nevapa. Clark Co.: Rio Virgin, Frémont, 2nd Exped. No. 424 (G, type, NY) ; Kyle Canyon, Charleston Mts., Clokey No. 7176 (PI. Exsicc. Gray. No. 723), Ripley & Barneby No. 3413 (CAS, RB); Erie, Tidestrom No. 8866 (NY); Good Springs, Jones in 1905 (PO). Lincoln Co.: Meadow Valley Wash, Jones in 1904 (PO); Caliente, Ripley & Barneby No. 3507 (RB); Highland Range, Ripley & Barneby No. 3499 (PO, RB); White Sage Valley, Train No. 2437 (NY); Pahroc 92 LEAFLETS OF WESTERN BOTANY __[VOL. IV, NO. 5 Range, Purpus in 1898 (PO) ; 20 miles N. of Pioche, Eastwood & Howell No. 7330 (CAS); Crystal Springs, Ripley & Barneby No. 4401 (RB); 25 miles E. of Groom, Train No. 2380 (UN). Nye Co.: between Sharp P. O. and Nyala, Jaeger in 1938 (PO); Currant, Bentley No. 20 (PO) ; NE. of Warm Springs, Eastwood & Howell No. 9453 (CAS, RB); Frenchman Flat, Ripley & Barneby No. 3464 (PO, RB) ; Spotted Range, Ripley & Barneby No. 3434 (CAS, RB). Esmeralda Co.: Lida, Tidestrom No. 9817 (G, NY) ; Klondyke, Ripley & Barneby No. 3655 (RB) ; Gold- field, Heller No. 9613 (NY, CAS) ; Gold Mt., Purpus No. 5954 (PO, UC), No. 5985 (PO, UC). Mineral Co.: Candelaria, Shockley No. 220 (G, type of Cystium griseolum, ND), No. 271 (G, cotype). Ca itFornta. Mono Co.: Devernois Ranch, White Mts., Petrson No. 12458 (RB). Inyo Co.: Cedar Flat, White Mts., Duran No. 3025 (CAS, G, PO, UC) ; White Mts., Purpus No. 5795 (PO, UC) ; N. of Big Pine, Keck No. 544 (UC) ; West- gard Pass, Hoffmann No. 10024 (SB), No. 10026 (SB), No. 15002 (SB) ; Darwin, Ferris No. 7908 (PO, UC); Darwin Mesa, Ferris & Bacigalupi No. 8063 (PO, UC), Hall & Chandler No. 7097 (PO, UC) ; Shepherd’s Canyon, Argus Mts., Jones in 1897 (G, NY, PO), Hall & Chandler No. 7073 (PO, UC); Maturango Peak, Purpus No. 5450 (G, PO, UC); Keeler, Ripley & Barneby No. 3768 (PO, RB) ; Olancha, Hall & Chandler No. 7349 (PO, UC) ; Little Lake, Hall & Chandler No. 7355 (UC), No. 7356 (PO, UC), No. 7357 (PO), No. 7358 (UC) ; Haiwee, Feudge No. 1872 (PO); Pleasant Canyon, Panamint Mts., Hall & Chandler No. 6934 (PO, UC); Dante’s Point, C. L. Hitchcock No. 12333 (PO, UC) ; Sur- prise Canyon, J. T. Howell No. 3923a (CAS); Daylight Pass, Gilman No. 6 (PO); Greenwater, ‘S. of Death Valley, Munz No. 16544 (CAS, PO). San Bernardino Co.: Cove Spring, Old Dad Mts., Jones in 1926 (PO); Halloran Spring E. of Baker, Ripley & Barneby No. 4286 (PO, RB) ; Cima, K. & T. S. Brandegce in 1915 (PO, UC) ; Barnwell, K. Bran- degee in 1911 (UC) ; Valley Wells, Ripley & Barneby No. 3298 (PO, RB) ; NW. of Fenner, Hoffmann No. 21805 (SB) ; Coliseum Mine, Clark Mts., Alexander No. 478 (UC); Mountain Pass, Eastwood & Howell No. 8914 (CAS); Tecopa Pass, Kingston Range, Alexander & Kellogg No. 2368 (UC) ; Fourth of July Canyon, New York Mts., Alexander & Kellogg No. 1329 (UC). The var. Fremontu, while always recognizable by the small purple flowers, is considerably polymorphic and includes several minor ecological races, separable on somewhat fluctuating criteria of stature and pubescence. In general there appears to be a gradual progression westward from Jones’s curiously glabrous collection from the Meadow Valley Wash in southeast Nevada, through the rather inconspicuously strigose-canescent type, to the populations with hirsutulous (Cystium griseolum) or appressed-silvery (Astragalus eremicus) herbage of southwest. Nevada and Owens Valley. In the last, the pod, with rare exceptions glabrous through- MARCH, 1945 ] PUGILLUS ASTRAGALORUM 93 out the rest of the range, is almost consistently strigillose, but I can find no satisfactory cleavage between these races. The type of Cystium griseolum Rydb. is a depauperate individual from rather high altitude (6000—6500 feet), with somewhat spread- ing and tangled pubescence, short peduncles, and condensed racemes, which is linked to the commoner phase by the dwarf, pre- dominantly annual race in western Nye and southern Esmeralda counties, Nevada (cf. Ripley & Barneby No. 3655). Rydberg erroneously described the flowers as “white tinged with rose,” in spite of Shockley’s note on the label stating that the petals, when fresh, were “light magenta, variable in shade,” and the isotype in herb. Greene. bears obviously purple flowers. This montane form is also represented by Duran No. 3025 and by several gatherings from the White Mts. Astragalus erenucus Sheld., of which I have not seen the type, was very minutely described by Sheldon, and is readily identified with the silvery-canescent form with minutely strigulose legume repeatedly met with on the alkaline meadows and mesas about Lone Pine, the type-locality, Olancha, and Darwin. The literature dealing with var. Fremontu is rather difficult to disentangle, as the name, in various combinations, has been made to cover widely divergent concepts, or the plant has been entirely submerged in var. Coulteri. Thus Watson (1871) con- founded it with the white-flowered var. macrolobus, while Jones stated its distribution as “common from Hawthorne, Nevada, .. . to Mexico,” and doubtless included both var. Kennedyi (e loc.) and var. variabilis, as shown by his figure of a plant from Victor- ville under this name (pl. 25). Jepson (1925) regarded A. Coul- teri as embracing all the southern California desert races, though later (1936) he admitted var. Fremontii to the flora of the state in a much more extended sense than admitted here, for specimens both of var. variabilis and of var. antonius are cited. The var. Fremonti, as here restricted, forms a natural unit intermediate between vars. macrolobus, Kennedyi, and variabilis. From the first it differs in the purple coloring and shorter calyx- teeth, and from the other two in the noticeably smaller corolla. An occasional specimen with white petals has been seen growing with the ordinary form. 94 LEAFLETS OF WESTERN BOTANY [ VOL. IV, NO. 5 8. var. FLORIBUNDUS Gray Astragalus lentiginosus Doug]. var. floribundus Gray, Proc. Amer. Acad. 6: 524 (1865) ; Jones, Contrib. W. Bot. 8:3 (1898), exclus. syn. A. lentiginosus Wats., Bot. King Explor. 65 (1871), pro parte, ex num. cit. ; Sheld., Minn. Bot. Stud. 1: 170 (1894), pro parte, non Doug]. A. lentiginosus var. ineptus Jones, Rev. Astrag. 124 (1923), pro max. parte, non A. ineptus Gray. Cystium floribundum (Gray) Rydb., N. Amer. Fl. 24: 411 (1929). C. ormsbyense Rydb., op. cit., p. 412. Pods membranous and shining, ovoid-acuminate to subglobose, 1—2 cm. long, strigose or glabrous, occasionally mottled, gathered into dense cylin- dric or subspherical clusters on short divergent peduncles: racemes 15—20- flowered, (1.5) 3—4 cm. long, equaling or sometimes shorter than the sub- tending leaf: calyx white-strigose, the tube 2.5—3.5 mm., the acute teeth 0.5—1.5 mm. long: petals white, the banner 8.5—10 mm., the wings 7.5—8.5 mm., the keel 6—7 mm. long: leaves 3—6 cm. long, with about 7 pairs of oblong, oblanceolate or obovate, glabrous or sparingly strigose, obtuse or acute leaflets, typically broad and flat, but in the upper leaves often narrow and conduplicate, distant : stems commonly much branched, rarely simple, prostrate or ascending, 2.5—4 dm. long. Illustrations. P1. I, figs. 30—36. Jones, Rev. Astrag., pl. 23 (as var. ineptus). Type-locality. Near Carson City, Nevada, the type collected by Dr. C. L. Anderson in 1864. Distribution. In sandy soil of sagebrush valleys, sometimes by road- sides, along the eastern base of the Sierra Nevada, from Ormsby County, Nevada, to Lake County, Oregon. Exsiccata. Nevapsa. Ormsby Co.: Carson City, Anderson in 1864 (G, type, NY), Anderson No. 157 (G), Jones in 1897 (PO) ; Eagle Valley, C. F. Baker No. 1275 (NY, type of Cystium ormsbyense, F, ND, PO, UC) ; Dayton to Carson, Eastwood No. 14831 (CAS); Empire City, Jones No. 2990 (G, NY, PO, UC). Lyon Co.: Silver City, Ripley & Barneby No. 4068 (PO, RB). Washoe Co.: Lemmon Valley, Kennedy No. 2072 (G, PO, UC) ; base of Peavine Mt., K. Brandegee in 1913 (UC) ; N. of Reno, Train No.7 (UN); Reno, Brandegee in 1888 (UC) ; Clark’s, E. of Reno, Saarni No. 118 (UC) ; Truckee Pass, Watson No. 258 in part (G). Catr- FORNIA. Sierra Co.: Sierra Valley, Bolander in 1872 (G), Lemmon No. 137 (F). Lassen Co.: Chat, Jones, 19 June 1897 (PO) ; Doyle, Ripley & Barneby No. 5678 (CAS, RB). Orecon. Lake Co.: SE. of Paisley, Hitchcock No. 6762 (NY). For a race of such limited geographic range, var. floribundus is unexpectedly variable and difficult to define, being intermediate in many respects between the vars. ineptus, salinus, and albifolius. The typical phase is nearly glabrous throughout, with broad, flat leaflets and rather long-acuminate pods, but this passes into plants with some or all of the leaflets conduplicate, or the whole plant MARCH, 1945 ] PUGILLUS ASTRAGALORUM 95 may be strigulose and the legume subglobose with short-triangular tip. The latter extremes, represented by Baker No. 1275 and Ripley & Barneby No. 4068, is Cystium ormsbyense Rydb., but cannot, I find, be separated by any constant character. Rydberg himself annotated only one other specimen beside the type as C. ormsbyense, viz. Jones No. 2990 (NY), which has glabrous (!) pods, at the same time referring another part of the same col- lection to C. floribundum: on the other hand, he determined the Bolander gathering from Sierra Valley, which has strigulose pods, as C. salinum, but the criterion of pubescence, even when strictly applied, is valueless in this case. I am inclined to regard var. floribundus as a robust, lowland derivative of the alpine var. ineptus, which it resembles in its short leaves and peduncles, in this respect finding an interesting paral- lel in var. albifolius: while sometimes agreeing with the latter in general habit, it may be separated on the character of the pubes- cence and by the straight or upturned, never declined, beak of the legume. From var. salinus, another close relative, it differs in the dense racemes, short subsessile leaves, and elongate internodes. 9. var. ALBIFOLIUS Jones Astragalus lentiginosus Dougl. var. albifolins Jones, Rev. Astrag. 124 (1923) ; Munz, Man. S. Calif. 270 (1935) ; Jeps., Fl. Calif. 2: 356 (1936). Cystium albifolium (Jones) Rydb., N. Amer. Fl. 24: 413 (1929). Astragalus albifolius (Jones) Abrams, Ill. Fl. Pac. St. 2: 598 (1944), non A. albifolius Freyn & Sint. (1893). Pods ovoid to subglobose with a short triangular declined beak, mem- branous-papery, strigose, 1—1.5 cm. long, in dense cylindric clusters: racemes densely 15—20-flowered, 1—2 cm. long, on peduncles about equal- ing the subtending leaf : calyx strigose, the tube 3 mm., the teeth 1.5—2 mm. long: petals white or purplish (rarely bright purple), the banner 8—9 mm., the wings 7--8 mm., the keel about 6 mm. long: leaves widely spreading at least the uppermost subsessile, 3—9 cm. long, with 7—8 pairs of narrowly oblong, oblanceolate or oval, discrete leaflets 5—15 mm. long, mostly con- duplicate, canescent on both faces (more strongly so above) with closely appressed, short hairs: stems 3—10 dm. long, prostrate, flexuous, branched or simple. Illustrations. P1. 1, figs. 25—29. Jones, Rev. Astrag., pl. 23. Abrams, Ill. Fl., fig. 2874. Type-locality. Lone Pine, Owens Valley, California, the type collected by M. E. Jones. 96 LEAFLETS OF WESTERN BOTANY __ [VOL. IV, NO. 5 Distribution. Alkaline flats and saline meadows, on the deserts at the — foot of the Sierra Nevada, from Inyo County south to Los Angeles County, California. Exsiccata. Cairornta. Sine loc., Hasse (NY), Pringle No. 14 in 1882 (G), Lemmon No. 69 in 1875 (NY). Inyo Co.: Lone Pine, Jones (PO, type, G, NY); Big Pine, Purpus No. 5760 (PO, UC) ; Kern Co.: Muroc, Hoffmann in 1929 (SB). Los Angeles Co.: Lancaster, K. Brande- gee in 1905 (PO, UC), Elmer No. 3627 (NY, PO), Davidson No. 3521 (LA) and in 1893 (LA, ND), Feudge No. 1133 (PO), Ripley & Barneby No. 5853 (CAS, RB), No. 5859 (CAS, RB). Jones called his var. albifolius a “remarkable variation,” but when compared with his material of var. floribundus from Carson City, Nevada, it will be seen nearly to approach that puzzling race, and it is probably another derivative of the alpine var. ineptus modified by similar ecological stimuli. The striking character of the declined beak to the pod, unique as far as I have observed in A. lentiginosus, was overlooked by Jones and Rydberg. In flower the plant is readily recognized by the long, prostrate, flexuous branches, relatively short, spreading, subsessile, canescent leaves, and by the condensed, shortly pedunculate racemes. The habital similarity between this variety and var. sesquimetralts, also a halo- phyte, is striking, though the flowers and legumes are very dif- ferent. : 10. var. sEMoTUS Jepson Astragalus lentiginosus Dougl. var. semotus Jeps., Fl. Calif. 2: 357 (1936). Pods ovoid, membranous, strigose, about 15 mm. long, in rather loose clusters much surpassed by the leaves: racemes 1—2 cm. long, rather lax: calyx white-strigose, the tube 3.5—4 mm., the teeth 1.5—2 (or perhaps more) long: petals whitish, the banner 11—12 mm., the wings about 9.5 mm., the keel 8 mm. long: leaves 5—9 cm. long, erect, with 6—10 (or 13, ex char.) pairs of remote, conduplicate, oblanceolate to obovate leaflets 3—9 mm. long, strigose-canescent or green and strigulose below, glabrate above, ciliate: stems procumbent, cespitose, about 1 dm. long. Illustration. Pl. II, figs. 5—6. Type-locality. Campito Mountain, White Mts., Inyo County, Cali- fornia, the type, W. L. Jepson No. 7280, not seen. Distribution. White Mts. of Nevada and California to the Inyo Mts., and (doubtfully) the east face of the Sierra Nevada in Inyo County, California. Exsiccata. Nervapa. Esmeralda Co.: Chiatovitch Creek, Duran No. 2776 (F, G, Br, NY, PO, UC). Catirornra. Inyo Co.: Silver Canyon near Laws, K. Brandegee in 1913 (PO), Heller in 1906 (Br) ; Waucoba Peak, Alexander & Kellogg No. 2511 (UC); Rock Creek Lake Basin, Peirson No. 9155 (RB). MARCH, 1945] PUGILLUS ASTRAGALORUM 97 I have not seen the type of var. semotus and my judgment of the race has been formed mainly by study of Duran No. 2776, cited in the original description. It appears to differ from var. ineptus, to which it is very closely related, in the much longer, filiform leaf-rachis, in the remote leaflets, and in the more closely appressed pubescence of straighter hairs. Jepson described the calyx-teeth as “1% to 34 as long as the tube” : in the Duran speci- mens they are half as long, but in Mrs. Brandegee’s plant, other- wise similar, they are considerably shorter. The same is true of Peirson’s gathering from the Sierra, which in addition has but - six or seven pairs of leaflets and should perhaps with greater justice be referred to the next. 11. var. 1nEPTUS (Gray) Jones Astragalus lentiginosus Doug]. var. ineptus (Gray) Jones, Rev. Astrag. 124 (1923), pro parte; Jeps., Fl. Calif. 2: 357 (1936), pro max. parte. A. ineptus Gray, Proc. Amer. Acad. 6:525 (1865) ; Hoover, Leafl. W. Bot. 3: 256 (1943) ; Abrams, Ill. Fl. Pac. St. 2: 599 (1944), pro parte. A. lentiginosus Wats., Bot. King Explor. 66 (1871), pro parte, quoad syn. ; Gray, Bot. Calif. 1: 147 (1880), quoad formam; Smiley, Univ. Calif. Pub. Bot. 9: 257 (1921), pro max. parte; Hall & Hall, Yosemite FI. 145 (1912), non Doug!. A. lentiginosus var. Fremontit Wats., Bibl. Ind. 196 (1878), pro parte. quoad syn. A. Fremontu Sheld., Minn. Bot. Stud. 1: 169 (1894), pro parte, quoad syn. A. lentiginosus var. floribundus Jones, Contrib. W. Bot. 8:3 (1898), pro parte, non Gray. Phaca inepta (Gray) Rydb., Mem. N. Y. Bot. Gard. 1: 246 (1900), quoad nomen. Cystiun ineptum (Gray) Rydb., Bull. Torr. Bot. Club 32: 659 (1906), quoad nomen. C. ineptum (Gray) Rydb., op. cit. 40:50 (1913), quoad nomen; N. Amer. Fl. 24: 413 (1929), sensu stricto. Pods membranous, ovoid, nearly straight or with slightly ascending beak, 10—17 mm. long, strigose (exceptionally glabrous), sometimes mottled : racemes rather loosely 10—15-flowered, 1—2 cm. long, the pedicels filiform: calyx loosely pubescent with curved hairs, the tube 3—4 mm., the acute teeth 1.5—2 mm. long: petals whitish or ochroleucous, or the wings tinged with pink, the banner 11 mm., the wings about 10 mm., the keel 8 mm. long: leaves 3—5 cm. long, on slender petioles, crowded, with 8—10 pairs of obovate-oblanceolate, conduplicate, and contiguous (or at least closely approximate ) leaflets 3—9 mm. long, 2—4 mm. wide, pubescent with spread- ing or curved hairs below, strongly ciliate, commonly glabrate above at least along the midrib: stems numerous, densely cespitose, prostrate, 1—2 dm. long. 98 LEAFLETS OF WESTERN BOTANY _ [VOL. IV, NO. 5 Illustrations. Pl. Il, figs. 1—4. Abrams, Ill. Fl., fig. 2879. Type-locality. Dry rocky mountain near Sonora Pass, the type col- lected by Brewer in 1863. Distribution. Exposed gravel slides or sandy banks along the easterly crest of the Sierra Nevada, between 7000 and 11,400 feet altitude, from Alpine County south to Inyo County, California. Exsiccata. CALIFORNIA. Alpine Co.: Sonora Peak, Wiggins No. 9312 (PO, UC); White Mt., Gifford No. 750 (UC). Mono Co.: near Sonora Pass, Brewer No. 1878 (G, type, UC); Sonora Pass, Wiggins No. 9255 (PO, UC), Peirson No. 11648 (PO), Eastwood & Howell No. 7533 (CAS), Ripley & Barneby No. 4077 (CAS, RB); Mono Pass, Congdon in 1895 (UC); Virginia Lakes Basin, Peirson No. 11169 (RB) ; Devil’s Gate, Ripley & Barneby No. 5810 (CAS, RB) ; Long Valley, J. T. Howell No. 14399 (CAS) ; Hilton Creek, Peirson No. 10743 (RB, UC) ; N. shore of Mono Lake, Peirson No. 10758 (RB, UC). Tulare Co.: Bonita Meadow, Hall & Babcock No. 5201 (UC). Inyo Co.: Bishop Creek, Dearing & Dearing No. 562 (SB); Inconsolable Range, Peirson No. 8544 (UC). Astragalus ineptus, as to the plant itself, has been much mis- understood by modern botanists, while the name has been widely and improperly applied to several diverse races of A. lentiginosus. First described from flowering material collected by Brewer, it was reduced by Watson first to A. lentiginosus (1871) and then to its var. Fremontii (1878). The earlier reduction was followed by Gray (1880), who regarded it as “‘a narrow-leaved and pubes- cent form,’ and more recently by Hall (1912) and Smiley (1921), while the second was copied by Sheldon into his Prelimi- nary List (1894). Jones, having at first (1898) referred it to var. floribundus Gray, later (1923) published the combination ' A. lentiginosus var. ineptus and reduced var. floribundus to that, but his concept seems always to have been floribundus in great part. Rydberg took up the name and combined it with Phaca (1900), then twice with Cystium (1905 and 1913), but applied it to A. salinus Howell (for which he later substituted Cystiwm heliophilum), and he was followed by Aven Nelson who treated as A. ineptus two different races from Montana and Wyoming (var. salinus and var. platyphyllidius). Not until 1929 did Ryd- berg restrict his C. ineptum to the plant from the Sierra Nevada. Jepson (1936), if one may judge from the cited distribution and altitude of occurrence, seems to have included in his var. imeptus some material which I regard as var. floribundus. The var. ineptus in exactly typical form is found at great alti- tudes, and it is only in material gathered between 9,000 and 11,400 MARCH, 1945] PUGILLUS ASTRAGALORUM 99 feet that the spreading pubescence and short peduncles are ob- served to be constant. As the plants descend the eastern face of the Sierra, the hairs become more appressed (Peirson No. 10743), or the racemes elongate and surpassing the leaf (Peirson No. 10758), or the stems elongate and the pods glabrous (Howell No. 14399), and thus appear to intergrade with var. floribundus. As I have seen it on exposed gravel ridges above Sonora Pass it forms a depressed, cespitose plant about a foot in diameter and an inch or two high, the numerous racemes of ochroleucous blossoms barely emerging from the crowded, ascending leaves. 12. var. SIERRZ Jones Astragalus lentiginosus Doug]. var. sierre Jones, Rev. Astrag. 124 (1923) ; Munz, Man. S. Calif. 269 (1936), pro max. parte; Jeps., Fl. Calif. 22357: (1936); Cystium sierre (Jones) Rydb., N. Amer. Fl. 24: 412 (1929). Astragalus sierre (Jones) Tidestr., Proc. Biol. Soc. Wash. 50:21 (1937). A. ineptus Abrams, Ill. Fl. Pac. St. 2: 599 (1944), pro parte, non Gray. Pods ovoid with a nearly straight triangular beak, papery-membranous, mottled, strigose, 15—20 mm. long : racemes 8—15-flowered, 1—2 cm. long: calyx strigose, the tube 4 mm., the rather obtuse, deltoid teeth 1 mm. long: petals whitish or the upper ones tinged with pink, the banner 11 mm., the wings about 10 mm., the keel 8 mm. long : leaves 2—4 cm. long, with slender petioles and 7—9 pairs of obovate, conduplicate, ciliate leaflets more or less strigose-canescent to glabrate below, glabrous above: stems numerous, matted, prostrate, very leafy, 1—3 dm. long. Illustrations. Pl. II, figs. 10—12. Jones, Rev. Astrag. pl. 23. Type-locality. In gravelly soil, Bear Valley, California, the type col- lected by Jones. Distribution. In stony meadows or open pinewoods of the Arid Tran- sition Zone, in the San Bernardino Mountains of southern California, alt. about 6000 feet. Exsiccata. CAatirornta. San Bernardino Co.: sine loc., Parry & Lem- mon No. 91 m 1876 (G, NY); Bear Valley, Parish No. 2350 (F, NY), Parish & Parish No. 1496 (F, G), No. 1867 (F, ND, NY), Abrams & McGregor No. 735 (NY), Jones in 1900 (PO, type), Davidson No. 2198 (LA); Bear Lake, Jones in 1926 (NY, PO), in 1934 (PO, UC), Ewan No. 1418 (SB), Munz No. 10468 (PO) ; Big Meadows, Feudge No. 1132 (PO) ; Baldwin Lake, Peirson No, 4558 (PO), Ripley & Barneby in 1936, (Kew Gardens) ; Holcomb Valley, Pierce in 1922 (PO); Sugarloaf Peak, Holmes in 1933 (LA). The var. sierre is very closely related to the preceding, which it exactly resembles in habit, and to which Abrams has lately re- ferred it, but the pubescence is commonly less copious and more closely appressed, while the calyx-teeth are obtuse and decidedly 100 LEAFLETS OF WESTERN BOTANY [ VOL. IV, NO. 5 shorter. Munz (1935, |. c.) admitted to the var. sterre a plant of very similar aspect from the mountains of Ventura Co., Califor- nia, here treated as var. tehatchapiensis: it is, no doubt, another close relative of the ineptus-sierre group, but the flowers are larger and more brightly colored, the pods are decidedly coria- ceous in texture with a more deeply intruded ventral suture, and the leaflets are not conduplicate. In the San Gabriel Mts., adjacent to the west, var. sierre passes into 13. var. ANTONIUS Barneby Astragalus lentiginosus Dougl. var. antonius Barneby, var. nov., preecedenti manifeste arcteque affinis, sed habitu sepissime laxiori, inter- nodiis foliisque duplo longioribus, foliolis semotis plerumque expansis et pube sericea appressiuscula utrinque strigoso-canescentibus, racemis longius pedunculatis necnon floribus paulo minoribus saturatius purpureo-tinctis absimilis —A. lentiginosus var. Fremontii Jones ex I. Jtn., Plant World 22: 108 (1919); Jeps., Fl. Calif. 2: 356 (1936), pro parte, ex num. cit. Pods in shap: and texture exactly as in the last, more loosely racemose, 14—24 mm. long: racemes 10—15-flowered, 1—3 cm. long, on rather long, straight peduncles, surpassing the leaves: calyx canescent with ascending hairs which may be all white or with an admixture of black, the tube 3—3,5 mm., the subulate teeth 0.5—0.9 mm. long: petals distinctly purple, the narrow banner 9—9.5 mm., the wings 8—8.5 mm. long, the obtuse, nearly straight keel 7—7.5 mm. long: leaves ascending, 4—8 cm. long, with 6—9 pairs of flat or very rarely conduplicate, elliptic or obovate, obtuse or emarginate leaflets 411 mm. long, canescent on both faces with straight, rather loosely appressed, silvery hairs: stems 1.5—3 dm. long, usually with well-developed internodes. Illustration. P1. II, figs. 7—9. Type. Swartout Valley, San Bernardino Co., California, J. B. Feudge No. 990 (Pomona College Herbarium No. 145409). Distribution. Valleys and slopes of Mount San Antonio (Old Baldy), east end of San Gabriel Mountains, in San Bernardino and Los Be counties, California, between'5000 and 8500 feet. Exsiccata. CALIFoRNIA. San Bernardino and Los Angeles Cos.: Swar- tout Valley, Feudge No. 990 (PO, type) ; Swartout Canyon, Hall No. 1485 (PO, UC); Blue Ridge, Swartout Valley, Munz No. 7681 (PO); Lone Pine Creek, Munz No. 4598 (PO); Prairie Fork of San Gabriel River, I. Johnston No. 1655 (PO, UC) ; Old Baldy Mt., Parish No. 1918 (F, ND, UC) ; Pinyon Ridge, Margaret Stason in 1927 (UC). Imperfect specimens of var. antonius were determined by Jones as A. lentiginosus var. Fremontii, which they superficially resemble, but the real affinity, as indicated by annotations of Munz in herb., is doubtless with the preceding. Material from the upper slopes of the Blue Ridge (Munz No. 7681) at 8450 feet altitude MARCH, 1945] PUGILLUS ASTRAGALORUM 101 is almost as condensed as var. sierr@, with short leaves and con- duplicate leaflets, but can still be separated on the character of the pubescence. The prevailing habit is much more lax, however, and the leaflets are commonly flat and remote. In this form it is easily confounded with the pubescent phase of var. Fremontit, but the racemes are fewer-flowered and shorter, the calyx-teeth and flower relatively smaller, and the habitat is quite different. This is another of the forms by which the varieties of A. lenti- ginosus are so closely interlocked as to make specific or even sub- specific segregation of major groups within the complex quite impracticable. 14. var. CHARLESTONENSIS (Clokey) Barneby Astragalus lentiginosus Doug]. var. charlestonensis (Clokey) Barneby, comb. nov. A. kernensis Jeps. subsp. charlestonensis Clokey, Madrofio 6: 218 (1942). Pods broadly ellipsoid to subglobose, the body 9—12 mm., the filiform beak 1—2 mm. long, the valves papery, strigose, mottled: racemes 2—6- fiowered, about 1 cm. long in fruit, on slender peduncles 2.5—5 cm. long: calyx strigose with white or some black hairs, the tube 3—4 mm., the teeth about 1 mm. long: petals narrow, white or ochroleucous, the banner 10—12 mm., the wings 9—10 mm., the purple-tipped keel 7—8 mm. long: leaves 3.5—5 cm. long, shortly petiolate, with 7—10 pairs of narrowly obovate to elliptic, obtuse, rather distant and rarely conduplicate leaflets 3—7 mm. long, strigose-hirsutulous below, glabrous above: stems cespitose, prostrate, slender, sparingly strigose, 5—15 cm. long from a stout perennial tap-root. Illustrations. Pi. 11, figs. 13—15. Clokey, op. cit., pl. 27, figs. k—o. Type-locality. Charleston Peak, Clark Co., Nevada, the type collected by Ira W. Clokey and Russell Bean. Distribution. Known only from limestone screes of Charleston Peak, at about 10,000 feet altitude. Exsiccata. Nevapa. Clark Co.: Charleston Peak, Clokey & Bean No. 7375 (F, PO, RB, and widely distributed, type-collection ) ; ibid., west slope, Jaeger in 1926 (PO). This variety and the next are closely related and should per- haps be accepted as a species distinct from A. lentiginosus, but the sole diagnostic character lies in the beak of the legume. In all other forms of the species this organ, when clearly defined from the body of the pod, is triangular or deltoid-acuminate in outline and usually somewhat compressed: in var. charlestonensis and var. kernensis, on the other hand, the beak is tubular-filiform and has the appearance of a persistent style. Clokey (1. c., 1942) stated that A. kernensis and its subspecies could be distinguished from A. lentiginosus by the narrow intrusion of the ventral suture 102 LEAFLETS OF WESTERN BOTANY [ VOL. IV, NOSS5 of the legume. An exactly similar intrusion, however, may be observed even in the typical race of A. lentiginosus, and it is espe- cially well marked in var. tehatchapiensis and several others. The flowers, pubescence, and stipules, as well as the general habit, are all closely similar to those of var. ineptus, and in an alliance such as § Diplocystium, where the pod is so extraordinarily polymor- phic and variable, I hesitate to segregate a species on carpological structure alone. 15. var. KERNENSIS (Jeps.) Barneby Astragalus lentiginosus Dougl. var. kernensis (Jeps.) Barneby, comb. nov. A. kernensis Jeps., Man. Calif. 569 (1925) ; Fl. Calif. 2: 357 (1936). A. lentiginosus var. latus, “a form with tiny flowers ... ,” Jones, Rev. Astrag. 125 (1923). Cystium kernense (Jeps.) Rydb., N. Amer. Fl. 24: 413 (1929). Astragalus ineptus Abrams, Ill. Fl. Pac. St. 2: 599 (1944), pro parte, quoad syn. cit., non Gray. Pods globose with a filiform beak, papery-membranous, sparingly strigose-hirsutulous, sometimes mottled, the body 6—8 mm., the beak 1—3 mm. long: racemes 3—6-flowered, about 1 cm. long in fruit, on peduncles shorter than the subtending leaf ; calyx white-strigose, the tube about 4 mm., the teeth 1—1.5 mm. long: petals narrow, whitish or the wings tinged with pink, the banner 9—11 mm., the wings 7—8 mm., the keel 6—7 mm. long: leaves 2—5 cm. long, with 4—7 pairs of obovate, obtuse to emarginate, commonly conduplicate leaflets 3—7 mm. long, pubescent beneath and ciliate with rather coarse hairs, glabrous above: stems very slender, ascending, canescent with loosely appressed hairs, 6—10 cm. long. Illustration. P1. II, figs. 16—18. Type-locality. Volcano Creek, Tulare Co., California, the type, W. L. Jepson No. 4930, not seen. Distribution. Whitney Plateau in the high Sierra Nevada of Tulare County, California, at about 8000 feet altitude. Exsiccata. Catirornta. Tulare Co.: Monatchy Meadows, Purpus No. 1871 (PO, UC) ; Volcano Creek, Hall & Babcock No. 5432 (UC), Peirson No. 2000 (CAS). The var. kernensis, the smallest and most delicate of the varie- ties of A. lentiginosus, is very local, being confined to that area of endemism in the southern Sierra Nevada to which also belong Abromnia alpina Bdg. and Oreonana Clementis (Jones) Jeps. It seems, like the preceding, to have been longer isolated than its near relatives, and furnishes an excellent example of the geo- graphical subspecies of some modern taxonomists. Jones referred the only specimen known to him to var. latus, an incomprehensible MARCH, 1945 ] PUGILLUS ASTRAGALORUM 103 determination which he followed up by reducing A. kernensis to A. lentiginosus sensu stricto (cf. Contrib. W. Bot. 18: 44,— 1933), while it has recently been referred by Abrams to var. ineptus. 16. var. PLATYPHYLLIDIUS (Rydb.) Peck Astragalus lentiginosus Dougl. var. platyphyllidius (Rydb.) Peck, Man. Pl. Oreg. 449 (1940), nomen; Madrofio 6: 135 (1941). A. lentiginosus var. cuspidocarpus Jones, Proc. Calif. Acad. Sci., II, 5: 673 (1895) ; Jeps., Fl. Calif. 2:357 (1936), pro parte, ex syn.; non A. cuspidocarpus Sheld., nec A. missouriensis var. cuspidocarpus (Sheld.) Jones. A. ineptus A. Nels., Erythea 6:54 (1898), ex num. cit., non Gray. A. lentiginosus Jones, Contrib. W. Bot. 8:3 (1898), pro parte, non Dougl. A. lentiginosus var. diphysus Jones, Rev. Astrag. 125 (1923), pro parte, non A. diphysus Gray. Cystium Merrillii Rydb., N. Amer. Fl. 24: 410 (1929). C. platyphyllidium Rydb., 1. c. C. cornutum Rydb., op. cit., 412, quoad typum, exclus. syn. Astragalus Merrillii (Rydb.) Tidestr., Proc. Biol. Soc. Wash. 50:21 (1937). A. araneosus Abrams, Ill. Fl. Pac. St. 2: 598 (1944), non Sheld. Pods highly variable in outline, length and inflation, ovoid or ovoid- acuminate to lanceolate-acuminate or subglobose, 1.5—3 (4) cm. long, usu- ally with a more or less upturned, laterally flattened beak, the valves cori- aceous, strigose or glabrous: racemes 10—12-flowered, the axis 1.5—3 cm. long, on peduncles about half as long as the subtending leaf : calyx strigose with black hairs, the tube 5—8 mm., the narrow acute teeth 2.5—4 mm. long : petals white or the keel purple-tipped, the banner 13—17 mm., the wings 11—14 mm., the keel 10—13 mm. long: leaves petiolate, with 6—9 pairs of broadly elliptic, obovate or suborbicular, obtuse or emarginate leaflets 6—16 mm. long, essentially glabrous or with a few scattered hairs on the margins and midribs : stems 1.5—3 dm. long, rather coarse, stramineous, subglabrous, ascending. Illustrations. P\. II, figs. 39—46. Jones, Rev. Astrag., pl. 24, fig. d (dubie). Abrams, Ill. F1., fig. 2876. Type-locality. Pine Creek, Gilliam Co., Oregon, the type collected by J. B. Leiberg. Distribution. Volcanic mesas and sagebrush plains, often in alkaline soil, from northeastern California through interior Oregon and northern Nevada to southern Idaho, southwest Wyoming, and northern Utah. Exsiccata. CAttrorNta. Lassen Co.: Eagle Lake, Jones in 1897 (PO) ; Hidden Valley near Susanville, Ripley & Barneby No. 5744 (CAS, RB). Modoc Co.: Goose Lake, Austin No. 253 (UC) ; E. side of Warner Mts., Mrs. Manning (NY). Orecon. Gilliam Co.: Pine Creek, Leiberg No. 171 (NY, type, G). Grant Co.: Humphrey Ranch, Henderson No. 5095 (G). Lake Co.: Crump Lake, Ripley & Barneby No. 6049 (CAS, RB). Harney Co.: Silvies River, near mouth of Emigrant Creek, Peck No. 3136 104 LEAFLETS OF WESTERN BOTANY _ [VOL. IV, NO. 5 (G) ; Lower Steens Mt., Cusick No. 1978 (F, G, NY, UC) ; Hoffman’s Ranch, Wildhorse Creek, Henderson No. 8781 (CAS) ; Wagontire, Ripley & Barneby No. 6070 (RB); Alvord Ranch, Jones No. 25406 (PO). Co. undetermined: John Day Valley, Howell No. 265 in 1880 (G); Muddy Station, John Day Valley, Howell in 1885 (NY, type of Cystium cornutum, F, UC) ; “Mudy,” Howell No. 384 (G) ; Anderson Ranch, Howell in 1885 (F). Nevapa. Elko Co.: near Deeth, Heller No. 9250 (F, NY); Cave Creek P. O., Mason No. 4788 (UC); Glencoe, Jones in 1891 (PQ). Washoe Co.: SW. of Vya, Ripley & Barneby No. 6014 (RB). Ivano. Owyhee Co. : House Creek, Nelson & Macbride No. 1781 (G, in part, NY). Bannock Co.: Soda Springs, Payson & Payson No. 1703 (G, NY). Blaine Co.: Arco, Payson & Payson No. 3104(G). Wyominc. Sweetwater Co.: Leucite Hills, Merrill & Wilcox No. 680 (NY, type of Cystium Merrillii, G) ; 35 miles N. of Point of Rocks, A. Nelson No. 7061 (G, in part). Sub- lette Co.: 25 miles W. of Big Piney, Payson & Payson No. 2657 (F, NY, PO, UC). Uinta Co.: Evanston, A. Nelson No. 2978 (NY), No. 4842 (NY, PO), No. 5385(PO). Uran. Summit Co.: Kimballs, Mrs. Clemens in 1908 (F, G). Into var. platyphyllidius, as will be seen from the synonymy, have been combined three Rydbergian Cystia, the reduction of which calls for some discussion. Cystium Merrill was separated from C. platyphyllidium by Rydberg on the dimensions of the calyx and the length of the corolla, the former being supposed to have the tube 7—8 mm. long and 3—4 mm. wide, as opposed to 4—5 mm. long and 3 mm. or less wide, and the corollas were described as “about 1.5 cm. long” and “10—12 mm. long” respectively. It seems that the measurements of the calyx of C. Merrillii were taken exclusively from the type, in which this organ is abnormally inflated: in any case no other specimen among those determined as C. Merrillii by Rydberg bears a calyx of such proportions, as may be seen by comparison with A. Nelson No. 2978, where the tube does not exceed 5 mm. in length, That Rydberg’s description was prepared exclusively from the type-sheet becomes almost certain when such a specimen as Payson & Payson No. 2657, also named by Rydberg, is found to have a legume not 3 cm. long as described for the species, but scarcely half that length. The de- scription of C. platyphyllidium, in contrast to that of C. Merrilli, does not even cover the type-specimen, Leiberg No. 171. No flowers remain, so it is impossible to check absolutely the length of the corolla, but the persistent calyx is about 9 mm. long (the tube 5.5 mm., the teeth nearly 4 mm., as opposed to 4 mm. and 2 mm. in the description), and the banner can scarcely have been } MARCH, 1945] PUGILLUS ASTRAGALORUM 105 less than 15 mm. long, certainly not as little as “about 10 mm.” Thus there seems to be no appreciable difference between the nor- mal flower (in contradistinction to that described from an abnor- mal type) of C. Merrilli and the typical flower (as opposed to that described from diverse elements) of C. platyphyllidium, and the essential difference between the two is reduced to one of comparative inflation in the pod. Rydberg’s real intentions in regard to C. platyphyllidium are not revealed by study of material named by him: they include such dissimilar plants as Leiberg No. 2301 (flowers as described for the species, but pods papery and quite typical var. salinus) and Heller No. 9250 (like the type but with very much larger flowers than described). Cystium cornutum, the third species proposed by Rydberg, is separated (1. c.) quite widely from the others on the criterion of the pubescent or “sparingly strigulose” pod, and this is, indeed, the sole character in which the types of C. platyphyllidiwm and C. cornutum differ. The author himself found difficulty in distin- guishing them. For example, there exists in the Gray Herbarium an interesting collection, Howell No. 384 from “Mudy”’ (i. e. probably Muddy Station in the John Day Valley, Oregon), gath- ered on May 12, 1885. Perhaps Rydberg took this for an isotype of C. cornutum, although that collection was made the following day : in any case he determined the plant as C. cornutum, in spite of the fact that the legume is glabrous! Another example of the practical value of this same character is furnished by Cusick No. 1978, which may bear either glabrous (F) or strigulose (G, ND) pods. It seems entirely mystifying, also, that Rydberg could de- scribe a species as having “pods fully 2.5 cm. long” and corolla “about 15 mm. long,’”’ and simultaneously determine under that name plants with flowers not exceeding 1 cm. and pods 13—15 mm. long. Examples of this are Heller No. 8062 and Austin & Bruce No. 2210 (var. carinatus), or Peck No. 9949 (var. typicus). Confusion about this race goes back further than Rydberg, the types of both Cystium platyphyllidium and C. cornutum having been discussed and determined by earlier authors. Thomas How- ell’s collection from John Day Valley was cited by Sheldon as belonging to his species A. araneosus, and the description of the flower as white, a character derived from this specimen but alien to the type, was incorporated by Rydberg into his account of 106 LEAFLETS OF WESTERN BOTANY _ [VOL. IV, NO. 5 Cystium araneosum in the N. American Flora. At the same time Sheldon identified Leiberg No. 171 as belonging to his A. cuspi- docarpus, a species unrelated to A. lentiginosus, and this gave rise to Jones’s combination A. lentiginosus var. cuspidocarpus which, although repudiated by the author, has since reappeared in Dr. Jepson’s Flora of California. In the Revision of Astragalus, Jones treated var. platyphyllidius as a form of var. diphysus, er- roneously quoting as a synonym A. ammolotus Greene, a name proposed to replace the already superfluous A. elatiocarpus Sheld. Professor Peck has recently recognized C. cornutum and C. pla- typhyllidium as separate varieties of A. lentiginosus, a disposition which I have failed to understand. He also published the combi- nation under which the plant is treated here as “A. lentiginosus var. platyphyllidium,”’ apparently not having realized the adjec- tival nature of the second epithet. While I regard C. platyphyllidium and C. cornutum as exactly Synonymous, the race segregated as C. Merrillu might be con- sidered distinct on the dimensions of the pod, which is as a rule considerably more inflated, ovoid rather than lance-ovoid, and perhaps slightly less firm in texture. Nevertheless the pod of Henderson No. 8781 from Harney County, Oregon, where var. platyphyllidius is not uncommon, is every bit as turgid as in the plant from Wyoming, and I can find no constant criterion corre- lated with distribution. As interpreted here, var. platyphyllidius is easily recognized by its large white flowers, partially black- strigose calyx with long teeth, the broad, thick, glabrous leaflets, and coriaceous pods. These, however, vary in outline between the extremes exhibited by Henderson No. 8781 (ovoid, much in- flated, 3 cm. long and with septum 7 mm. deep), Henderson No. 5095 (lanceolate, to 4 cm. long, with septum only 2 mm. deep), and Payson No. 2657 (subglobose, 13—15 mm. long, the septum about 5 mm. deep), but the texture of the valves is always leathery. A few specimens show signs of intergradation with var. salinus and var. carinatus. A similar form with papery pods may be dis- tinguished as 17. var. TOYABENSIS Barneby Astragalus lentiginosus Dougl. var. toyabensis Barneby, var. nov., precedenti persimilis, et preter legumen membranaceum subdiaphanum, nec coriaceum, atque foliola angustiora, apice minute apiculata, fere omnino congruens.—A. diphysus Wats., Bot. King Explor. 65 (1871), pro parte, MARCH, 1945] PUGILLUS ASTRAGALORUM 107 non aliorum. Cystium platyphyllidium Rydb., N. Amer. Fl. 24: 410 (1929), pro parte. Pods obliquely ovoid, membranous, glabrous, 2 cm. long : racemes dense, about 3 cm. long : calyx black-strigose, the tube 5 mm., the teeth 2.5—4 mm. long : petals white, the banner 13—17 mm., the wings 11—15 mm., the keel 10—12 mm. long: leaflets 10—20 mm. long, 4—8 mm. wide, rounded and minutely apiculate at apex : otherwise as the preceding. Illustration. P1. III, figs. 1—4. Type. Toyabe Mountains, Nevada, Sereno Watson No. 253 (Gray Her- barium). Distribution. Known only from the neighborhood of the type-locality. Exsiccata. Nevapa. Nye (? or Lander) Co.: Toyabe Mts., Watson No. 253 (G, type, NY), Tidestrom No. 10851 (G). This variety differs from var. platyphyllidius chiefly in the membranous texture of the pod, and in the somewhat narrower, apiculate leaflets. The type was named A. diphysus by Watson, and was referred by Rydberg to his Cystium platyphyllidium. Further material is needed to establish the validity of this entity. 18. var. IDRIENSIS Jones Astragalus lentiginosus Dougl. var. idriensis Jones, Contrib. W. Bot. 10:63 (1902) ; Rev. Astrag. 124 (1923). Cystium idriense (Jones) Rydb., N. Amer. Fl. 24: 414 (1929). Astragalus lentiginosus var. cuspidocarpus Jeps., Fl. Calif. 2: 357 (1939), pro parte. A, idriensis (Jones) Abrams, Ill. Fl. Pac. St. 2: 599 (1944), pro parte. Pods obliquely ovoid-lunate, with upturned beak or strongly incurved, the ventral suture prominent, the valves not much inflated, leathery, stri- gose, mottled, 12—14 mm. long: racemes about 10-flowered, 1—2 cm. long, on peduncles about 5 cm. long which conspicuously exceed the leaves : calyx strigose with mixed white and black hairs, the tube 4+—4.5 mm., the obtuse teeth 0.5—0.75 mm. long: petals purple, the banner 12—14 mm., the wings 10—11 mm., the keel 9—10 mm. long: leaves 2—4 cm. long, spreading, the upper subsessile, with 8—9 pairs of flat, emarginate, ciliate leaflets, 5—9 mm. long, glabrous above and sparingly strigose beneath: stems rather slender, prostrate, about 2 dm. long, numerous from a stout woody tap-root. Illustrations. P1. II, figs. 19—23. Jones, Rev. Astrag., pl. 24, and pl. 72 (section incorrect). Abrams, Ill. FI., fig. 2880. Type-locality. New Idria, California, the type collected by Alice East- wood in 1893. Distribution. San Carlos and Santa Lucia ranges in Monterey and San Benito counties, California. Exsiccata. CALIFoRNIA. San Benito Co.: New Idria, Eastwood in 1893 (PO, type, CAS, UC). Monterey Co.: Tassajara Hot Springs, Elmer No. 3288 (PO, fragments). 108 LEAFLETS OF WESTERN BOTANY _ [VOL. IV, NO. 5 This rare variety has recently been reduced by Jepson (1. c.) to his var. cuspidocarpus of interior northern California (cf. var. carinatus), and it must be admitted that the pods, with their acute ventral suture, are undeniably similar. Nonetheless the larger purple flowers, shortly racemose on peduncles much exceeding the leaves, and the longer calyx with proportionately shorter, ob- tuse teeth, suffice to distinguish var. idriensts. It is much more - closely related to var. tehatchapiensis, from which it differs essen- tially only in the long peduncles and shorter pod. Indeed it would seem from the range cited by Abrams (1. c.) for A. idriensis that he included there a part of the material referred below to var. tehatchapiensts. Rydberg’s statement that the type locality was not given in the original publication is incorrect; it is clearly stated in the opening sentence of the description. 19. var. TEHATCHAPIENSIS (Rydb.) Barneby Astragalus lentiginosus Dougl. var. tehatchapiensis (Rydb.) Barneby, comb. nov. Cystium tehatchapiense Rydb., N. Amer. Fl. 24: 414 (1929). Astragalus Fremontii Heller, Muhlenb. 2: 84 (1905), e num. cit., non Gray. A. lentiginosus var. nigricalycis Munz, Man. S. Calif. 270 (1935), pro parte, e syn. cit.; Jeps., Fl. Calif. 2: 357 (1936), pro parte, non Jones. A. lentiginosus var. sierre Munz, op. cit. 269, pro parte, non Jones. A. tehachapiensis (Rydb.) Tidestr., Proc. Biol. Soc. Wash. 50:21 (1937) ; Abrams, Ill. Fl. Pac. St. 2: 598 (1944). A. idriensis Abrams, op. cit. 599, pre parte, non A. lentiginosus var. idriensis Jones, sens. strict. Pods obliquely ovoid-acuminate, incurved or with ascending beak, leathery, strigose, mottled, 2—2.5 cm. long: racemes 10—15-flowered, sub- capitate or in fruit 1—2 cm. long, on pedumcles about equaling the leaves: calyx strigose with all black or some white hairs, the tube 4—5 mm., the obtuse, subulate to triangular teeth 0.75—1 mm. long: petals purple or the wings pale, the banner 12—15 mm., the wings 10—14 mm., the keel 9—11 min. long: leaves 4—5 cm. long, on short petioles, with 8—10 pairs of flat, obovate-emarginate to obcordate leaflets 5—10 mm. long, glabrate on either face or the margins and midribs commonly strigose-ciliate with scattered hairs: stems numerous, 1.5—2 dm. long, subglabrous to strigulose, leafy, prostrate from a stout perennial tap-root. Illustrations. Pi. II, figs. 24-29. Abrams, Ill. Fl. fig. 2873. Type-locality. Tehatchapi (sic), Kern County, California, the type collected by A. A. Heller in 1905. Distribution. Meadows and brushy slopes in the mountains about the head of the San Joaquin Valley, California. MARCH, 1945] PUGILLUS ASTRAGALORUM 109 Exsiccata. CattrorniA. Kern Co.: Tehachapi, Heller No. 7833 (NY, type, F, G, UC), Eastwood No. 3247 (CAS, NY in part), K. Brandegee in 1905 (UC), Jones in 1903 (PO, UC) ; N. side of Tehachapi Pass, Munz No. 8917 (PO); Tehachapi Valley, Davy No. 2184 (UC). Ventura Co.: Lockwood Valley, Hoffmann No. 6982 (SB), No. 10864 (SB); Cuddy Valley, Hall No. 6317 (UC) ; Mt. Pinos, Hall No. 6460 (UC) ; Seymour Creek, Mt. Pinos, Munz No. 6968 (NY, PO, UC); Chuchupate Camp, Mt. Frazier, Epling & Wheeler No. 1818 (PO, UC). Santa Barbara Co.: San Rafael Mts., H. C. Ford in 1889 (G). This variety is rather unstable in the size of the flower, and perhaps contains two minor races. The typical element, restricted to the Tehachapi region, has the calyx-tube 5 mm. long with very obtuse teeth, all the petals purple, and the keel 1O—11 mm. long. In the material from about Mt. Frazier, on the other hand, the calyx-tube rarely exceeds 4 mm. in length, with rather more acute teeth and fewer black hairs, the petals are paler, and the keel varies between 9and 10mm. The latter form much resembles var. sierre Jones, and was referred there by Munz (1. c.), but it differs in the flat leaflets, longer petals, and the coriaceous texture of the larger and more arcuate pod. The same thing was apparently in- cluded by Abrams in his A. idriensis (q. v.), another close rela- tive. The plants from Tehachapi were originally determined by Heller (1. c.) and Jones (in herb.) as var. Fremontu, but since Rydberg’s segregation of Cystium tehatchapiense the name at least has been reduced by Munz (as an intergrade) and by Jepson (altogether) to var. migricalycis Jones. It can be separated from the former by its larger flower and leathery legume, from the latter by its merely strigose and never villosulous calyx, and from both by the subcapitate racemes and prostrate stems. It seems to be much more closely related to var. idriensis and var. sierre, both montane races of cespitose habit, but the length of the peduncle and texture of the pod will serve to distinguish it. Rydberg’s spelling of the trivial “tehatchapiense’”’ was evidently intentional, and must be retained in the new combination, although both Jep- son and Tidestrom have corrected it to “tehachapiense.” It is probable that the pod-section figured by Jones as var. C oultert, fig. a, on Plate 25 of the Revision of Astragalus, in reality illus- trates this variety. From the mountains of Ventura County we have also three interesting collections which apparently represent hybridization between var. nigricalycis and var. tehatchapiensis. These plants 110 LEAFLETS OF WESTERN BOTANY _ [VOL. IV, NO. 5 agree with the latter in the dimensions of the flower and prostrate habit, but the petals are ochroleucous, the leaves and stems longer, the pubescence much more copious throughout, while the pod is less coriaceous and canescent. To this form belongs Abrams & McGregor No. 204 (G, NY) from the Frazier Borax Mine, which was mistaken by Rydberg for Cystium vulpinum (var. macrolobus (Rydb.) Barneby) and led him to report that race from California. Another gathering is Hall No. 6455 (UC) from North Fork, Mt. Pinos, while the third is a single individual discovered by the writer on the south side of Mt. Pinos at about 6000 feet, Ripley & Barneby No. 4086 (PO, RB). - 20. var. LaTuS (Jones) Jones Astragalus lentiginosus Dougl. var. latus (Jones) Jones, Rev. Astrag. 125 (1923), exclus. pl. Purpusiana. A. diphysus var. latus Jones, Zoe 3: 287 (1893) ; Proc. Calif. Acad. Sci. II, 5:675 (1895), exclus. syn. A. latus (Jones) Jones, Zoe 4: 272 (1894) ; Sheld., Minn. Bot. Stud. 1: 169 (1894). A. araneosus Tidestr., Contrib. U. S. Nat. Herb. 25: 325 (1925), pro parte, ex syn., non Sheld. Cystium latum (Jones) Rydb., N. Amer. Fl. 24: 415 (1929). Pods broadly ovoid to subglobose, with a short-triangular straight or moderately upturned beak, coriaceous, glabrous, 1—2.5 cm. long, with septum produced into the very apex and thus completely bilocular : racemes 5—10-flowered, I1—2 cm. long, on slender, arcuate peduncles a half shorter than the leaves: calyx sparingly black-strigose, the tube 5.5—7.5 mm., the linear-lanceolate teeth 2—3.5 mm. long: petals purple, the banner 15—18 mm., the wings 11.5—16 mm., the keel 10.5—13.5 mm. long: leaves 5—9 cm. long, with 6—8 pairs of broadly elliptic, obovate or suborbicular, obtuse or _emarginate, flat leaflets, 6—11 mm. long, glabrate and sparingly ciliate: stems prostrate, 5—10 cm. long, from a slender, branched caudex and strong perennial root. Illustrations. Pl. Il, figs. 30-34. Jones, Rev. Astrag. pl. 24. Type-locality. Schell Creek Range (later precised as Aurum), Nevada, the type collected by Jones in 1891. Distribution. Mountains of White Pine County, Nevada. Exsiccata. Nevapa. White Pine Co.: Aurum, Jones in 1891 (PO, type), in May, 1893 (F, NY, PO, UC), and July, 1893 (G, NY, PO); Schellbourne, Jones in 1891 (PQ); Ward Mt., Egan Range, about 8000 ft., Ripley & Barneby No. 4020 (CAS, RB). With its short stems, shortly beaked pods, and large purple flowers, var. latus is a distinct and conspicuous plant. Jones first described it as a variety of A. diphysus Gray, but after renewed MARCH, 1945 | PUGILLUS ASTRAGALORUM III study in the field raised it to specific rank, only to return it first to its original status and finally to varietal position under A. lentiginosus. At one period he included (1895) in this variety A. salinus Howell, and in the Revision (1923) mysteriously re- ferred here a collection of var. kernensis (Jeps.) Barneby. In his very detailed and complete description of the plant in 1894 he erroneously described the apex of the pod as ‘‘not two-celled,” but I have found that the septum is always produced into the beak of the pod, and it is so portrayed in Jones’s figure. The statement (Jones, 1923) that this variety is “rather common in eastern Nevada” is not supported by the material available in herbaria. 21. var. CHARTACEUS Jones Astragalus lentiginosus Dougl. var. chartaceus Jones, Proc. Calif. Acad. Sci. II, 5: 673 (1895). A. lentiginosus var. diphysus, “a form intergrading with the var. latus,’ Jones, Rev. Astrag. 125 (1923). Cystium salinum Rydb., N. Amer. Fl. 24: 411 (1929), pro parte, e syn., non A. salinus Howell. Pods obliquely ovoid to lance-ovoid, long-acuminate, 2—2.5 cm. long, coriaceous, glabrous, not mottled, the apex unilocular: racemes subcapi- tate, 1—2 cm. long in fruit, on peduncles about half as long as the subtending leaf: calyx sparingly strigose, the tube 6.5 mm., the teeth about 2 mm. long: petals apparently purple, the banner 16 mm., the wings 14.5 mm., the keel 13 mm. long: leaves 4—6 cm. long, with 5—6 pairs of glabrous or minutely puberulent leaflets 6—12 mm. long: stems slender, flexuous, pros- trate, about 1 dm. long. Illustrations. Pi. If, fig. 35. Jones, Rev. Astrag. pl. 24 (as var. di- physus, fig. 2a). Type-locality. Ephraim, Utah, the type collected by Jones. Distribution. Known only from the type-collection. Exsiccata. Utau. Sanpete Co.: Ephraim, Jones No. 5627m (PO, type). The var. chartaceus 1s something of a puzzle, for it is known only from a single very mature specimen without fresh flowers. | have presumed, nevertheless, from traces of pigment in the fallen petals that they were originally purple. From var. latus, which it resembles closely in habit, var. chartaceus differs in the long-acuminate, not abruptly apiculate, legume, and the some- what shorter leaves with fewer leaflets, but with further know]- edge these may well prove not to be diagnostic. The habitat of var. chartaceus, described by Jones as “clay land in the Sanpete Valley about 6000 feet,” is unlike that of var. latus, a native of 112 LEAFLETS OF WESTERN BOTANY [ VOL. IV, NO. 5 sweet soil in mountain valleys. It seems that Rydberg never studied authentic material of the variety, for he could hardly have confused it with A. salinus Howell, a plant with small white flowers and papery legume. 22. var. ARANEOSUS (Sheld.) Barneby Astragalus lentiginosus Doug]. var. araneosus (Sheld.) Barneby, comb. nov. A. araneosus Sheld., Minn. Bot. Stud. 1: 170 (1894), sensu strict. A. palans var. araneosus (Sheld.) Jones, Proc. Calif. Acad. Sci. II, 5: 675 (1895), quoad nomen, exclus. num. cit. Cystium araneosum (Sheld.) Rydb., Bull. Torr. Bot. Cl. 40:50 (1913) ; Fl. Rocky Mts. 492 (1917); N. Amer. FI. 24: 410 (1929). Astragalus lentiginosus var. palans Jones, Rev. Astrag. 126 (1923), pro parte, quoad formam cit., non alibi. A. lentiginosus var. diphysus Jones, |. c., pro parte. Cystium diphysum Rydb., op. cit. (1929), p. 414, pro parte, e loc. cit., non sensu strict. Pods long-acuminate, narrowly ovoid or if broadly so then with an elongate unilocular beak, often much incurved, broadest below the middle, (1.5) 2—3 (4) cm. long, the valves coriaceous, glabrous or very rarely strigulose in youth, mottled: racemes 10—18-flowered, 2.5—3.5 cm. long in fruit, about equaling the leaf: calyx commonly black-strigose, or at least with some black hairs, the tube 4.5—6 mm., the teeth 1—3 mm. long: petals purple, the banner 15—18 mm., the wings 12—14.5 mm., the keel 11.5—13 mm. long: leaves 6—11 cm. long, with 7—11 pairs of obovate, elliptic to oblanceolate, obtuse or emarginate leaflets 7—18 mm. long, glabrous above, canescent-strigose, strigulose or rarely glabrate below: stems commonly strigose, 1.5—3.5 dm. long, ascending from a stout perennial root. Illustrations. P\. III, figs. 11—14. Jones, Rev. Astrag., pl. 24 (as var. diphysus, figs. a, b). Type-locality. Frisco, Utah, the type collected by Jones in 1880. Distribution. Sagebrush valleys and foothills, often in alkaline soil, from southern Utah northward along the Sevier to Great Salt Lake, and westward into central and northern Nevada. Exsiccata. Utan. Sine loc.: Ward in 1875 (G), Siler in 1877 (G). Iron Co.: Cedar City, Eastwood & Howell No. 9331 (CAS); S. of Cedar City, Blood No. 4412 (F, G, UC), Maguire & Blood No. 4411 (F, PO, UC), Ripley & Barneby No. 4367 (RB) ; Parowan, Eastwood & Howell No. 9340 (CAS, RB). Beaver Co.: Frisco, Jones No. 1807 (F, G, NY, PO, type collection of A. araneosus) ; Milford, Jones in 1880 (NY, PO), Rydberg & Carlton No. 6267 (NY) ; Beaver, Vansell in 1940 (CAS, RB). Piute Co.: Kingston, Jones No. 5322d (PO) ; Marysvale, Jones No. 5338m (NY, PO); Sevier River Canyon, Eastwood & Howell No. 622 (CAS, NY, PO); Bryce Junction to Manti, Nelson & Nelson No. 2915 (NY, UC); N. of Bryce Canyon cutoff, C. L. Hitchcock No. 2957 (F). Sevier Co.: Glenwood, Ripley & Barneby No. 4761 (RB) ; Fish Lake Mt., Cottam MARCH, 1945] PUGILLUS ASTRAGALORUM 113 No. 4518 (PO) ; Fish Lake Valley, Purpus No. 5860 (PO). Wasatch Co.: Midway, Eastwood & Howell No. 519 (CAS). Salt Lake Co.: Salt Lake, Wooton in 1906 (PO). Tooele Co.: Desert Mt., E. of Detroit, Jones in 1891 (PO, UC). Co. undetermined: Fort Hamilton, Jones No. 5209 (F, PO). Nevapa. Elko Co.: Pilot Peak near Montello, Ripley & Barneby No. 4616 (RB); N. of Wendover, Train No. 3810 (UN); Wendover, Eastwood & Howell No. 358 (CAS). White Pine Co.: Baker, Ripley & Barneby No. 3575 (CAS, RB); Ely, Ripley & Barneby No. 3541 (CAS, RB) ; Shoshone, Ripley & Barneby No. 3535 (RB) ; Aurum, Jones in 1893 (PO); Centerville, Jones in 1893 (PO). Eureka Co.: W. of Eureka, Ripley & Barneby No. 6234 (CAS, RB). Lander Co.: Kingston Canyon, Goodner & Henning No. 229 (UN); Austin, Jones in 1882 (PO), Ripley & Barneby No. 4540 (RB) ; SE. of Austin, Vansell in 1942 (CAS). Nye Co.: Pancake Range, Ripley & Barneby No. 3652 (RB). As more fully stated in discussion of the next, var. araneosus is somewhat doubtfully distinct from. var. diphysus, but as a rule it can be distinguished by the characters given in the key. Although occasional specimens occur with glabrous herbage (e. g., those from Austin, Nevada), or the pod shortly beaked (Ripley & Barneby No. 4761), or the calyx entirely white- strigose, the other criteria are in each case valid. The tendency toward curvature of the legume is especially marked in part of Jones’s gathering from Detroit, Utah, but the upper suture is nearly always concave throughout. Only in Jones’s specimens from Aurum do the pods exceed 3 cm. in length. Astragalus araneosus was described from a mixture, the cotype being a specimen of var. platyphyllidius which later formed the basis of Cystium cornutum Rydb., and Sheldon’s statement that the flowers were white was derived from the latter element. Rydberg (1929, 1. c.) repeated this error, and placed his C. ara- neosum far from its nearest allies: apparently he never realized its true identity, for he referred most of the available material to C. diphysuam. Jones (passim) regarded the extreme and typical forms as intergrades between vars. diphysus and palans, and the material in his herbarium is determined under either name. 23. var. pIPHysSUS (Gray) Jones Astragalus lentiginosus Dougl. var. diphysus (Gray) Jones, Proc. Calif. Acad. Sci. II, 5: 673 (1895), pro max. parte; Rev. Astrag. 125 (1923), exclus. syn. et fig. omn. A. diphysus Gray, Mem. Amer. Acad., ser. nov., 4: 34 (1849) ; Proc. Amer. Acad. 6: 193 (1863) ; Bot. Ives Exped. 10 (1860) ; Walp. Ann. 2: 396 (1851-2) ; Torr. & Gray, Pac. R. R. Rep. 44: 80 (1857), non op. cit. 114 LEAFLETS OF WESTERN BOTANY [ VOL. IV, NO. 5 2:120 (1855) ; Wats., Bot. King 435 (1871), non op. cit. 65; Bran- degee, Bull. U. S. Geol. Geogr. Surv. 2: 234 (1876) ; Rothrock, Rep. Geogr. Surv. (Wheeler Exped.) Bot. 93 (1878), pro parte; Sheld., Minn. Bot. Stud. 1: 171 (1894) ; Coult. & Nels., Man. Rocky Mt. Bot. 284 (1909); Woot. & Standl., Contrib. U. S. Nat. Herb. 19: 361 (1915) ; Tidestr. & Kitt., Fl. Ariz. & N. Mex. 217 (1941), pro parte; Kearn. & Peeb., Fl. Pl. Ariz. 487 (1942), pro parte, exclus. syn. A. diphysus var. albiflorus Gray (1849), 1. c.; vix Bot. Ives Exped. 10 (1860). Tragacantha diphysa (Gray) O. Kze., Rev. Gen. 944 (1891). Astragalus MacDougali Sheld., Minn. Bot. Stud. 1: 169 (1894) ; Tidestr. & Kitt., l. c.; Kearn. & Peeb., 1. c. A. lentiginosus var. MacDougali (Sheld.) Jones, Proc. Calif. Acad. Sci. II, 5: 673 (1895) ; Rev. Astrag. 124 (1923). Cystium diphysum (Gray) Rydb., Bull. Torr. Bot. Club 32: 659 (1905) ; Fl. Rocky Mts. 491 (1917) ; N. Amer. FI. 24 (1929), pro parte, exclus. loc. C. MacDougali (Sheld.) Rydb., op. cit. (1929), p. 415. Pods variable in shape and size, typically much inflated, obliquely ovoid but with nearly straight sutures and slightly upturned beak, sulcate on either suture, or, not uncommonly, less inflated, subglobose to rather narrowly ovoid and rarely arcuate, 1—2.5 (3) cm. long, the valves cori- aceous, glabrous, often mottled: racemes 10—20-flowered, commonly short and dense, 2—4 cm. long in fruit, or rarely looser and up to 6 cm. long, on slightly longer peduncles: calyx strigose with white, very rarely a few black, hairs, the tube 46 mm., the teeth 1.25—2.5 mm. long: petals purple, the banner 13—17 mm., the wings 11—13.5 mm., the keel 10—12.5 mm. long: leaves 5—10 (14) cm. long, petiolate, with 8—10 pairs of oblanceo- late, obovate or elliptic leaflets 7—18 mm. long, glabrous on either face or with a few scattered hairs beneath: stems spreading or ascending, often flexuous, glabrous or sparingly strigulose, 2—4 dm. high. Illustrations. Pl. III, figs. 5—10. Clements, Rocky Mt. Flrs., ed. 3, pl. 29, fig. 25 (1928). Jones, Rev. Astrag., pl. 23 (as var. MacDougalt), not pl. 24 (as var. diphysus). Rydb., Amer. Journ. Bot. 17: tab. XVI, fig. v (1930). Type-locality. Plains around Santa Fe, New Mexico, the type collected by Fendler in 1847. Distribution. Arid plains, dunes, and talus slopes, from central New Mexico in the Rio Grande Valley and southwestern Colorado westward across northern Arizona to Yavapai and Coconino counties. Exsiccata. Cotorapo. SW. Colorado, Brandegee No. 1283 (UC). New Mexico. Santa Fe Co.: Santa Fe, Fendler No. 146 in 1847 (G, type, F, NY), No. 147 (G, type of var. albiflorus), Dr. Edwards in 1849 (NY), Heller & Heller No. 3541 (G, ND, NY), Arséne No. 17074 (F). San Miguel Co.: Santa Fe to Glorieta, Eggleston No. 19996 (G, NY). Berna- . lilo Co.: Albuquerque, Bigelow in 1883 (NY), Herrick in 1894 (NY), Kammerer No. 39 (NY). Sandoval Co.: 39 miles N. of Albuquerque, MARCH, 1945 | PUGILLUS ASTRAGALORUM 115 Wiegand & Upton No. 3626 (F). Socorro Co.: Magdalena, Eggleston No. 16208 (G, NY), No. 16201 (NY), Diehl No. 392 (PO); W. of Magda- lena, Eggleston No. 14325 (G). McKinley Co.: Satan Pass, Ripley & Barneby No. 5279 (RB). Co. undetermined: Mexican Springs, Babcock No. 14312 (G). Arizona. Apache Co.: Holbrook to St. Johns, Eggleston No. 15718 (G, NY). Navajo Co.: Holbrook, Zuck in 1897 (NY), Ward in 1901 (NY), Ripley & Barneby No. 5240 (RB) ; Holbrook to Winslow, Mrs. Fred Stone No. 393 (NY), Peebles No. 11996 (CAS, SAC) ; Wins- low, Peebles & Fulton No. 9589 (SAC). Coconino Co.: Dennison, Rusby in 1883 (F, NY); 35 miles E. of Flagstaff, Wiegand & Upton No. 3628 (F) ; Meteor Crater, Ripley & Barneby No. 5238 (RB) ; Flagstaff to Hol- brook, Nelson & Nelson No. 3125 (NY, UC); 20 miles E. of Flagstaff, Peebles & Smith No. 13964 (NY, SAC), Nelson & Nelson No. 2126 (NY, UC) ; Flagstaff, MacDougal No. 144 (F, G, NY, PO, UC); Wupatki Ruins, Peebles & Parker No. 14631 (NY); S. of Cameron, McKelvey No. 4482 (G) ; Cameron to Navajo Bridge, Nelson & Nelson No. 3858 (UC), Peebles No. 13011 (SAC) ; Aztec Ruins, Little Colorado, Purpus No. 1905 (UC); Cedar Ridge, Ripley & Barneby No. 4871 (RB); Grand Falls, E. G. Smith No. 12009 (SAC), No. 12007 (SAC); Leupp, Peebles & Fulton No. 9563, No. 9567 (SAC) ; 50 miles S. of (Lee’s) Ferry, Jones in 1890 (PO). Yavapai Co.: Ash Fork, Rusby No. 570 (G, UC), No. 571 (NY). Banghart’s Ranch, Rusby No. 570 (G, UC, type collection of A. MacDougali) ; Chino Valley, Loomis No. 6904 (SAC). The var. diphysus is very closely related to var. australis and var. araneosus : indeed it is often difficult, without indication of locality, to separate individual specimens, and the three* should perhaps be united. In general (though exceptions exist) var. diphysus differs from var. araneosus in the less curved, more abruptly beaked and never long-acuminate legume, in the lack of black hairs on the calyx, and in the commonly glabrous stems and herbage. The var. australis, on the other hand, has much looser racemes, chartaceous-membranous, never leathery, pods, and often larger flowers. Vague and unsatisfactory though they be, these three races are well separated geographically, but the same cannot be said of var. MacDougali, here reduced to var. diphysus. In New Mexico and northeastern Arizona the pods of var. diphysus are commonly much inflated and about 2 cm. long; but farther west, and particularly in the poor soil of volcanic mesas about Flagstaff, the entire plant becomes depauperate, the flower is smaller, and the pod is sometimes less than 1 cm. in length and globose-apiculate, or narrowly ovoid and somewhat curved. This is A. MacDougali Sheld. and at its smallest extreme (e. g. MacDougal No. 144) it appears distinct : however, speci- 116 LEAFLETS OF WESTERN BOTANY _ [VOL. IV, NO. 5 mens of an intermediate character in the size of both flower and legume (and these not correlated) are more numerous than those of typical dimensions, and, particularly near Holbrook, they occur together. North of Flagstaff occur plants with small, narrowly ovoid pods disposed in loose racemes (e. g. McKelvey No. 4485, Ripley & Barneby No. 4871) which seem to approach var. palans, and Kearney & Peebles (I. c.) cite a similar gathering (Jones, S. of (Lee’s) Ferry) as A. mokiacensts. The name A. diphysus, in various combinations, has been used very loosely in the literature to cover much wider concepts than are admitted here. By various authorities it has been applied to var. platyphyllidius, var. araneosus, var. salinus, var. australis, var. yuccanus, and others: more detailed citations will be found under these varieties and in the text. Astragalus diphysus var. albiflorus Gray was an albino form found growing with the type near Santa Fe, though according to Jones (Contrib. W. Bot. 8: 4,—1898) some of the material so named by Gray belongs to a form of A. Pattersonii Gray: I have not verified this point. 24. var. SESQUIMETRALIS (Rydb.) Barneby Astragalus lentiginosus Dougl. var. sesquimetralis (Rydb.) Bantety comb. nov. Cystium sesquimetrale Rydb., N. Amer. Fl. 24: 414 (1929). Pods rather narrowly ovoid, gently falcate with ascending beak, 2—2.5 cm. long, coriaceous, strigose, mottled: racemes dense, 2—3 cm. long, on peduncles of about the same length: calyx white-strigose, the tube 4—5 mm., the subulate teeth 2—2.5 mm. long: petals purple, the banner about 14 mm., the wings 11.5—12 mm., the keel 10 mm. long: leaves spreading, subsessile, 4—5 cm. long and shorter than the internodes, with 5—8 pairs of oblanceo- late to oblong leaflets 12—15 mm. long, 4—5 mm. wide, strigose on either face or glabrate above: stems very elongate, 5—7 dm. long, prostrate, flexuous. Illustration. Pl. II, figs. 36—38. Type-locality. Soda Springs, Nevada, the type collected by Shockley. in 1882. Distribution. Known only from the type station. Exsiccata. Nevapa. Esmeralda Co.: Soda Springs, Shockley No. 278 (G, type, NY, fragments, ND, UC). This variety is similar in aspect to vars. albifolius and flori- bundus, but differs in the much larger, purple flower, and in the longer coriaceous pod, while it is unique in the great length of the stems and relatively long internodes. Field-notes made by MARCH, 1945 ] PUGILLUS ASTRAGALORUM 117 Shockley indicate that the plants were found “in alkaline soil” (UC), “moist ground” (G), or “creeping among grasses” (ND), and they are described as ‘“‘spreading on the ground, 5 feet in diameter.”’ It appears to be a remarkable race, one which in an- other section of Astragalus would deserve, in fact, specific status : more collections, however, are needed. 25. var. AUSTRALIS Barneby Astragalus lentiginosus Doug!l. var. australis Barneby, var. nov., var. yuccano Jones necnon var. diphyso (Gray) Jones manifeste affinis, sed ab hoc racemo laxiori fructifero saltem 8 cm. longo, leguminisque valvulis tenuioribus, membranaceo-chartaceis nec coriaceis, ab eo floribus evidentius purpureo-tinctis, carina longiori 12—15 mm., nec 10—11.5 mm. tantum, longa, et legumine longius acuminato magis arcuato et sepissime minus inflato, diversa—A. lentiginosus var. palans Jones, Contrib. W. Bot. 17: 27 (1930), non alibi. Pods membranous to thinly chartaceous, obliquely and sometimes narrowly ovoid, commonly much curved, strongly or but little inflated, glabrous or rarely pubescent, 13—20 mm. long : racemes lax from the begin- ning, many-flowered, 8—12 cm. long in fruit, the peduncles usually stout and of about the same length: calyx strigose with white hairs, the tube 4—5 mm., the teeth 1.5—2 mm. long: petals pale purple (the color rather fugacious), the banner 14—16 mm., the wings 13—15 mm., the keel 12—14 (15) mm. long: leaves 8—12 cm. long, with 9—13 pairs of obovate to oblong leaflets (7) 10—21 mm. long, glabrous above and finely, sometimes copiously, strigose beneath: stems coarse, ascending from a stout perennial root, 3—6 dm. high. Illustration. Pi. III, figs. 15—19. Type. Near Robles, east side of the Baboquivari Mts., Pima County, Arizona, 4ven & Ruth Nelson No. 1537 (New York Botanical Garden). Distribution. Sandy plains and desert washes, from southeastern Ari- zona and southwestern New Mexico to northern Sonora and Chihuahua. Exsiccata. Arizona. Pima Co.: Robles, Nelson & Nelson No. 1537 (NY, type, UC), No. 1534 (NY, UC) ; between Ajo and Sells, Nelson & Nelson No. 1370 (UC); Sells, Buss in 1937 (SAC); Sells to Tucson, Loomis No. 903 (SAC); Papago Reservation, Jones No. 26183 (G, LA, PO) ; Coyote Mts., Ripley & Barneby No. 4244 (RB, PO), Harrison & Kearney No. 3752 (SAC); Tucson, Jones No. 28255 (PO). Pinal (? Pima) Co.: Oracle to Tucson, Peebles No. 6881 (NY). Graham Co.: Tanque, Eggleston No. 19867 (G). Greenlee Co.: Duncan, J. N. Rose No. 11736 (NY), Davidson No. 146 (LA). Cochise Co.: San Pedro River near Fairbank, Ripley & Barneby No. 4234 (RB) ; Bowie, Eastwood No. 8599 (G); Chiricahua Mts., Toumey in 1897 (PO, depauperate) ; Bernardino, Leding 12L (SAC). New Mexico. Hidalgo Co.: near Steins, Ripley & Barneby No. 4216 (RB) ; Lordsburg, Davidson No. 1427 (LA). Dona Ana Co.: Pefia Blanca, Curtin No. 20 (F). Sonora. Boca Grande, 118 LEAFLETS OF WESTERN BOTANY [ VOL. IV, NO. 5 Capt. E. K. Smith (NY). Curmuanua. Lake Gusman, Wright in 1852 (G). The var. australis differs chiefly from var. diphysus in the greater development of all its parts, in the loose raceme evidently much surpassing the leaf, and in the chartaceous to thinly mem- branous texture of the legume. The shape of the latter is subject to a fluctuation in size and turgescence similar to that described for var. diphysus. For the most part the pod is ovoid and strongly inflated, then resembling that of var. diphysus, but averaging smaller and with a more abruptly upturned and proportionately longer beak. In southern Arizona, however, the inflation is com- monly less marked and in the extreme form represented by Jones No. 28255 the pod is scarcely more than turgid, acuminate from the base, strongly falcate throughout, and not exceeding 15 mm. in length. A similar specimen, Jones No. 26183, reported by the collector (Jones, 1930, 1. c.) as var. palans, is anomalous in bear- ing strigulose pods, and is unusually pubescent in the stems and herbage. The var. australis is also related to var. yuccanus, which it resembles in habit, pubescence, and a tendency to produce “cari- nate” flowers, with the keel prolonged to equal or surpass the wings, or even, on occasion, to attain the length of the banner: the larger flower, however, and the purple coloring of the petals will usually serve to distinguish it. The specimen from Sonora is referred here only with doubt, for it has no fruit. It is of historical interest in that it formed one of the three elements, all different, upon which Gray founded his A. Arthu-Schottu (var. borreganus Jones). 26. var. YUCCANUS Jones Astragalus lentiginosus Dougl. var. yuccanus Jones, Contrib. W. Bot. 8:3 (1898) ; Rey. Astrag. 127 (1923), pro max. parte. Cystium yuccanum (Jones) Rydb., N. Amer. Fl. 24: 407 (1929). Astragalus yuccanus (Jones) Tidestr., Proc. Soc. Wash. 48:40 (1935). A. Fremontii var. yuccanus (Jones) Tidestr., in Tidestr. & Kitt., Fl. Ariz. & New Mex. 216 (1941). A. diphysus Kearn. & Peeb., Fl. Pl. Ariz. 487 (1942), pro parte, e syn. cit., non Gray. Pods subglobose or broadly ovoid with abruptly upturned triangular beak, bladdery-inflated, membranous, shining, glabrous or minutely strigu- lose, 17—25 mm. long: racemes lax, (6) 8—12 cm. long in fruit, on stout peduncles about equaling the subtending leaf: calyx rather loosely pubes- cent to merely strigose with white or some black hairs, the tube 3.75—4.5 mm., the teeth 1—1.5 mm. long: petals narrow, white, ochroleucous or MARCH, 1945] PUGILLUS ASTRAGALORUM 119 occasionally flushed with mauve, the banner 11.5—13.5 mm., the wings 11— 12.5 mm., the keel 10.5—11.5 mm. long: leaves spreading, (6) 8—14 cm. long, with 7—10 pairs of elliptic to broadly obovate, emarginate leaflets 1—2 cm. long, glabrous above, strigose beneath with very fine white hairs or in youth, like the stems, canescent: stems erect or strongly ascending, 1.5—4 dm. high, from a winter-annual or short-lived perennial root. Illustrations. P1. III, figs. 34—37. Jones, Rev. Astrag., pl. 26. Type-locality. Yucca, Arizona, the type collected by Jones in 1884. Distribution. Sandy foothills and desert washes of western Arizona, from Mohave south to Pima and Maricopa counties. Exsiccata. Arizona. Mohave Co.: Yucca, Jones No. 3886 (PO, type, F, G, NY), Ripley & Barneby No. 3380 (RB); Oatman to Kingman, Harrison & Kearney No. 7613 (F, NY, SAC); Kingman, Kearney & Peebles No. 11143 (SAC) ; Hackberry, Jones in 1884 (PO), in 1903 (PO). Yavapai Co.: Congress Junction, Jones in 1903 (PO). Maricopa Co.: Wickenburg, Palmer No. 589 (F, NY), Peebles No. 8474 (F, PO), Ripley & Barneby No. 4261 (RB), No. 4262 (PO, RB). Pima Co.: Gunsight Pass, Fosberg No. 904 (PO). This variety, which in the past few years has been raised to specific rank by Tidestrom and reduced to synonymy with A. di- physus by Kearney and Peebles, is rather easily recognized by its ochroleucous or white flowers (occasionally tinged with mauve), the narrow banner which is very short in proportion to the keel, the loose racemes on their stout, rather strict, peduncles, and the subglobose, papery legume. In the area immediately north and east of Kingman, Arizona, are found plants with purple flowers and coriaceous pods which seem to approach or are per- haps the result of hybridization with var. mokiacensis (q. v.), while to the southwest it is not clearly distinguished from var. australis. In the middle of its range, as in the open sandy valleys about Yucca and Oatman, the plants of var. yuwccanus sometimes occur in enormous numbers, forming between the bushes of Larrea an almost pure association over hundreds of acres. 27. var. VITREUS Barneby Astragalus lentiginosus Doug]. var. vitreus Barneby, var. nov. legu- mine vesicario-inflato tenuiter membranaceo var. salino (Howell) Barneby, a quo floribus majusculis purpureo-tinctis differt, haud absimilis. A var. diphyso (Gray) Jones legumine haud coriaceo, racemoque laxo, a var. Ken- nedyi (Rydb.) Barneby floribus pallidis calycisque parce strigosi nec patule pubescentis dentibus brevioribus, necnon a var. stramineo (Rydb.) Barneby | imprimis foliis utrinque glabratis aliisque notulis recedit. Pods ovoid and much inflated, with a very short triangular beak, or less inflated, lanceolate-ovoid, long-acuminate, and strongly arcuate, the 120 LEAFLETS OF WESTERN BOTANY [ VOL. IV, NO. 5 valves papery-membranous, shining, glabrous, not mottled: racemes 10—20- flowered, rather lax, 5—6 cm. long in fruit, on peduncles about as long: calyx strigose with rather scattered black or white hairs, the tube 4 mm., the teeth 1.25—1.5 mm. long: petals very pale but distinctly tinged or veined with purple, the banner 13—15 mm., the wings 11—12 mm., the keel 10—11 mm. long: leaves 6—8 cm. long, petiolate, with 7—8 pairs of dark green, oblong to obovate, emarginate leaflets 6—12 mm. long, sparingly ciliate, glabrous on either face or with a few scattered hairs beneath: stems ascending, rather slender, striate, sparingly strigose, 1.5—3 dm. high. Illustration. P1. III, figs. 30—33. Type. In sand desert, 5 miles west of Leeds, Washington County, Utah, Bassett Maguire & H. L. Blood No. 4413 (Pomona College Herbarium No. 211065). Distribution. Badlands along the Virgin River in southwest Utah, and in adjacent Arizona. Exsiccata. Utan. Washington Co.: W. of Leeds, Maguire & Blood No. 4413 (PO, type, UC); Leeds, Vansell in 1940 (CAS, RB); E. of Leeds, C. L. Hitchcock No. 3025 (F, PO); Washington, Eastwood & Howell No. 9155 (CAS, RB); NE. of Washington, Tidestrom No. 9385 (G) ; St. George, Cottam, Stanton & Harrison No. 4018 (PO). Arizona. Coconino Co.: Kaibab, Jones No. 25443 (PO). Mohave Co.: W. of Fre- donia, Peebles & Parker No. 14687 (SAC); Pipe Springs, Peebles No. 13071 (SAC). The affinities of var. vitreus are not immediately apparent, for it combines the facies of the northern representatives of A. lenti- ginosus with some of the characters of the southerly forms. In habit and in the legume it resembles var. salinus, but the flowers are much larger and at least purple-tinged: in this last respect it approaches var. diphysus (from which it differs in the membra- nous pod and lax racemes) or var. Kennedyi (from which the glabrous herbage and sparsely strigulose calyx immediately dis- tinguish it). Perhaps it is most nearly related to var. stramineus, which it resembles in the proportions and pale coloring of the petals, but the pubescence is entirely different and the pod of a much thinner texture. With one exception the material at hand is very uniform, and the broadly ovoid, strongly inflated pods with their ventral suture nearly straight or a little convex are constant in size and outline. The Hitchcock collection, however, bears narrowly ovoid-acumi- nate, less inflated pods, which are arcuate and gradually acuminate almost from the base, and the ventral suture is concave. At first sight this appears very distinct, but it differs in no other detail from the prevailing form of var. vitreus: analogous examples of MARCH, 1945] PUGILLUS ASTRAGALORUM 121 fluctuation in the inflation of the pod have been remarked in many of the varieties of A. lentiginosus. 28. var. KENNEDYI (Rydb.) Barneby Astragalus lentiginosus Doug!l. var. Kennedyi (Rydb.) Barneby, comb. nov. Cystium Kennedyi Rydb., N. Amer. FI. 24: 407 (1929). Astragalus Fremontii subsp. eremicus Abrams, Ill. Fl. Pac. St. 2: 597 (1944), pro parte, e syn. cit. Pods rather narrowly ovoid-acuminate with long upturned beak, inflated, membranous, not mottled, (1.5) 2.5—3 cm. long: racemes 10—30-flowered, commonly very lax, 5—10 cm. long in fruit, on stout peduncles about as long: calyx loosely pubescent with white or often some black hairs, the tube (3.5) 4—5 mm., the teeth (1.5) 2—2.5 mm. long: petals bright purple, the banner 12—15 mm., the wings 11—12.5 mm., the keel (8) 10—11 mm. long: leaves 5—10 cm. long, with 7—10 pairs of oblong, oblanceolate or obovate leaflets 7—20 mm. long, woolly-pubescent (typically) on both faces with tangled pubescence of fine, curly hairs, or (exceptionally) glabrate above or on either surface: stems stout, erect, subtomentose to glabrate, (1) 2—5 dm. high. Illustration. P1. III, figs. 20—25. Type-locality. Carson Sink Region, Nevada, the type collected by P. B. Kennedy. ; Distribution. Alkaline dunes and sandy valleys, middle western Nevada. Exsiccata. Nevapa. NW. Nevada, Lemmon in 1875 (G). Washoe Co.: Pyramid Lake, Kennedy No. 1019 (NY, UC), Ripley & Barneby No. 4501 (RB), K. Brandegee in 1913 (PO), Lemmon No. 137% (F); Romola Station, Pyramid Lake Road, Mathias No. 1227 (UC); W. of Reno, Mason No. 4602 (UC) ; Wadsworth, K. Brandegee No. 14 in 1913 (PO), Jones in 1930 (PO); N. of Wadsworth, Archer No. 6195 (NY, UN) ; NW. of Wadsworth, Hendrix No. 855 (UC). Ormsby Co.: Empire City, Jones in 1881 (PO). Mineral Co.: N. end of Walker Lake, Ripley & Barneby No. 4440 (RB); S. side of Walker Lake, Heller No. 10915 (F, G, NY, UC); Hawthorne, Jones in 1882 (PO); Mina, Ripley & Barneby No. 4436 (RB). Churchill Co.: Carson Sink region, Kennedy No. 1691 (NY, type, UC). Nye Co.: 20 miles W. of Tonopah, Eastwood & Howell, No. 9482 (CAS, RB); W. of Rattlesnake, Ripley & Barneby No. 4427 (PO, RB). The var. Kennedyi, as typically developed in sandy, alkaline soil in the neighborhood of Pyramid and Carson lakes, is ex- tremely distinct, differing from var. Fremontii in the much larger flowers, elongate calyx-teeth, tangled pubescence, and very loose racemes of long-acuminate pods. In the south, however, there is evidence of intergradation, and the last two collections cited are of an intermediate character, the racemes more con- 122 LEAFLETS OF WESTERN BOTANY [ VOL. IV, NO. 5 densed, the pubescence shorter or almost lacking, the flowers and calyx smaller. This race was first recognized by Rydberg as a species of Cystiugn, Jones having determined some of his col- lections as var. diphysus, others as var. Fremontii, and those of Mrs. Brandegee (which are decidedly depauperate) as var. palans. 29. var. STRAMINEUS (Rydb.) Barneby Astragalus lentiginosus Doug]. var. stramineus (Rydb.) Barneby, comb. nov. Cystium stramineum Rydb., N. Amer. Fl. 24: 409 (1929). C. Coulteri Rydb., Fl. Rocky Mts. 419 (1917), pro parte, e loc. cit., non A. Coulteri Bth. Astragalus Coulteri Tidestr., Contrib. U. S. Nat. Herb. 25: 325 (1925), pro parte; Tidestr. & Kitt., Fl. Ariz. & New Mex. 217 (1941), pro parte, e loc. cit., non Bth. A. Fremontii Kearn. & Peeb., Fl. Pl. Ariz. 486 (1942), e num. cit.; non Gray. Pods broadly ovoid with upturned beak, chartaceous, inflated, strigose, green or at length stramineous, mottled, 1—2 cm. long: racemes 15—18- flowered, lax and becoming 5—9 cm. long in fruit, on peduncles of about “he same length: calyx white-strigose or -sericeous, the tube 3—4 mm., the teeth 1.5—2 mm. long: petals pale purple, quickly fading brownish, the banner 11—12 mm., the wings 9.5—10.5 mm., the keel 9—10 mm. long: leaves 5—8 cm. long, shortly petiolate, with 6—8 pairs of rather distant, oblong-obovate, emarginate leaflets 5—19 mm. long, green and glabrous or finely pubescent above, more or less canescent-strigose below, particularly in youth: stems erect or ascending from an annual or short-lived perennial root, canescent or at length merely strigose, flexuous, 1.5—4 dm. high. Illustration. Pl. III, figs. 26—29. Type-locality. Southeastern Utah, the type collected by Edward Palmer in 1870. Distribution. Sandy valleys or on dunes, from southeastern Utah to northwestern Arizona and adjacent Nevada. Exsiccata. Utan. SE. Utah, Palmer in 1870 (NY, type) ; S. Utah, Bishop in 1874 (PO), Palmer No. 116 (G, NY). Arizona. Mohave Co.: Beaver Dam, Peebles & Parker No. 14767 (G, NY, SAC); Littlefield, Eastwood & Howell No. 9037 (CAS) ; Mesquite to Littlefield, Kearney & Peebles No. 13190 (SAC) ; Arizona strip, Maguire & Blood No. 4414 (PO, UC), No. 4415 (F, PO, UC). Nevapa. Clark Co.: Mesquite, Ripley & Barneby No. 4295 (PO, RB). The var. stramineus is a somewhat uncertain entity, partly owing to the fragmentary nature of the type, which is exactly matched only by Bishop’s collection at Pomona. Both of these, from localities not precisely noted, have a rather curious pubes- cence made up of long, straight, silky hairs, and the upper surface MARCH, 1945] PUGILLUS ASTRAGALORUM 123 of the leaflets is glabrous. The more copious material from Ari- zona and Nevada, while essentially similar in the flower and legume, has a vesture of shorter hairs, and the leaflets are less obviously bicolored, and often sparingly pubescent above. Ryd- berg described the petals as white, but this is doubtfully correct. In the Bishop collection there are distinct traces of pigment still visible in the petals, and in the rest the color is pale when fresh and fugacious in the herbarium. This variety is very similar to var. variabilis of the Mohave Desert in California, differing chiefly in the relative length of the calyx-teeth, paler flowers and somewhat different aspect of the plant itself. The areas of distribution of these two races are quite disjunct, being separated by a gap of at least two hundred miles. It has been treated by Tidestrom as A. Coulteri and by Kearney and Peebles as A. Fremontii, but it is probably more closely related to var. vitreus than to any other. 30. var. VARIABILIS Barneby Astragalus lentiginosus Dougl. var. variabilis Barneby, var. nov., inter var. Coulteri (Bth) Jones, var. Fremontii (Gray) Wats. et var. migri- calycem Jones fere exacte intermedia, sed a prima corollz minoris carina 8—10 mm. tantum longa, a secunda corolla manifeste majori, a tertia petalis saturate purpureis nec ochroleucis facile separanda. Proles valde poly- morpha, pube adspectuque variabilis—A. Arthu-Schottti Gray, Proc. Amer. Acad. 6: 209 (1863), pro parte, exclus. typ.; op. cit. 7: 337 (1867), pro max. parte; Abrams, Ill. Fl. Pac. St. 2:597 (1944), pro max. parte. A. Coulteri, A. lentiginosus var. Coulteri, A. lentiginosus var. Fremontii auct. calif. pro parte, non A. Coulteri Bth. mec A. Fremontti Gray sens. strict. Cystium Arthu-Schottii Rydb., N. Amer. Fl. 24: 409 (1929), pro max. parte, exclus. typ. cit. C. pardalotum Rydb., op. cit., 415. Pods obliquely ovoid with upturned triangular beak, much inflated, mem- branous to rather firm in texture, sericeous to strigose with loosely ap- pressed or incumbent hairs, never glabrous, 12—25 mm. long: racemes 12—30-flowered, (3.5) 5—12 cm. long and rather loose in fruit, surpassing the subtending leaf: calyx silvery-strigose, silky or hirsutulous-tomentose with all white, black and white, or all black hairs, the tube 3—4.5 mm., the teeth 0.75—1.75 mm. long: petals purple, the banner 10—14 mm., the wings 9—11.5 mm., the keel (shorter than, equaling or often exserted from be- tween the wings) 7.5—10 (10.5) mm. long: leaves 5—9 cm. long, spread- ing, with 6—10 pairs of obovate or oblong leaflets 5—15 mm. long, densely silvery-canescent or white tomentose on both faces to green and strigose or glabrate above: stems commonly tomentose-canescent, at least in youth, rarely green and loosely strigose, ascending, coarse or rather slender, 1.5—4 dm. high, from an annual or short-lived perennial root. 124 LEAFLETS OF WESTERN BOTANY [ VOL. IV, NO. 5 Illustrations. Pl. IV, figs. 1—8. Jones, Rev. Astrag., pl. 25 (as var. Fremontii,a). Abrams, Ill. FI., fig. 2870. Type. Victor (now Victorville), San Bernardino Co., California, M. E. Jones in 1903 (Pomona College Herbarium No. 25684 and 25685). Distribution. Sandy plains and desert washes of the Mohave Desert, California, from southern Inyo Co. southeast to Baker, San Bernardino Co., and westward to the head of the San Joaquin Valley. Exsiccata. CALIFORNIA. San Bernardino Co.: Mohave River, Frémont (NY), Palmer in 1876 (NY); Rock Spring, Mohave River district, Palmer No. 104 in 1876 (UC) ; Mohave Desert, Parish & Parish No. 4957 (NY). Dale Dry Lake, Munz No. 15640 (CAS, PO) ; Windmill Tank, Little San Bernardino Mts., C. L. Hitchcock No. 12235 (PO, UC) ; War- rens Well, Munz & Johnston No. 5174 (PO) ; Keyes Ranch, Jones in 1931 (PO), Gilman No. A4 (PO), No. A7 (PO), Munz & Johnston No. 5272 (PO); Lucerne Valley, Lefebure No. 487 (SB); Camp Cady, Cooper (G) ; Rabbit Springs, Parish No. 4957 (NY); W. of Old Woman Well, Ferris & Bacigalupi No. 8116 (UC); Old Woman Springs, Ripley & Barneby No. 4276 (RB, PO); Barstow, Jones in 1924 (PO), Hall & Chandler No. 6839 (PO, UC), Spencer No. 1424 (G), No. 1962 (PO), Parish No. 9668 (UC), M. S. Baker No. 8870 (CAS), Munz No. 2579 (PO), Heller & Heller No. 2982 (NY), Eastwood No. 17982 (CAS) ; Garlic Springs, Munz & Keck No. 7903 (G, PO); Leach Spring, Jaeger in 1940 (PO); Adelanto, Parish No. 11823 (UC), Clokey & Templeton No. 5756 (NY, PO) ; Copper Mt., Spencer No. 1768 (G) ; N. of Hinkley, Clokey & Anderson No. 6686 (F, LA, NY, PO); Barstow to Kramer, Hoffmann No. 10869 (SB); Kramer Junction, Constance & Mason No. 2112 (UC); E. of Victorville, Jaeger in 1932 (PO), Johnston in 1920 (PO), Ripley & Barneby No. 3264 (RB), Eastwood No. 18714 (CAS) ; Victorville (Victor), Jones in 1903 (PO, type, UC), in 1926 (PO), Parish No. 9225 (PO), Munz & Youngberg No. 15179 (PO), Hall No. 6203 (PO, UC), Feudge No. 76 (PO), John Roos No. 487 (PO), Mason No. 3067 (UC), Ripley & Barneby No. 4282 (RB); Hesperia, Spencer No. 368 (G, PO), Munz No. 4445 (PO); Cajon Pass, Cooper (G). Los Angeles Co.: Palmdale, Elmer No. 3641 (G, NY, PO), Abrams & Mc- Gregor No. 516 (G, NY, PO); Pallett Creek, Munz No. 6907 (PO). Kern Co.: Muroc, Hoffmann No. 8499 (SB), Crum No. 1809 (UC); Walker Pass, Purpus No. 5723 (PO, UC) ; N. of Rosamond, Abrams No. 11158 (F); Mohave, Eastwood No. 3218 (CAS, G, NY); Rosedale, Abrams in 1900 (PO); Ricardo, Munz No. 12466 (F, PO, UC); N. of Ricardo, J. T. Howell No. 4981 (CAS), Hoffmann No. 9567 (SB) ; El Paso Range, Abrams No. 11894 (PO). Inyo Co.: Intake, Cottonwood Creek, Kerr No. 565-5 (CAS). Form fg. San Bernardino Co.: Lavic, Munz, Johnston & Harwood No. 4166 (PO), No. 4288 (PO) ; Ludlow, Hall No. 6112 (UC) ; Baker, Jones in 1934 (PO, UC); Afton Canyon, Jaeger in 1940 (PO) ; Pisgah Crater, Ripley & Barneby No. 5209 (RB). MARCH, 1945] PUGILLUS ASTRAGALORUM 125 Form y. Inyo Co.: Salt Wells, Hall & Chandler No. 6889 (G, PO, UC) ; Lone Pine, Purpus No. 5741 (PO, UC), Hall & Chandler No. 7177 (PO, UC); Lone Pine to Keeler, Ripley & Barneby No. 3767 (RB) ; Keeler, Ripley & Barneby No. 3768 (RB). Kern Co.: Walker Pass, Purpus No. 5403 (G, UC), Ripley & Barneby No. 3777 (PO, RB), No. 3778 (RB, fma. floribus albis). San Bernardino Co.: Ophir Mine, Slate Mts., Epling, Ellison & Anderson in 1930 (CAS, F, UC) ; SW. of Trona, Gould No. 985 (NY); Trona, Epling & Anderson in 1931 (CAS, UC). Form §. Kern Co.: S. of Willow Springs, Munz No. 10055 (NY, PO, UC) ; Rosamond to Mohave, Abrams No. 11794 (PO, UC) ; Bakersfield, Jones in 1906 (PO), in 1934 (PO) ; 18 miles N. of Bakersfield, Munz No. 9015 (PO, UC); SE. of Shafter, Bracelin No. 867 (PO, CAS). In restricting var. Coulteri to the race with large flowers endemic to the Colorado Desert, and admitting to var. Fremont only those plants with short keel and glabrous or sparingly strigu- los¢ pod from the northern and far eastern Mohave Desert to Nevada, a large block of specimens of intermediate character from the intervening territory is left without a name, and it is for this highly polymorphic aggregate that the name var. variabilis is proposed. The disposition of this material has always presented a problem, and in any large herbarium sheets may be found filed both under var. Coulteri and var. Fremontiu. The very earliest collection (Frémont), in flower only, was included by Gray in his A. Arthu-Schottii, the type of which, as I have attempted to show elsewhere (Proc. Calif. Acad. Sci. IV, 25: 158 sequ.), be- longed to var. borreganus Jones, and the first fruiting collection formed the basis of Gray’s later description (1867, 1. c.) of the pod of that species. In subsequent years both Gray and Watson, having this plant in mind, reduced A. Arthu-Schottiui to synonymy with A. Coulteri, and since that time the limits of A. Coulteri and the point of its passage into 4d. Fremonti have become pro- gressively more nebulous and difficult of definition. So vague, indeed, are the authorities dealing with this immediate race that I have not thought it worth while to list the references in the literature which, to greater or less extent, belong to var. variabilis. An example of this vagueness may be found in Jones’s Revision. Here the range of A. lentiginosus var. Coulteri is given as “common on the Mohave and Colorado deserts, and Owens Valley,” and that of var. Fremontii as “common from . Darwin and the Death Valley region to Mexico,” indicating that both varieties occur throughout the entire desert region of 126 LEAFLETS OF WESTERN BOTANY _ [VOL. IV, NO. 5 southern California. On plate 25 of the same volume at least part of the figures corresponding with these same two varieties were drawn from specimens collected at Tehachapi and Victor- ville respectively, and while neither could possibly portray the original A. Coulteri and A. Fremontt, at least the latter belongs to var. variabilis. Rydberg, to whom there was available a comprehensive suite of specimens from the western Mohave Desert, annotated them promiscuously as Cystium eremicum (e. g. Eastwood No. 3281, Munz No. 10005, etc.), C. Coulteri (the old Frémont gathering from the Mohave River), and C. Arthu-Schotti (Parish No. 4957, Abrams & McGregor No. 516, but not, surprisingly, the Frémont cotype of A. Arthu- Schottii), and on another specimen collected by Davidson erected a fourth species, C. pardalotum. This last was supposed to belong to the group Diphysa of Cystium, and to have glabrous pods and short, dense racemes : from the fragments of the type in the New York Botanical Garden it is apparent that the pods are strigulose and the racemes lax. Jepson in his Manual (1925) referred all the desert forms (except, of course, var. albifolius) to A. Coultert, but in his Flora (1936) all but A. Coulteri sensu strictissimo are shifted to var. Fremontit. : It might be argued that the var. variabilis should somehow be divided between the vars. Fremontii and Coulteri, and perhaps var. nigricalycis, to each of which it is about equally related. Nonetheless it does represent, at least in the center of its range, an entity demonstrably distinct from all three, and for this reason, apart from considerations of taxonomic convenience, it well merits recognition. In the Victorville-Barstow region the typical form of var. variabilis is abundant and well defined, the pubescence and size of the corolla being intermediate between vars. Coulteri, Fre- montii, and nigricalycis, but towards the periphery of its area it passes insensibly into three subsidiary forms which have been separated in the Exsiccata as £, y, and 6. The form £ from the central Mohave Desert has the sericeous vesture, even to the pods, of var. Coulteri, but still differs in the shorter calyx teeth and smaller flowers: the form y, from Owens Valley south to Walker Pass, is again silky-villous throughout, with rather slender, flexuous stems from an annual root, and elongated, often MARCH, 1945] PUGILLUS ASTRAGALORUM 129 zigzag, racemes, and thus approaches the more pubescent phase of var. Fremontii which was recognized by Sheldon as A. eremicus: while form 8, from both sides of the mountains in southern Kern County, is separable from var. nigricalycis only by the purple color of the corolla. These minor races, which form the links con- necting var. variabilis on all sides with its nearest relatives, have not been considered worthy of taxonomic rank, though in a more elaborate classification they might be recognized as forme. The carinate condition of the flower, already observed in var. carinatus, in which the keel is elongated to equal or quite com- monly to surpass the wings, has been collected only about Hes- peria, Victorville, and Barstow. Even here it does not entirely replace the normal state, and is perhaps more frequent in years of heavy rainfall. 31. var. NIGRICALYCIS Jones Astragalus lentiginosus Doug]. var. nigricalycis Jones, Proc. Calif. Acad. Sci. II, 5: 674 (1895) ; Rev. Astrag. 124 (1923) ; Munz, Man. S. Calif. 270 (1935) ; Jeps., Fl. Calif. 2: 356 (1936). A, lentiginosus var. Fremontti Jones, Zoe 4: 272 (1894), non alibi. A. Fremontti Heller, Muhlenb. 2: 84 (1905), e num. cit., non Gray. A. nigricalycis (Jones) Abrams, Ill. Fl. Pac. St. 2: 598 (1944). Pods ovoid with ascending beak, firm in texture but chartaceous, mottled, villosulous with white incumbent hairs, 2—3.5 cm. long: racemes 12—25- fiowered, at first crowded, becoming lax and 5—8 cm. long in fruit, com- monly surpassing the leaf: calyx villous with black (or very rarely white) hairs, the tube 4—5 mm., the teeth 1—2 mm. long: petals pale yellow or ochroleucous, the banner 13—15 mm., the wings 11—13.5 mm., the keel 9.5—12 mm. long: leaves 7—13 cm. long, with 9—11 pairs of green, oblong to obovate leaflets 8—20 mm. long, densely villous with spreading hairs on either surface or somewhat glabrate above: stems commonly white-villous- tomentose, coarse, flexuous, fistulose, 2—5 dm. high, diffuse . Illustrations. P\. IV, figs. 12—13. Jones, Rev. Astrag., pl. 23 and 24 (the septum faulty). Jepson, Fl. Calif. 2: fig. 209. Abrams, Ill. FI., fig. 2872. Type-locality. “Bakersfield,” or as labelled on the type-sheet (PO), San Emigdio, Kern County, California, the type collected by Alice East- wood in 1893. Distribution. Plains and foothills of the San Joaquin Valley, Cali- fornia, along the Temblor Range from southwestern Fresno County south to Santa Barbara and Kern counties. Exsiccata. CALIFORNIA. Kern Co.: San Emigdio, Eastwood in 1893 (PO, type); mouth of San Emigdio Canyon, Davy No. 1982 (UC); Bakersfield, Eastzvood in 1893 (UC, cotype), Vasey in 1874 (G), Davy No. 1714 (UC), L. Benson No. 3337 (UC); Tejon Pass to Bakersfield. 128 LEAFLETS OF WESTERN BOTANY __ [VOL. IV, NO. § Munz No. 9002 (PO); SE. of Bakersfield, Ora Clark No. 1846 (LA); S. of Bakersfield, Munz No. 11652 (PO) ; Wheeler Ridge, L. S. Rose No. 35655 (CAS) ; Grapevine, Hilend & Newsom in 1927 (PO) ; Fort Tejon, Wm. Kennedy in 1876 (F) ; Reed’s Station, L. S. Rose No. 37051 (LA); Wasco, Munz No. 10110 (PO, UC) ; Taft, Hoffmann in 1928 (SB), Munz No. 13642 (PO, UC), L. Benson No. 2956 (UC) ; Arvin, Clokey No. 6680 (F, LA, NY, PO) ; W. of Blackwell’s Corner, Schreiber No. 2479 (UC); San Diego Canyon, Temblor Range, Eastwood & Howell No. 4108, (CAS). Ventura Co.: Ozena P. O., Chester Dudley in 1936 (CAS) ; between Ozena and Frazier Park, Blanche Clear in 1940 (CAS). Santa Barbara Co.: base of Temblor Range, Ferris No. 9073 (CAS); Cuyama Valley, L. S. Rose No. 36062 (CAS, F, PO, UC), Keck No. 2247 (CAS). San Luis Obispo Co.: Painted Rocks, Carriso Plain, Eastwood in 1896 (G, NY). Kings Co.: N. of Devils Den, J. T. Howell No. 5889 (CAS) ; Avenal, Eastwood & Howell No. 2053 (CAS, G, NY). Fresno Co.: Zapato Chino, Brandegee in 1893 (PO, UC); Oil City, Heller No. 7587 (CAS, F, G, NY, UC), Ripley & Barneby No. 3240 (RB) ; W. of Kerman, Hoover No. 287 (UC) ; Coalinga, Constance No. 2087 (NY, UC) ; Alcalde, Brandegee in 1890 (UC). Throughout the greater part of its range the var. nigricalycis presents the two characters of black-villous calyx and ochro- leucous flowers in constant association, and it is preferable to restrict the name to cover this form alone, although there are in- cluded above two specimens, Munz No. 11652 and the Blanche Clear gathering, in which the calyx is white-hairy. In the Bakers- field region and on the desert side of the mountains in Kern County, plants occur which are scarcely separable from var. nigri- calycis except for the fact that the petals are red or purple and that the calyx may be villous with either black or white hairs. Specimens corresponding with this form have been enumerated under var. variabilis as fma §. Ina case of this nature where two races, more or less distinct in the middle of their ranges, are morphologically confluent at the point where their areas touch or overlap, the disposition of the intergrading material must always be artificial and a matter of opinion. But the existence of such intergrades cannot be said to invalidate either variety. 32. var. CHSARIATUS Barneby Astragalus lentiginosus Doug]. var. cesariatus Barneby, Proc. Calif. Acad. Sci. IV, 25: 161 (1944). Pods obliquely ovoid, somewhat arcuate, the ventral suture a little con- cave, and the dorsal strongly convex, abruptly acuminate and the beak up- turned, the valves coriaceous, mottled, sparingly strigose-hirsutulous with MARCH, 1945] PUGILLUS ASTRAGALORUM 129 incumbent black and white hairs : racemes 15—20-flowered, at first crowded but at length laxer and 3.5—7 cm. long in fruit, on stout peduncles shorter than the leaves: calyx copiously but rather loosely black-strigose, the tube 6 mm., the lanceolate acute teeth 2.5 mm. long: petals brilliant purple, the broad banner 17—19 mm., the wings 14—15 mm., the keel 12—13 mm. long: leaves shortly petiolate or the upper subsessile, 8—10 cm. long, with 11—13 pairs of obovate-cuneate to oblanceolate, truncate or emarginate leaflets 8—18 mm. long, glabrous above, ciliate and sparsely hirsutulous below with ascending or incumbent white or some black hairs: stems numerous from a heavy perennial root, flexuous, prostrate, glabrate, stramineous, 4 dm. long or more. Illustrations. P1. IV, figs. 8—11. Barneby, op. cit., pl. 17, figs. 29—30. Type-locality. Temblor Range, above McKittrick, Kern Co., Cali- fornia, the type collected by H. D. Ripley and R. C. Barneby. Distribution. Known only from the type-station, on grassy slopes in the foothills of the Inner Coast Range. Exsiccata. CaLirornta. Kern Co.: Temblor Range, Ripley & Barneby No. 3243 (CAS, type, PO, RB). The var. cesariatus is in some degree intermediate between var. nigricalycis and var. tehatchapiensis, and was described as most closely related to the latter, though differing in the larger, more deeply colored flower, much longer calyx-teeth, and more pubescent pod of rather different shape. I now believe that it is more justly placed next to var nigricalycis, for the raceme is in- clined to be lax in fruit, and the pubescence, while much less copious, is of a similar character. It may be quickly distinguished, however, by the larger purple corolla, more coriaceous legume, and prostrate stems. In nearly all the varieties of A. lentiginosus some nigrescent villi are found either on the calyx, bracts, or pedi- cels, but var. c@sariatus is unique in bearing black hairs on the rachises, petiolules, and even the leaflets of the upper leaves. 33. var. COULTERI (Bth.) Jones Astragalus lentiginosus Dougl. var. Coulteri (Bth.) Jones, Contrib. W. Bot. 8: 4 (1898) ; Rev. Astrag. 127 (1923), pro max. parte, exclus. syn. A. Coulteri Bth., Pl. Hartw. 307 (1848) ; Gray, Proc. Amer. Acad. 6: 233 (1863) ; Bot. Calif. 1: 146 (1880), exclus. syn., non op. cit. 2: 442; Wats., Bot. King Explor. 66, 435 (1871), exclus. syn.; Jeps., Man. Calif. 568 (1925), pro parte; Fl. Calif. 2: 356, sens. strict., exclus. syn.; Munz, Man. S. Calif. 270 (1935), exclus. syn.; Abrams, III. FI. Pac. St. 2: 597 (1944). Tragacantha Coulteri (Bth.) O. Kze., Rev. Gen. 944 (1891). Cystium Coulteri (Bth.) Rydb., Bull. Torr. Bot. Club 40:50 (1913) ; FI. Rocky Mts. 491 (1917), pro parte, exclus. loc. plur.; N. Amer. FI. 24: 408 (1929), sens. strict. 130 LEAFLETS OF WESTERN BOTANY _ [VOL. IV, NO. 5 Pods obliquely ovoid with short ascending beak, about 2 cm. long, the valves firmly chartaceous or subcoriaceous, densely canescent, mottled: racemes 13—25-flowered, in fruit 6—9 cm. long, on stout, widely spreading, often flexuous peduncles of about the same length: calyx rather loosely silky-villous with straight hairs, white or sometimes nigrescent, the tube 4—5 mm., the teeth 2—3 mm. long: petals bright purple, the banner 14—15 mm., the wings 12—13 mm., the keel 10—11.5 mm. long : leaves 5—10 cm. long, with 7—10 pairs of oblong to obovate leaflets 5—15 mm. long, silvery- sericeous throughout or the lowest sometimes greenish but always densely pubescent: stems stout, flexuous, white-tomentose or -sericeous, 2—3 dm. high from a biennial or short-lived perennial root. Illustrations. P1. IV, figs. 14—17. Jones, Rev. Astrag. pl. 25 (exclus. fig.a). Abrams, Ill. Fl., fig. 2868 Type-locality. “Juxta Monterrey,” according to the original publi- cation, but certainly on the Colorado Desert in California: the type col- lected by Thomas Coulter. Distribution. Colorado Desert, California. Often reported from Mexico, but no specimens have been seen from south of the Salton Sea. Exsiccata. CAatirorntA. Southern California, sine loc., Parry & Lem- mon No. 87 in 1876 (F, G, NY, UC: the last is erroneously labelled “Mo- have River District”), Parry No. 1 in 1876 (G), Lemmon No. 150 in 1876 (F). Riverside Co.: San Gorgonio Pass, Lemmon No. 589 (G); Ban- ning, Gilman No. 47 (UC); Whitewater, Vasey in 1881 (F, ND, NY), Parish & Parish No. 25 (ND), Parish No..6119 (F), No. 19189 (UC), Spencer No. 1501 (F), Hall No. 5757 (F, G, NY, PO, UC), Munz No. 2307 (PO), I. Johnston No. 1065 (NY, PO) ; Palm Springs, Jones in 1903 (PO), Eastwood No. 3116 (CAS, G, NY), Spencer No. 743 (NY), No. 1655 (G), Grant No. 6747 (CAS, UC). No. 6714 (PO, UC), Templeton & Clokey No. 1072 (LA), No. 4710 (PO), Ripley & Barneby No. 4271 (RB); Garnet Station to Edom, Abrams No. 11091 (F); Garnet, J. T. Howell No. 3439 (CAS), Ripley & Barneby No. 4271a (RB) ; Indian Wells, Munz & Keck No. 4963 (PO); Indio, Ora Clark No. 1734 (LA) ; Paloverde, Hall No. 5958 (UC) ; Boulder Well, Schellenger No. 17 (NY). The type of 4. Coulteri, in the herbarium of Trinity College, Dublin, has not been studied, but Bentham’s admirable descrip- tion leaves little doubt as to the identity of Coulter’s plant. The indication of the type-locality as Monterey has a number of paral- lels, as remarked by Jepson (FI. Calif. 2: 350) : Coulter is known to have passed through the desert on his way to the Colorado River, and it was doubtless there that he first encountered Dalea mollis Bth., Phaca Crotalarie Bth., and A. Coulteri, all of which were erroneously labelled ‘“Monterey.” Since its publication the name A. Coulteri (or A. lentiginosus var. Coulteri) has almost universally been applied to wider concepts than are admitted here. Only Rydberg (sub Cystio), in the N. American Flora, and MARCH, 1945] PUGILLUS ASTRAGALORUM 131 Jepson (1936), have up to the present restricted it to the large- ' flowered, sericeous race endemic to the Colorado Desert, and of these two only Rydberg repudiated the long-established error of admitting as a synonym A. Arthu-Schottii Gray. This species, originally described from a mixture of var. borreganus and var. variabilis, and subsequently defined in the latter sense, was finally reduced by Gray to A. Coulteri, and this synonymy has persisted in the literature (cf. discussion of var. borreganus). This handsome variety is very closely related to var. variabilis, but differs essentially in the relatively longer calyx-teeth, larger flowers, and, for the most part, in its silvery pubescence. On the other hand certain collections of var. borreganus from Imperial County, California, reveal in their somewhat atypically inflated legume another close affinity. Jones’s statement (1923) that the var. Coulteri “is almost certainly a hybrid between lentiginosus and aridus” is completely without foundation. 34. var. BORREGANUS Jones Astragalus lentiginosus Dougl. var. borreganus Jones, Contrib. W. Bot. 8:3 (1898) ; Rev. Astrag. 126 (1923). A. Arthu-Schotttt Gray, Proc. Amer. Acad. Sci. 6: 209 (1863), pro max. parte, non alibi; Barneby, Proc. Calif. Acad. Sci. IV, 25: 158 (1944). A. agninus Jeps., Man. Calif. 577 (1925), Fl. Calif. 2: 355 (1936) ; Har- rison & Kearney, Journ. Wash. Acad. Sci. 22: 227 (1932) ; Munz, Man. S. Calif. 270 (1935); Kearn. & Peeb., Fl. Pl. Ariz. 487 (1942) ; Abrams, Ill. Fl. Pac. St. 2: 597 (1944). Cystium Arthu-Schottii (Gray) Rydb., N. Amer. Fl. 24: 407 (1929), saltem quoad typ. cit. ; C. agninum (Jeps.) Rydb., op. cit. 408. Pods lanceolate-acuminate to very narrowly and obliquely ovoid, fal- cate, subterete or a little obcompressed and shallowly sulcate ventrally, 1.5—2 cm. long, the valves membranous in texture, heavily strigose or seri- ceous, at times faintly mottled, the septum sometimes not quite reaching the ventral suture: racemes 15—35-flowered, lax and 6—12 (25) cm. long in fruit, on peduncles about equaling the leaf : calyx white-silky, villosulous or sericeous, often with some black hairs, the tube 3—3.5 mm., the teeth 0.75—1.5 mm. long: petals purple, the banner about 13 mm., the wings 10— 12.5 mm., the broad keel 9.5—11.5 mm. long: leaves 6—15 cm. long, with 7—9 pairs of obovate, obtuse or emarginate leaflets 6—15 mm. long, silky- canescent on either face: stems copiously white- or silvery-pubescent, erect, flexuous, 1.5—3 dm. high, stout or slender, often solitary, from an annual or perhaps rarely biennial root. Illustrations. P1. 1V, figs. 18—22. Jones, Rev. Astrag., pl. 25. Abrams, Ill. Fl., fig. 2869. 132 LEAFLETS OF WESTERN BOTANY [ VoL. IV, NO. 5 Type-locality. Borregos Springs, southeastern California, the type col- lected by Charles Orcutt. Distribution. Sandy desert valleys of the lower Colorado Desert in California, adjacent Arizona and Sonora, northwards along the Colorado River to the borders of Nevada. Exsiccata. CALIFORNIA or Sonora. Diluvial banks of the Colorado, Schott, Mex. Bound. Surv. 256 (G, type of A. Arthu-Schottii, NY) ; Colo- rado Desert near Fort Yuma, Schott in 1855 (NY). Sonora. Colorado Desert, Schott in 1855 (F); San Luis near Yuma, Harrison & Kearney, No. 8434 (F, SAC). Arizona. Yuma Co.: Harrison & Kearney No. 8431 (G), Beckett & Nobles No. 8695 (F, G, SAC) ; Yuma Desert, Nelson & Nelson No. 1288 (NY, UC); Yuma Mesa, Peebles & Harrison No. 5029 (SAC); Dome to Yuma, Kearney No. 3910 (SAC). Ca.irornia. San Diego Co.: Borrego Springs (or Wells), Orcutt (PO, type of var. borre- ganus), Jepson No. 8883 (NY, type-collection of A. agninus, fragments), Eastwood No. 2680 (G), Brandegee in 1895 (UC, erroneously annotated “type of var. borreganus” by Jones), Jones in 1906 (F, NY, PO, UC). Imperial Co.: Yuma Road at Ogilby turn-off, Johansen & Ewan No. 7142 (PO); dunes E. of Brawley, Jones in 1926 (PO). San Bernardino Co.: Silver Lake, J. T. Howell No. 3586 (CAS) ; Kelso, K. & T. S. Brandegee in 1915 (UC) ; Nipton, Ripley & Barneby No. 3360 (CAS, PO, RB). In a previous paper (I. c., 1944, sub A. Arthu-Schotti) I have attempted to give an account of the history of var. borre- ganus, and it will not be necessary to go over the same ground again. At that time I regarded the plant as a distinct species, but after further reflection I am persuaded that it is too closely re- lated to var. Coulteri and var. variabilis to be kept apart from the A. lentiginosus complex. There are so many examples in § Diplocystium of a race with bladdery-inflated legume passing gradually into a close relative with the legume scarcely more than turgid, that the degree of inflation can no longer be regarded as specifically diagnostic. Moreover, there are specimens such as Johansen & Ewan No. 7142 which, while doubtless best referred here on the dimensions of the flower and the obviously annual root, combine a certain degree of inflation in the pod reminiscent of var. Coulteri or var. variabilis with the membranous texture of the pod proper to var. borreganus. Jones’ statement, however, that “this shades directly into var. Coulteri” seems to be exagger- ated, while his reduction (Contrib. W. Bot. 15:69, and op. cit. 18: 44,—1933) of A. agninus to A. aridus Gray was an unlucky guess, prompted either by ignorance or malice. MARCH, 1945 ] PUGILLUS ASTRAGALORUM 133 35. var. uRSINUS (Gray) Barneby Astragalus lentiginosus Doug]. var. ursinus (Gray) Barneby, comb. nov. A. ursinus Gray, Proc. Amer. Acad. Sci. 13: 367 (1878) ; Sheld., Minn. Bot. Stud. 1: 147 (1894) ; Rydb., Fl. Rocky Mts. 494 (1917) ; Tidestr., Contrib. U. S. Nat. Herb. 25: 325 (1925) ; Tidestr. & Kitt., Fl. Ariz. & N. Mex. 213 (1941). A. lentiginosus var. mokiacensis Jones, Rev. Astrag. 126 (1929), pro parte, non A. moktacensis Gray. Tium ursinum (Gray) Rydb., N. Amer. Fl. 24: 398 (1929); Bull. Torr. Bot. Club 57: 408 (1931). Pods strongly ascending, lance-oblong, abruptly acuminate, slightly arcuate, subterete.and about 3 mm. in diameter, 10—13 mm. long, the valves leathery and glabrous, the septum produced only halfway across the cavity: racemes lax, 12—15-flowered, in fruit 4—6 cm. long, about equaling the peduncle: calyx sparingly strigose with black and white hairs, the tube 3—3.5 mm., the teeth about 0.75 mm. long : petals purple, the narrow banner 13—15 mm., the wings 11.5 mm., the keel 10 mm. long: leaves 6—9 cm. long, petiolate, with 7—8 pairs of obovate, obtuse or emarginate leaflets 5—11 mm. long, glabrous or with a few scattered hairs below: stems erect, flexuous, strigose, 1.5—3 dm. long. Illustration. P1. IV, figs. 31—33. : Type-locality. Bear Valley in the south-central part of Utah, the type collected by Palmer in 1877. Distribution. Known only from the type-locality. Exsiccata. Utan. Iron or Garfield Co.: Bear Valley, Palmer in 1877 (G, type, NY). The var. ursinus, although it has been more often confused with var. mokiacensis than with var. palans, is, nevertheless, at least as nearly related to the latter, and with further collections they may even prove to be confluent. As known at present only from the type, var. ursinus differs from var. palans in the smaller calyx with proportionately shorter teeth, as well as in the shorter, subterete and strongly ascending pod, characters all recognized by Rydberg in his revision of Tium (1931, 1. c.). It also re- sembles some of the depauperate forms of var. diphysus such as Peebles & Parker No. 14631, but the same criteria will serve to distinguish it. Astragalus ursinus was always treated by Jones as a synonym of A. mokiacensis, except for a brief period when he described under this name an unrelated plant later recognized as A. ensiformis Jones. Eventually the two binomials became so identified in his mind, that in his review of Tidestrom’s Flora of Utah and Nevada he actually criticized (Contrib. W. Bot. 15:21,—1929) the author for reporting A. ursinus from Utah. 134 LEAFLETS OF WESTERN BOTANY _ [VOL. IV, NO. 5 36. var. PALANS (Jones) Jones Astragalus lentiginosus Dougl. var. palans (Jones) Jones, Contrib. W. Bot. 8:4 (1898) ; Rev. Astrag. 125 (1923), pro max. parte, exclus. syn. A. palans Jones, Zoe 4: 37 (1893) ; Sheld., Minn. Bot. Stud. 1: 152 (1894) ; Rydb., Fl. Rocky Mts. 495 (1917); Tidestr. & Kitt., Fl. Ariz. & N. Mex. 213 (1941) ; Kearn. & Peeb., Fl. Pl. Ariz. 486 (1942). A. palans var. araneosus Jones, Proc. Calif. Acad. Sci. II, 5: 675 (1895), quoad num. cit., non A. araneosus Sheld. ' A. amplexus Pays., Bot. Gaz. 60:378 (1915). Hamosa amplexa (Pays.) Rydb., Fl. Rocky Mts. 496, 1063 (1917). Tium palans (Jones) Rydb., N. Amer. Fl. 24: 397 (1929) ; Bull. Torr. Bot. Club 57: 406 (1931). T. amplexum (Pays.) Rydb., op. cit. 1929, p. 398; 1931, lc. | Pods spreading or rarely ascending, lanceolate or narrowly ovoid- acuminate, falcate or nearly straight, slightly compressed or obcompressed, sulcate dorsally and often also ventrally, the valves coriaceous to firmly chartaceous, rugose or smooth, strigulose or commonly glabrous, mottled or stramineous, the septum sometimes incomplete at the center of the pod: racemes 10—20-flowered, typically rather lax but not uncommonly con- densed, 2—12 cm. long in fruit on peduncles 3—8 cm. long: calyx strigose with white or black hairs, the tube (4) 5—7 mm., the narrow, acute teeth 1.25—3 (4) mm. long: petals purple, the banner 15—17 mm., the wings 12—15.5 mm., the keel 11—14 mm. long: leaves shortly petiolate or the uppermost subsessile, 5—9 cm. long, with 6—9 pairs of obovate to elliptic leaflets 5—15 mm. long, glabrate on either face or sparsely pubescent below : stems spreading or erect, flexuous, strigillose to glabrate, 15—30 cm. long. Illustrations. Pl. IV, figs. 23—26. Jones, Rev. Astrag. pl. 25. Type-locality. Montezuma Canyon, Utah, the type collected by Alice Eastwood in 1892. Distribution. Along the Colorado River and its tributaries from the Grand Canyon to southern Utah and southwestern Colorado. Exsiccata. Cotorapo. Mesa Co.: Grand Junction, Osterhout No. 6142 (PO). Montrose Co.: Long Park near Naturita, Payson No. 335 (NY, isotype of A. amplexus). Utau. Grand Co.: Thompson’s Springs, Purpus No. 6476 (UC). San Juan Co.: Montezuma Canyon, Eastwood in 1892 (PO, type, CAS, G, UC) ; Bluff (or Bluffs), Rydberg & Garrett No. 9914 (NY), Jones in 1919 (PO), Cottam No. 2555 (F), Ripley & Barneby No. 5381 (RB); Goosenecks of the San Juan, Cronquist No. 1104 (UC). Emery Co.: San Rafael Swell, Jones in 1914 (PO). Washington Co.: Rockville, Jones No. 5215e (PO), No. 5218 (F, G, UC), No. 5218a (PO). Arizona. Coconino Co.: Grand Canyon, Macbride & Payson No. 945 (G) ; Grand View Trail, Grand Canyon, Eastwood No. 5748 (F, G). The var. palans is extremely variable in the length of the raceme, in the size of the flower and calyx, and in the develop- ment and curvature of the legume. The type has rather large corollas, lax racemes, and elongate pod curved into half a circle MARCH, 1945] - PUGILLUS ASTRAGALORUM 135 on recurved pedicels, but this passes insensibly into forms with patent or ascending pedicels, short and claw-like or longer and merely lunate pods, sulcate on either or both sutures, condensed or loose racemes, small or large flowers, and leaflets strigose below or glabrous. Astragalus amplexus was described from a rather more pubescent state, but Rydberg’s key characters of flower-size and pubescence do not hold. For example, in the Grand Canyon, plants otherwise similar may have either glabrous (Eastwood No. 5748 )or strigose (Macbride & Payson No. 945) legumes, while the dimensions of the flower given by Rydberg (1929, p. 386) in his key to Tium, viz. “corolla 13 mm. long” for T. palans and “16 mm. long” for T..amplexum, are immediately corrected in his descriptions of the species (op. cit., 397, 398) to “about 2 cm. long” and “2 cm. long” respectively, the latter being more nearly correct in either case. Although at first sight appearing quite distinct from A. lentigi- nosus, and so treated by all authorities except Jones, the var. palans is in reality closely allied to var. diphysus and var. ara- neosus. It is particularly suggestive to remark the manner in which var. diphysus of northern Arizona and eastwards and the pseudovicariant var. araneosus of Utah and Nevada are isolated from each other by a broad band of territory along the Colorado River, from the Grand Canyon nearly to its source, which is occupied to the exclusion of either by var. palans. Each of these varieties is widespread and fairly constant in the middle of its area, but as they approach the Colorado intermediates arise, and it is only by assuming an arbitrary dimensional limit to the in- flation of the legume that either can be separated from var. palans. That this limit has been vague in the past is evidenced by Jones’s treatment. At one point he identified his southern Utah col- lections of var. palans as A. araneosus, and reduced (1895, 1. c.) the latter to a variety of A. palans. Subsequently, however, he treated A. araneosus as an intergrade between vars. diphysus and palans, and similar material in his herbarium is promiscuously determined as either one or the other. 37. var. OROPEDII Barneby Astragalus lentiginosus Doug]. var. oropedii Barneby, var. nov., hinc leguminis glabri forma racemoque denso var. diphyso (Gray) Jones, illinc habitu, foliis glaucescentibus glabris ac caulibus elongatis var. Wilsons 136 LEAFLETS OF WESTERN BOTANY _ [VOL. IV, NO. 5 (Greene) Barneby comparanda, sed ab ea dentibus calycinis elongatis nec- non leguminis pericarpio membranaceo, ab hac legumine inflato breviori floribusque purpureis manifeste minoribus diversa. Pods spreading, obliquely ovoid, inflated, ventrally sulcate and with ventral suture prominently intruded, 1.3—2.5 cm. long, the valves mem- branous, glabrous, mottled, the nearly straight beak acuminate and uni- locular: racemes 10—20-flowered, dense in both flower and fruit, 1—3 cm. long, on peduncles shorter than or rarely equaling the subtending leaf, the peduncles usually widely divergent at maturity: calyx strigose with mixed black and white hairs, the tube 4.5—5.5 mm., the narrow acute teeth (2.5) 3—5 mm. long: petals purple, the banner (12) 13—15 mm., the wings about 12 mm., the keel 10—11 mm. long: leaves subsessile or very shortly petio- late, 5—10 cm. long, with 7—10 pairs of suborbicular, broadly oblong or ovate, obtuse to subacute leaflets 0.5—2 (2.5) cm. long, these glaucescent, flat and essentially glabrous, but sometimes (like the rachis) sparingly ciliate: stems rather slender, flexuous, glabrous except at the extremities, ascending or prostrate, (1) 2—3 dm. long. Type. Kaibab Trail to Roaring Springs, Grand Canyon National Park, Coconino County, Arizona, 22 September 1938, fruct., Eastwood & Howell No. 7064 (Herb. Calif. Acad. Sci. No. 262056). Also ibid., 23 June 1933, flor., Eastwood & Howell No. 1054 (Herb. Calif. Acad. Sci. No. 211208, cotype). Distribution. Kaibab Plateau, Arizona, particularly about the North Rim of the Grand Canyon. Exsiccata. Arizona. Coconino Co.: Kaibab Trail to Roaring Springs, Eastwood & Howell No. 7064 (CAS, type), 1054 (CAS, cotype) ; Cliff Spring, 1 mile from Cape Royal, Peirson No. 7428 (CAS); North Rim of Grand Canyon, Eastwood & Howell No. 930 (CAS, PO) ; Jacob’s Lake to Fredonia, Peebles No. 13059 (SAC). The var. oropedii is especially interesting in that it combines features on the one hand of vars. palans and Wilsonu, races characterized by a lanceolate, barely turgid legume, and on the other of vars. diphysus and latus, in which the pod is ovoid and strongly inflated. Having in mind the form of var. moktacensis with inflated pod, I was at first reluctant to recognize in var. oropedit more than an aberrant, perhaps hybrid, form of var. palans, but the sum of characters presented by the material now before me suggests that it should be recognized as a distinct and rather well-defined variety. In habit, and especially in the ample, glabrous foliage, relatively short peduncles, and long, flexuous stems, var. oropedii most closely resembles var. Wilsonii, which occupies a similarly restricted area in the same life-zone to the south, and which I take to be its nearest relative : the purple flower is almost that of var. palans, which occurs also in the Grand MARCH, 1945 ] PUGILLUS ASTRAGALORUM 137 Canyon, but at a lower elevation: while in size and contour of the pod it approaches, or at least recalls, var. diphysus and var. latus, from which the membranous texture of the valves, different facies, as well as the longer calyx-teeth (in the first case) and uni- locular beak of the pod (in the second) combine to distinguish it. The var. oropedu might also be confused, especially in im- mature specimens, with var. vitreus : the dimensions of the petals and calyx, however, are quite different, and it occupies a distinct area in the Transition Zone far removed in altitude, if not in longi- tudinal distance, from the badlands along the Virgin River to which var. vitreus seems to be confined. 38. var. MOKIACENSIS (Gray) Jones Astragalus lentiginosus Dougl. var. mokiacensts (Gray) Jones, Rev. Astrag. 126 (1923), pro max. parte, exclus. syn. cit. A. mokiacensis Gray, Proc. Amer. Acad. 13: 367 (1878); Sheld., Minn. Bot. Stud. 1: 147 (1894) ; Tidestr. & Kitt., Fl. Ariz. & N. Mex. 213 (1941), exclus. syn.; Kearn. & Peeb., Fl. Pl. Ariz. 486 (1942). Tium mokiacense (Gray) Rydb., N. Amer. F1. 24: 398 (1929) ; Bull. Torr. Bot. Club 57: 406 (1931). Pods strigose or glabrous, woody or leathery, typically lanceolate in outline and abruptly cuspidate, scarcely inflated, straight or curved through the fourth part of a circle, 2—2.5 cm. long, 5—6 mm. in diameter, the septum mostly incomplete (sometimes traversing only half the width of the cavity), or, atypically, inflated and rather narrowly and obliquely ovoid with triangular-acuminate beak, 1 cm. or more in diameter, the valves thinner and the septum complete : racemes 12—18-flowered, lax and 7—12 cm. long in fruit, on stout, stramineous peduncles of about the same length: calyx strigose with white and some black hairs, the tube (4.5) 5—7 mm., the teeth 1—2.5 mm. long: petals purple, drying violet, the banner (14) 15—19 mm., the wings (13) 14—17 mm., the keel (12) 14—17 mm. long: leaves 1—12 cm. long, with 6—9 pairs of elliptic or obovate emarginate leaflets 8—20 mm. long, glabrous on either face or commonly only above, and strigose with fine white hairs beneath: stems coarse, glabrate or canescent in youth, 2—5 dm. high, erect from a perennial root. Illustrations. Pl. IV, figs. 34—39. Jones, Rev. Astrag. pl. 25. Type-locality. Mokiak Pass on the borders of Utah and Arizona, the type collected by Palmer in 1877. Distribution. NW. Arizona and\ adjacent Nevada: also reported by Tidestrom and by Kearney & Peebles from SW. Utah. Exsiccata. Arizona. Mohave Co.: Mokiak Pass, Palmer No. 105 (G, type, NY, PO, fragments) ; Grand Canyon of the Colorado near Fort Mohave, Lemmon No. 3116 (G); Peach Springs, Lemmon No. 3326 138 LEAFLETS OF WESTERN BOTANY _ [VOL. IV, NO. 5 (G, PO, fragments, UC). Nevapa. Clark Co.: Mica Spring, Jones No. 5058 (NY, PO, UC). Form 8. Arizona. Mohave Co.: Pipe Spring, Jones No. 5272d (PO) ; Chloride, Jones in 1903 (PO); Peach Springs, Jones in 1904 (PO), Kearney & Peebles No. 11094 (PO, SAC), Ripley & Barneby No. 3430 (CAS, RB). Although A. mokiacensis was first described more than sixty years ago, it remains very little known, and the material available is scanty. The type-collection, which Rydberg (1931) errone- ously stated was without fruit notwithstanding Gray’s careful de- scription of the legume, bears strigose, lanceolate pods, shallowly sulcate ventrally and incompletely bilocular. This gathering is matched exactly only by the two obtained by Lemmon along the lower reaches of the Grand Canyon. Jones’ material from Mica Spring, Nevada, is almost exactly similar, but has glabrous legumes and slightly shorter calyx. These were the only speci- mens of Tium mokiacense known to Rydberg in 1931, and they are still the only ones which really match the type. On the plateau around Peach Springs, however, occurs a plant indistinguish- able from var. mokiacensis in habit, but with ovoid, inflated, com- pletely bilocular pod and slightly smaller flower. The material belonging to this form has been enumerated above as form 8, and the dimensions of the legume have been included in the description of the variety. Jones regarded it as var. yuccanus “intergrading towards var. mokiacensis”: in certain respects it is intermediate between the vars. mokiacensis and yuccanus and may well represent hybridization between the two, but until it is better known it may be disposed of in an inconclusive manner. The variety in its strict sense is very closely related to var. palans, var. Wilsonii, and var. ursinus, and all four should per- haps be united. The flowers, however, turn violet in the her- barium, rather than the reddish-purple of var. palans, and are usually larger and more loosely racemose : the peduncles are much longer and the petals more brightly colored than in var. Wilsoni: while both flowers, pods, and particularly the calyces, are all conspicuously longer than in var. ursinus. The literature dealing with var. mokiacensis is fairly straight- forward, except for the fact that Jones always included in it A. ursinus Gray. Rydberg has suggested that this error arose from Gray’s misdetermination of the Lemmon collections as / MARCH, 1945] PUGILLUS ASTRAGALORUM 139 A. ursinus. This may be so, but it scarcely explains Jones’ state- ment (Contrib. W. Bot. 15:21,—1929) that “A. ursinus .. . actually grows only in Arizona bordering the Grand Canyon,” when the type of that species came from Utah. 39. var. WiLtsonm (Greene) Barneby Astragalus lentiginosus Dougl. var. Wilsonii (Greene) Barneby, comb. nov. A. Wilsonii Greene, Pitt. 3: 196 (1897) ; Kearn. & Peeb., Fl. Pl. Ariz. 486 (1942). ‘Tium Wilsoni (Greene) Rydb., N. Amer. FI. 24: 398 (1929) ; Bull. Torr. Bot. Club 57: 406 (1931). Astragalus lentiginosus var. mokiacensis Jones, Rev. Astrag. 126 (1923), pro parte, e syn. cit., non A. mokiacensis Gray. A. mokiacensis Tidestr. & Kitt., Fl. Ariz. & N. Mex. 213 (1941), pro max. parte, e syn. & descr., non Gray. Pods strongly ascending, lanceolate and gradually acuminate, slightly upcurved, subterete and either ventrally or dorsally sulcate, 2.5—3 cm. long, 4—6 mm. in diameter, coriaceous, smooth, glabrous or rarely strigulose, commonly mottled: racemes 7—12-flowered, rather dense even in fruit, on stout peduncles a little shorter than the leaf and 4—7 cm. long: calyx stri- gose with white or a few black hairs, the tube 6—6.5 mm., the teeth 1.5—3 mm. long: petals white or pinkish, the banner 16—19 mm., the wings about 15 mm., the keel about 13 mm. long: leaves subsessile or the lower shortly petiolate, 8—11 cm. long, with 8—12 pairs of rather thin, ovate, elliptic or obovate, obtuse leaflets 8—25 mm. long, glabrous on either face: stems diffuse or prostrate, flexuous, nearly glabrous, 2.5—5 dm. long. Illustration. Pl. IV, figs. 27—30. Type-locality. “Northern Arizona,” actually near Flagstaff, the type collected by Norman C. Wilson in 1893. Distribution. In open pinewoods of the Mogollon Rim in Coconino and Yavapai counties, Arizona. Exsiccata. Arizona. Coconino Co.: near Flagstaff, Wilson in 1893 (ND, type), MacDougal No. 140 (F, G, NY, PO, UC) ; Oak Creek, Rusby in 1883 (NY) ; Walnut Canyon, Purpus No. 23 (UC) ; Cosnino near Flag- staff, Jones in 1884 (PO). Yavapai Co.: N. of Cottonwood, Whiting No. 1058 (SAC). The var. Wilsonu has been confused by Jones and Tidestrom with the preceding, which it resembles in the strongly ascending pod, and of which it may be a montane derivative, but its diffuse stems, pale flowers, and more condensed racemes are conspicu- ously different. In gross habit it is quite similar to var. oropedii from the pinewoods of the Kaibab Plateau to the north, but in 140 LEAFLETS OF WESTERN BOTANY [ VOL. IV, NO. 5 that race the legume is spreading, inflated, and membranous. Kearney & Peebles (1. c.) doubtfully included in their 4. Wilsoniu some material from Maricopa County, Arizona, treated below as a new variety. 40. var. MARICOP2 Barneby Astragalus lentiginosus Dougl. var. maricopz Barneby, var. nov., legumine lineari-lanceolato coriaceo vix obliquo adscendenti var. Wilsonii (Greene) nob. manifeste affinis, sed caulibus strictioribus minus flexuosis, racemo plurifloro plerumque laxissimo, necnon corolla ochroleuca breviori diversa.—A. Wilsonii Kearn. & Peeb., Fl. Pl. Ariz. 486 (1942), pro parte. Pods as in the preceding, about 2 cm. long, castaneous and not mottled: racemes (9) 15—20-flowered, when well developed 10—20 cm. long and lax in fruit, rarely shorter and 3—5 cm. long, on stout and rather strict peduncles 5—9 cm. long, shorter than the subtending leaf: calyx stri- gose with white or some black hairs, the tube 5—6 mm., the teeth 1.5—2 mm. long: petals ochroleucous, the banner 13—15 mm., the wings 12—13 mm., the keel 11—11.5 mm. long: leaves shortly petiolate, 7—15 cm. long, with 7—11 pairs of rather thick, subglaucescent, obovate to broadly elliptic, obtuse, truncate (rarely apiculate) leaflets 7—22 mm. long, essentially gla- brous or with a few scattered hairs beneath when young : stems mostly erect, stout, not flexuous, 3—4 dm. high, glabrous. Type. Roadside near Tempe, Maricopa County, Arizona, G. J. Harrison No. 1790 (U. S. National Herbarium No. 1870814). Distribution. Lower reaches of the Salt River, Arizona. Exsiccata. Ar1IzoNA. Maricopa Co.: near Tempe, Harrison No. 1790 (U. S. Nat. Herb.) ; Cave Creek Dam, Peebles, Harrison & Kearney No. 3684 (SAC) ; Fish Creek, Peebles, Harrison & Kearney No. 5238 (SAC). In the shape and texture of the legume the var. maricope resembles the preceding very closely, but the long open raceme of smaller ochroleucous flowers, as well as the stout, erect stems distinguish it immediately. The gathering from Fish Creek, how- ever, shows some transition toward var. Wilsonii in general habit, though still differing in the size and color of the corolla. When in flower alone it might be taken for a very glabrous form of var. yuccanus Jones, thereby revealing another of the manifold cross- relationships between the races of 4. lentiginosus with strongly inflated and uninflated pods. 1 Material of var. maricope, kindly communicated by Mr. Peebles from the herbarium of the U. S. Field Station, Sacaton, Arizona, was received after completion of the plates, and it has been found impossible to insert an illustration. MARCH, 1945] PUGILLUS ASTRAGALORUM 14! ExcLtupEp NAMES AND SPECIES ASTRAGALUS BAJAENSIS Sheld., Minn. Bot. Stud. 1: 169 (1894). This species was erroneously described as having bilocu- lar pods, and was placed by the author in his § Lentiginosus be- tween A. latus Jones and A. Fremontu Gray. As observed by Jones (Contrib. W. Bot. 10: 62,—1902), the plant is closely allied to A. Hornu Gray, and the legume, though inflated, is wholly devoid of septum. ASTRAGALUS BOISEANUS A. Nels., Bot. Gaz. 53: 233 (1912). Cystium boiseanum (A. Nels.) Rydb., Bull. Torr. Bot. Club 40:50 (1913). A. eremiticus var. spencianus Jones. Astragalus boiseanus, a plant with stipitate pods and closely related to A. eremiticus Sheld., was at one time referred by Rydberg to his genus Cystium. Ina subsequent paper Rydberg (Bull. Torr. Bot. Club 57:400,—1931) explained that he mistook A. boiseanus for that race of A. lentiginosus for which he proposed instead the name Cystium heliophilum, but which he eventually recognized as A. salinus Howell (A. lentiginosus var. salinus (Howell) Barneby ). ASTRAGALUS BryantTiI Barneby, Proc. Calif. Acad. Sci. IV, 25: 156, pl. 17, figs. 1O—18 (1944). In the original publication it was suggested that A. Bryant was most closely related to A. mokiacensts Gray (var. mokiacensis (Gray ) Jones), an opinion which may or may not be correct. The pod, however, is strongly compressed, a condition not met with elsewhere in § Diplocystium, and until the species is better known I prefer to exclude it from the section. ASTRAGALUS CUSPIDOCARPUS Sheld., Minn. Bot. Stud. 1 : 147 (1894). A. lentiginosus var. cuspidocarpus (Sheld.) Jones, quoad nomen. dA. missouriensis var. cuspidocarpus (Sheld.) Jones. Xylophacos cuspidocarpus (Sheld.) Rydb. The cotype of this species, Leiberg No. 171, is a collection of A. lentiginosus var. platyphyllidius (Rydb.) Peck, a part of which served as the type of Cystium platyphyllidium Rydb., but the typical element of the species bears a unilocular pod similar to that of A. cibarius Sheld. Jones’s combination A. lentiginosus var. cuspidocarpus referred only to the Leiberg element, to which the name cannot be validly applied. This point is more fully dealt with in the text under var. carinatus and var. platyphyllidius. 142 | LEAFLETS OF WESTERN BOTANY _ [VOL. IV, NO. 5 ASTRAGALUS DIAPHANUS Dougl. ex Hook., Fl. Bor.-Amer. 1:151 (1831). A. drepanolobus Gray, Proc. Amer. Acad. 19:75 (1883). A. lentiginosus var. diaphanus (Dougl.) Jones, Proc. Calif. Acad. Sci. II, 5:675 (1895); Rev. Astrag. 123, pl. 23 (1923). ren, Astragalus diaphanus Dougl., partly on account of the in- adequate description and partly because of the lack of authentic material in this country, has always remained obscure to Ameri- can botanists. Gray, to whom the plant was unknown at the time of his Revision in 1863, having studied the type in England in 1868, reduced the species (Bot. Calif. 1: 147,—1880) to A. lenti- ginosus, a disposition followed by Watson and others. In 1894, Sheldon (Minn. Bot. Stud. 1: 168) revived the name, placing it without comment in his § Lentiginosus. In the following year, Jones (1895, 1. c.), without further investigation of the facts, reduced A. diaphanus to varietal status under A. lentiginosus, only to submerge it altogether (Contrib. W. Bot. 8: 3,—1898) with that species, and finally, in the Revision of Astragalus, to reinstate it as a variety. In 1927, Rydberg (Bull. Torr. Bot. Club 54:21) suggested that in reality A. diaphanus, far from being a form of A. lentiginosus, was an earlier synonym of the very different 4. drepanolobus Gray; not having seen the type, however, he prudently placed the name as a doubtful synonym of Hamosa drepanoloba (Gray) Rydb., and it was treated thus in the North American Flora. In response to my enquiries for information, Mr. N. Y. Sand- with of the Royal Herbarium at Kew kindly undertook to study and report on the type of A. diaphanus which is preserved there among the Hooker collections, and after careful consideration of the literature and comparison of the Douglas specimens of the species both with the type of A. lentiginosus and an isotype of A. drepanolobus, he concluded that Rydberg’s supposition was indeed correct, and that all but a small part of the type of A. di- aphanus belongs to A. drepanolobus. From Mr. Sandwith’s copious notes, a copy of which has been’deposited in the Her- barium of the California Academy of Sciences, it appears that the type-sheet of A. diaphanus contains three elements, which were separated and annotated by Gray. One of these, in flower only and not closely agreeing with Hooker’s description, was correctly MARCH, 1945] PUGILLUS ASTRAGALORUM 143 identified by Gray as A. lentiginosus and matched with a fruiting specimen of the same collected in Oregon by Burke. The second, also in flower, and the third, from which the description of the pod must have been prepared (and hence best regarded as the actual type), are both A. drepanolobus. Against the fruiting specimen Gray wrote in pencil, “Not identified. May keep the name,” but in the twelve years which elapsed between his study of the type of A. diaphanus and his reduction of it to A. lentigt- nosus, his analysis of the elements seems to have been forgotten. In the Gray Herbarium there are preserved two fragments under the name A. diaphanus, one being half of a pod from the Douglas collection, the other a small fruiting branch taken from the Burke gathering mentioned above. It seems evident that Gray’s concept of A. diaphanus had gradually been transferred from the fruiting to the flowering element of the type-sheet, and thence to the quite different material from Oregon, thereby eliminating the essential A. diaphanus in favor of A. lentiginosus. ASTRAGALUS LENTIGINOSUS Var. CUSPIDOCARPUS Jones. Cf. A. cuspidocarpus. ASTRAGALUS LENTIGINOSUS Var, DIAPHANUS Jones. Cf. A. di- aphanus. ASTRAGALUS PeErIrRson1II Munz & McBurney, Bull. S. Calif. Acad. Sci. 31:67 (1932). This species, like A. bajaensis above, was described as an ally of A. Coulteri Bth., on the supposition that the pod was bilocular. In reality the septum is lacking, and the affinities of the species lie with the series Candidissim@ of Rydberg’s Phaca. Jones (Contrib. W. Bot. 18: 37,—1933) re- duced A. Peirsonii to A. Coulteri, although it does not closely resemble that species even superficially. ASTRAGALUS PLATYTROPIS Gray, Proc. Amer. Acad. 6: 526 (1865). Cystium platytrope (Gray) Rydb., Bull. Torr. Bot. Club 40: 50(1913). This species was described as a relative of A. lenti- ginosus, and was referred by Rydberg to Cystinm, although Jones had already suggested that its true place in the genus was else- where. Ina previous note (Proc. Calif. Acad. Sci. IV, 25 : 166,— 1944) I have attempted to show that it differs from the rest of § Diplocystium in several important morphological particulars, and have proposed for it a separate section, § Orocystium. 144 LEAFLETS OF WESTERN BOTANY __ [VOL. IV, NO. 5 AstRAGALUS RotHrocki Sheld., Minn. Bot. Stud. 1:174 (1894). At one period Jones (Proc. Calif. Acad. Sci. II, 5:673,— 1895) reduced this species to A. MacDougaliu Sheld., but later (Rev. Astrag. 156,—1923) recognized its true relationship to A. prelongus Sheld., from which it appears at best to be not more than varietally distinct. CysTIUM BOISEANUM (A. Nels.) Rydb. Cf. Astragalus boise- anus. CysTIUM PLATYTROPE (Gray) Rydb. Cf. Astragalus platy- tropis. INDEX TO SYNONYMY (New names appear in bold-face type, synonyms in italic type. Numbers correspond with the numbered varieties of A. lentt- ginosus, those in bold-face type referring to the essential entry in the text. ES stands for Excluded Species.) Astragalus lentiginosus 1, 6, 8, 11, 16. agninus 34. var. albifolius 9. albifolius 9. var. antonius 13. amplexus 36. var. araneosus 22. araneosus 16, 20, 22. var. australis 25. Arthu-Schotttu 30, 34. var. borreganus 34. bajaensis ES. var. cesariatus 32. boiseanus ES. var. carinatus 2. Bryantii ES. var. charlestonensis 14. Coulteri 7, 29, 30, 33. var. chartaceus 21. var. Fremontti 7. var. cornutus 16. cuspidocarpus 2, 16, ES. var. Coulteri 7, 30, 33. diaphanus ES. var. cuspidocarpus 2, 16, 18, ES. diphysus 5, 17, 23, 26. var. diaphanus ES. var. albiflorus 23. var. diphysus 16, 21, 22, 23. var. latus 20. var. floribundus 5, 8, 11. drepanolobus ES. var. Fremontii 6, 7,11, 13, 30, 31. eremicus 7. var. idriensis 18. Fremonttt 7, 11, 19, 29, 31. var. ineptus 5, 8, 11. subsp. eremicus 7. var. Kennedyi 28. var. yuccanus 26. var. kernensis 15. heliophilus 5. . var. latus 5, 15, 20. tdriensis 18, 19. var. MacDougali 23. ineptus 5, 11, 16. var. macrolobus 6. kernensis 15. var. maricopz 40. subsp. charlestonensis 14. var. mokiacensis 35, 38. latus 20. var. nigricalycis 19, 31. MARCH, 1945] PUGILLUS ASTRAGALORUM var. oropedii 37. var. palans 22, 25, 36. var. platyphyllidius 16. var. salinus 5. var. scorpionis 1, 3. yar. semotus 10. var. sesquimetralis. 24. var. sierre 12, 19. var. stramineus 29. var. tehatchapiensis 19. var. toyabensis 17. var. tremuletorum 4. var. typicus 1. var. ursinus 35. var. variabilis 30. var. vitreus 27. var. Wilsonii 39. var. yuccanus 26. MacDougali 23. Merrillii 16. missouriensis var. cuspidocarpus ES mokiacensis 38. nigricalycis 31. palans 36. var. araneosus 22, 36. Peirsonii ES. platytropis ES. Rothrockii ES. salinus 5. sterre 12. tehatchapiensis 19. ursinus 35. Wilsonii 39. yuccanus 26. Cystium agninum 34. albifolium 9. araneosum 22. Arthu-Schottii 30, 34. boiseanum ES. cornutum 2, 16. Coulteri 29, 33. diphysum 22, 23. eremicum 7. floribundum 8. ’ Fremonti 7. griseolum 7. heliophilum 5. idriense 18. meptum 5, 11. Kennedyi 28. kernense 15. latum 20. lentiginosum 1, 2. MacDougali 23. macrolobum 6. Merrillii 16. nigricalyce 31. ormsbyense 8. pardalotum 30. platyphyllidium 16, 17. platytrope ES. salinum 1, 5, 21. scorptonis 3. sesquimetrale 24. sierr@ 12, stramineum 29. tehatchapiense 19. vulpinum 6. yuccanum 26, Hamosa amplexa 36. Phaca inepta 5, 11. lentiginosa 1. Tium amplexum 36. mokiacense 38. palans 36. ursinum 35. Wilsontt 39. Tragacantha Coulteri 33. diphysa 23. lentiginosa 1. 145 146 LEAFLETS OF WESTERN BOTANY __ [VOL. IV, NO. 5 FIGURES OF ASTRAGALUS LENTIGINOBUS (Leaves and racemes x %. Pods, flowers, and sections x 1) Prats I Fig. 1—5. Var. typicus Barneby. 1, leaf and raceme; 2 and 3, vertical section of pod; 4, cross section of pod; 5, flower. Fig. 6--12. Var. carinatus Jones. 6, leaf and raceme; 7, vertical section of pod; 8, cross section: 9, pod; 10 and 11, flowers; 12, leaflet. All from the type except figures 9 and 10. Fig. 13—17. Var. tremuletorum Barneby. 13, leaf and raceme; 14 and 17, vertical section of pod; 15, cross section; 16, flower. All from the type except figure 17. Fig. 18-—24. Var. scorpionis Jones. 18 and 19, leaf and raceme; 20 and 22. vertical section of pod; 21 and 24, flowers; 23, stipulesx 1%. Figures 18 and 24 from the type. Fig. 25—29. Var. albifolius Jones. 25, leaf and raceme; 26, pod; 27, vertical section of pod; 28, cross section; 29, flower (from type). Fig. 30—36. Var. floribundus Gray. 30, leaf and raceme; 31 and $4, flowers; 32, leaf; 33, pod; 35, vertical section of pod; 36, cross section. Figures 35 and 36 from the type; figures 33 and 34 from the type of Cystium ormsbyense. Fig. 37—41. Var. macrolobus (Rydb.) Barneby. 37, leaf and raceme; 38. vertical section of pod ; 39, cross section ; 40, flower (from type of Cystium vulpinun) ; 41, flower (from type of C. macrolobum). Fig. 42—45. Var. salinus (Howell) Barneby. 42, leaf and raceme; 43, section of pod; 44 and 45, flowers. . Fig. 46—50. Var. Fremontii (Gray) Wats. 46 and 47, leaf and raceme; 48, vertical section of pod; 49 and 50, flowers. Figure 50 from the type of C. agriseolum. ~ PLATE II Fig. 1—4. Var. ineptus (Gray) Jones. 1, leaf and raceme; 2, vertical section of pod; 3, cross section; 4, flower. Fig. 5 and 6. Var. semotus Jeps. 5, leaf and raceme; 6, flower. Fig. 7—9. Var. antonius Barneby. 7, leaf and raceme; 8, vertical section of pod; 9, flower. All from the type. Fig. 10—12. Var. sierre Jones. 10, leaf and raceme; 11, vertical section of pod; 12, flower. All from the type. Fig. 13—15. Var. charlestonensis (Clokey) Barneby. 13, leaf and raceme ; 14, vertical section of pod; 15, flower. All from the type. Fig. 16—18. Var. kernensis (Jeps.) Barneby. 16, leaf and raceme; 17, vertical sect'on of pod ; 18, fiower. Fig. 19—23. Var. idriensis Jones. 19, leaf and raceme; 20, pod; 21, cross section; 22, flower; 238, vertical section of pod. All from the type except figure 23. Fig. 24—29. Var. tehatchapiensis (Rydb.) Barneby. 24, leaf and raceme ; 25, vertical section of pod; 26, pod; 27, cross section of pod; 28 and 24, flowers. Figure 29 from the type. Fig. 30—34. Var. latus Jones. 20, leaf and raceme; 31, vertical section ot pod; 32, cross section of pod ; 33, pod; 34, flower. All from the type except figure 33. Fig. 35. Var. chartaceus Jones. 35, vertical section of pod (from the type). Fig. 36—28. Var. sesquimetralis (Rydb.) Barneby. 36, leaf and raceme ; 37, vertical section of pod; 38, flower. All from the type. Fig. 39-46. Var. platyphyllidiuws (Rydb.) Peck. 39, leaf and raceme; 4(0—42, vertical section of pod ; 43 and 44, flowers; 45 and 46, calyx. Figures 24 and 45 from the type; figures 40 and 43 from the type of Cystiuimn cornu- tum; figures 41 and 44 from the type of C. Merrillii. (Baplanation of Plates ITT and IV on page 147.) PLATE I. plate I Pirate Il. pate I. | 7? MARCH, 1945 | PUGILLUS ASTRAGALORUM 147 FIGURES OF ASTRAGALUS LENTIGINOSUS (Leaves and racemes x 1%. Pods, flowers, and sections x 1) PLATE III Fig. 1—4. Var. toyabensis Barneby. 1, vertical section of pod; 2, cross section; 3, flower; 4, leaflet. All from the type. Fig. 5—10. Var. diphysus (Gray) Jones. 5, leaf and raceme ; 6—8, verti- cal section of pod; 9, cross section of pod; 10, flower. Figures 5 and 10 froin the type. Fig. 11—14. Var. araneosus (Sheld.) Barneby. 11, leaf and raceme (from the type) ; 12, vertical section of pod; 13, pod; 14, flower. Fig. 15—19. Var. australis Barneby. 15, leaf and raceme; 16, vertical section of pod; 17, cross section; 18, leaflet; 19, flower. Figures 16 and 17 from the type. Fig. 20—25. Var. Kennedyi (Rydb.) Barneby. 20, leaf and raceme; 21, vertical section of pod ; 22, cross section ; 23 and 25, flower ; 24, lower surface of leafiet. All from type except figures 20 and 25. Fig. 26—29. Var. stramineus (Rydb.) Barneby. 26, leaf and raceme; 27, cross section of pod; 28, pod (from the type) ; 29, flower. Fig. 30—33. Var. vitreus Barneby. 30, leaf and raceme; 31, cross section of pod; 32, pod; 33, flower. Figures 30 and 31 from the type. Fig. 34—37. Var. yuccanus Jones. 34, leaf and raceme (from the type) ; 35, pod; 36, flower; 37, lower surface of leaflet. PLaTEe IV Fig. 1—8. Var. variabilis Barneby. 1, leaf and raceme; 2 and 3, vertical section of pod ; 4—7, flowers ; 8, calyx. Figure 1 from the type; figure 5 from the type of Cystium pardalotum. Fig. 9—11. Var. cesariatus Barneby. 9, leaf and raceme; 10, cross sec- tion of pod; 11, flower. All from the type. Fig. 12 and 13. Var. nigricalycis Jones. 12, leaf and raceme; 13, flower (from the type). Fig. 14—17. Var. Coulteri (Bth.) Jones. 14, leaf and raceme; 15, flower; 16, vertical section of pod ; 17, cross section. Fig. 18—22. Var. borreganus Jones. 18, leaf and raceme; 19 and 20, vertical section of pod; 21, cross section; 22, flower. Figure 19 from the type of A. agninus ; figure 22 from the type of A. Arthu-Schottii. Fig. 23—26. Var. palais (Jones) Jones. 23, leaf and raceme; 24 and 25, vertical section of pod; 26, flower. Figures 23 and 26 from the type. Fig. 27—30. Var. Wilsonii (Greene) Barneby. 27, leaf and raceme; 28, vertical section ot pod; 29, cross section; 30, flower. All from the type except figure 27. Fig. 21—33. Var. wrsinws (Gray) Barneby. 31, leaf and raceme; 32, verti- cal section of pod; 33, flower. All from the type. Fig. 34—39. Var. mokiacensis (Gray) Jones. 34, leaf and raceme; 35, pod; 36 and 37, cross section ; 38, flower (from the type) ; 39, vertical section of pod. 148 LEAFLETS OF WESTERN BOTANY _ [VOL. IV, NO. 5 NEW WESTERN PLANTS—VI BY ALICE EASTWOOD Some time before his death, A. A. Heller sent to the Her- barium of the California Academy of Sciences two manzanitas to _ which he had given specific names which he had planned to pub- lish in this journal. One of these from Amador County I have taken the liberty of naming in his honor; the other retains the characteristic specific name on his label. Arctostaphy!os Helleri Eastwood, spec. noy. Procumbens (?), omnino glabra, ramis gracilibus; foliis oblongis vel suborbiculatis, circa 15 mm. longis, 8—10 mm. latis, apice acutis obtusisve et mucronatis, basi obtusis ; petiolis gracilibus, 3—4 mm. longis; floribus roseis, in racemis vel paniculis parce ramosis, pedicellis erectis, circa 2 mm. longis; bracteis subulatis, minutis, circa 1 mm. longis; sepalis 5, orbiculatis, minute ciliatis, circa 0.75 mm. diametro; corolla circa 3 mm. longa; staminibus 5 mm. longis, antheris et filamentis equilongis, lata basi filamentorum ciliata ;-ovario glabro, 1 mm. ‘ longo; fructu ignoto. Type: Herb. Calif. Acad. Sci. No. 300047, collected by A. A. Heller, No. 15859, Jan. 29, 1941, about three miles southeast of Ione, Amador County, California, in a very dense growth of other manzanitas, especially A. viscida Parry. It is related to the group ~ which includes A. Hookeri G. Don, A. franciscana Eastwood, and A. densiflora M. S. Baker. The flowers are smaller than those of any other species of this genus. The leaves are dull and rather thick. Without the fruit it cannot be definitely placed in this small group and its habit is uncertain. Arctostaphylos laxiflora Heller, spec. nov. Erecta, divaricate ramosa, glabra; foliis oblongis vel orbiculatis apice obtusis et mucronulatis vel paulum emarginatis, pallido-viridibus vix glaucis, maximis 40 mm. longis, 30 mm. latis, basi obtusis, petiolis circa 1 cm. longis; floribus roseis in pani- culis laxifloris glabris, divaricate ramosis, ramis aliquis 1 dm. longis; pedi- cellis filiformibus, 2—4 mm. longis; bracteis ovatis acuminatis brevioribus pedicellis ; sepalis orbiculatis glabris, circa 2 mm. diametro; corolla circa 9 mm. longa; antheris circa 1 mm. longis, brevioribus filamentis et cornibus, basi lata filamentorum villosa; ovario glabro; fructu ignoto. Type: Herb. Calif. Acad. Sci. No. 279926, collected by A. A. Heller, No. 15493, February 6, 1940, east of Paradise, Butte County, California, on Neal Road near the junction of the first road east of the Bille Road, at an old sawmill site in the Transition Life Zone. This belongs near A. Manzanita Parry of which it MARCH, 1945 | NEW WESTERN PLANTS | 149 may be a variation. It differs in the smooth pale green leaves and the laxly flowered, long-branched, smooth panicles. Arctostaphylos nitens Eastwood, spec. nov. Fruticosa, divaricate ramosa; caulibus senioribus glabris, fuliginosis asperisque, caulibus juniori- bus glanduloso-villosis ; foliis viridissimis, oblongis vel oblongo-ovatis, apice acutis, basi obtusis truncatisve, maximis circa 3 cm. longis, 10—12 mm. latis, coriaceis, supra nitentibus et minute reticulatis, infra leviter et minute glanduloso-puberulis, petiolis circa 5 mm. longis, planis, costatis, glanduloso- puberulis et leviter glanduloso-villosis ; paniculis nutantibus, subsessilibus ; bracteis foliaceis, nitentibus, aliquando puberulis et villosis, lineari-lance- olatis, infimis 15 mm. longis, longioribus supremis, pedicellis glabris, brevi- oribus bracteis; sepalis orbiculatis glabris; corollis 7 mm. longis, pallide roseis vel albis; ovario leviter villoso. _ Type: Herb. Calif. Acad. Sci. No. 315577, collected April 9, 1941, along the roadbank near top of hill south of bridge over Chetco River, Curry County in the southwest corner of Oregon, by Milo S. Baker, No. 9938. This manzanita is at once noticeable among the foliaceous- bracted group by the glossy very green leaves which become brittle when dry. The reticulation is fine and conspicuous. In old leaves it is even somewhat rugose on the upper surface. The lower sur- face appears smooth with the glandular indument apparent under alens. The bark surface of old branches is not smooth and glossy as is general in this genus but is rough but not shreddy and in color a dull blackish-brown. The glandular hairs and underlying puber- _ ulence are more or less evident on the young stems, petioles, and inflorescence. In my key to the genus (Leafl. West. Bot. 1: 108- 113), this species comes next to A. intricata Howell from which it differs in the bright green and glossy leaves, the flowers 7 mm. long, and the indument glandular-puberulent and somewhat glandular-hairy. Lupinus Noldekz Eastwood, spec. noy. Caules perennes, supra ramosi, graciles, tenuiter adpresse sericei; foliolis 5 vel 6, linearibus, conduplicatis, maximis circa 3 cm. longis, 1—2 mm. latis, acutis tenuiter adpresse sericeis ; petiolis gracilibus zequilongis vel paululum longioribus foliolis; stipulis dis- cretis, attenuatis, 2—3 mm. longis ; racemis superantibus folia et longioribus pedunculis, laxe et subverticillate floridis; bracteis plerumque persistenti- bus, zquilongis pedicellis, gracilibus, attcnuatis; floribus circa 11—13 mm. longis, albis; tubo calycis 3 mm. longo, 2 mm. lato, calcare 2 mm. longo, labio inferiore deltoideo, obtuso, 3 mm. longo, circa 2 mm. lato, labio superi- ore 1 mm. longo latoque, truncato sed dentibus duobus separatis; vexillo albo, brunneo-maculato, obovato, 10 mm. lato, 13 mm. longo, dorso Sericeo, reflexo ; alis sericeis prope apicem ; carina ciliata, tecta alis. 150 LEAFLETS OF WESTERN BOTANY _ [VOL. IV, NO. 5 Type: Herb. Calif. Acad. Sci. No. 280932, collected July, 1938, by Anita Noldeke, near Hot Creek, Mono County, Cali- fornia. Miss Noldeke notes that the flowers are “very fragrant and delicate.” This dainty white-flowered lupine is related to L. calcaratus Kellogg by the shape and color of the flower, but in this the flower is more slender and the inflorescence more open. The very narrow leaflets, slender branching habit, and denser indument mark it as distinct. The Herbarium of the California Academy of Sciences is indebted to Miss Noldeke for many specimens from that little- known region of Mono County. It is a pleasure to name this lupine in her honor. STUDIES IN PHACELIA—IV BY JOHN THOMAS HOWELL CoNCERNING THE Type Locality oF PHACELIA LEIBERGII. One of the species in Phacelia sect. Microgenetes that has not been generally accepted is P. Leibergit Brand (Das Pflanzenr. IV. 251: 128) but in my revisional study of the section I have found it to represent a distinct specific unit intermediate between P. bicolor Torr. and P. glandulifera Piper. From an excellent series of speci- mens loaned to me by Prof. M. E. Peck I was able to work out the distribution of the species as it occurs in central and middle eastern Oregon but I was puzzled by the collection data accompanying the type specimen which was loaned to me from Rocky Mountain Herbarium by Prof. C. L. Porter. The type, Leiberg No. 321, as well as numerous isotypes, is labelled as follows: “Plain bet. [between] Pineville and Bear Buttes, 1110 m.,” Washington, June 25, 1894, but neither Pineville nor Bear Buttes could be located in that state in an attempt to fix the region of the type locality. The puzzle resolved itself at last when I recently bor- rowed, through Dr. J. M. Greenman, the Missouri Botanical Gar- den specimens in sect. Microgenetes. Among the specimens of P. Leibergii is one collected by Leiberg on June 25, 1894, on “Sandy plains, alt. 1110 m., between Prineville and Bear Butte, Crook County, Oregon.” Without doubt this specimen is part of the type collection and furnishes the correct data. Another speci- men which is probably part of the type collection of P. Leibergit MARCH, 1945 | STUDIES IN PHACELIA 151 is in the Washington State College Herbarium loaned to me by Prof. Marion Ownbey, but in this there is a slight variation in the data which read: ‘Pumice desert south of Prineville, June, 1894.” From those two specimens it seems clearly evident that the type collection of P. Letbergit was not made in Washington as origin- ally stated (Brand, I. c.) but rather in central Oregon not far dis- tant from stations where Prof. Peck has more recently collected the species. SoME CIRCUMSTANTIAL Evipence. A collection of Phacelia bicolor Torr. reputedly made by J. G. Lemmon and wife in 1884 at Fort Mohave, Arizona, has been the basis for tentatively includ- ing the species in the flora of that state (Kearney and Peebles, Flow. Pl. and Ferns Ariz., p. 734). In my study of Phacelia sect. Microgenetes I have examined specimens of this collection in the University of California Herbarium and in the United States Na- tional Herbarium and, from a detail of the latter, it seems to me quite evident that the plant did not come from Arizona. An ex- traneous leaf which adheres to the base of the right hand plant of the specimen in the United States National Herbarium and which was obviously pressed when the Phacelia was pressed is the leaf of Purshia tridentata (Pursh) DC. If the inference is correct that this leaf and the phacelia came from the same place, then the pha- celia did not come from Fort Mohave. Although Purshia is known in the Arizona mountains farther east, it is as certain that the pha- celia did not come from the mountains of Arizona as it is certain that the purshia did not grow in the Sonoran desert along the lower Colorado River. Also it is quite unlikely that the collection was made in the Mohave Desert as has been suggested by Kearney and Peebles (1. c.) because in this Lemmon collection the filaments are puberulent while in all collections that I have examined from _ the Mohave Desert the filaments are glabrous. Moreover the purshia in the Mohave region is Purshia glandulosa Curran, a species with glandular-pitted leaves that are quite different from the leaf adhering to the plant of Phacelia bicolor. Rather, there is every reason to believe that this collection of Phacelia bicolor in- cluding the purshia leaf was made in northeastern California or northwestern Nevada where both it and Purshia are known to grow and where Lemmon collected abundantly. SoME DIstTrIBUTIONAL Notes. Among many hundreds of 152 LEAFLETS OF WESTERN BOTANY _ [VOL. IV, NO. 5 specimens that I have examined during the past year in my revis- ional study of Phacelia sect. Microgenetes, about eight collections seem to represent records of species in states of the United States and Mexico from which the species have not been reported in the literature. These are listed briefly as follows: PHACELIA AFFINIS Gray. Heretofore known from Lower Cali- fornia, California, Nevada, Arizona, and Utah, this species may now be reported from New Mexico: Peloncillo Mts., Hidalgo County, Ripley & Barneby No. 4218a (CAS*). PHACELIA GLANDULIFERA Piper. The only collections of this species that I have seen from Wyoming (where it has been con- fused with P. Ivesiana Torr.) are two specimens collected by Prof. Aven Nelson well over forty years ago: Point of Rocks, Sweet- water County, A. Nelson in 1898 (RM) ; Granger, Sweetwater County, A. Nelson No. 4696 (M). Although P. glandulifera is very common in southern Idaho and northern Nevada, no speci- men of it has been seen from Utah. PHACELIA IvEsIANA Torr. All specimens that I have seen from Idaho that have been referred to this species have proved to be P. glandulifera Piper except one: 13 miles west of Glenn’s Ferry, Elmore County, Davis No. 1939 in part (UI). The other part of the collection is P. glandulifera. I cannot find that P. Ivesiana has been reported from New Mexico but it may be recorded from that state on the basis of the following collection: Ojo Caliente, about 13 miles southwest of Zuni Pueblo, McKinley County, Hodge in 1919 (US). PHACELIA NEGLECTA M. E. Jones. Heretofore this species has been reported from Lower California as P. pachyphylla Gray, a Californian species which to the present is not known south of the international boundary. The Lower Californian collection of P. neglectais Orcutt No. 2035 from west of Indian Wells (US). No collection of P. neglecta has been seen from Sonora, Mexico, but it must grow there since the species has been collected in the Organ Pipe Cactus National Monument in Arizona less than a mile north of the international boundary: 1 mile north of Sonoyta, Pima County, McDougall No. 93 (UA). _. * The following symbols indicate the institutions where specimens herein cited have been seen: CAS, California Academy of Sciences; M, Missouri . Botanical Garden; RM, Rocky Mountain Herbarium, University of Wyo- ming; UA, Univ ersity of Arizona; UI, University of Idaho, Southern Branch; US, United States National Herbarium. LEAFLETS of WESTERN BOTANY Available A limited number of copies of volumes I, II, and III at $5.00 each. Single copies of most numbers in these volumes are available at $.40 each. LINDLEY’s GLOSSOLOGY An excerpt from this most useful work, reissued with an introduction by Alice Eastwood. $.50. ALICE EastTwoop’s Key to the Common Families of Flowering Plants in California and Guide for the Analysis and Description of Flowering Plants. $.25. Address: Joun Tuomas HoweELi California Academy of Sciences, Golden Gate Park San Francisco, California Vot. IV No. 6 LEAFLETS of WESTERN BOTANY CONTENTS PAGE The Wild Flower Gardens of San Francisco inthe 1890’s_ . 153 ALICE EAstwoop Marin County Miscellany—III . . . . . . . 156 JoHn THomas Howe. emeemus An Calitornia |»..\))+) ee Re! NOL JoHn THomAs Howe. The Introduction of Spartina alterniflora to Washington with Oyster Culture Mh ant Wane pc See Tueo. H. SCHEFFER A Genus New to North America . . . . . .. 164 C. V. Morton anp J. T. Howeiu New Varieties of Two Well-known Californian Plants . 166 ALICE EAstwoop Some Noteworthy Californian Plant Collections - . . 168 JoHn THomas Howe. Studies in Rosacew, Tribe Potentillee . . . . . = I7i JoHN Tuomas Howe. This number published with funds from the California Botanical Club SAN FRANCISCO, CALIFORNIA May 21, 1945 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. PORN Au RU UG UU ul UU Uk PRETTY RUAUU UDO OU LA LOPE Gear ba Owned and published by Aticr Eastwoop and JoHN THomas Howey MAY, 1945 | WILD FLOWER GARDENS 153 THE WILD FLOWER GARDENS OF SAN FRANCISCO IN THE 1890'S BY ALICE EASTWOOD In Dr. Behr’s reminiscences of the San Francisco flora, pub- lished in the second volume of Zoe in 1891 and dating back to the early 1850's, several species were listed that had disappeared by 1892. In that year in the same volume of Zoe, Katharine Bran- degee published a list of species collected in 1892 in San Francisco by members of the California Botanical Club which had been founded the preceding year. This list numbered 585 species of flowering plants, ferns, and fern allies. Those recorded by Dr. Behr and no longer to be found were marked by an asterisk. They were the small unidentified pines on Lone Mountain; the box- elder, Acer californicum ; California laurel, Umbellularia califor- nica; marestail, Hippuris vulgaris ; large-flowered dogwood, Cor- nus Nuttall; buck-bean, Menyanthes trifoliata; pitcher sage, Sphacele calycina; stream orchis, Epipactis gigantea; cotton grass, Eriophorum gracile; grape-fern, Botrychium sp.; lady-fern, Asplenium Filix-foemina (Athyrium). Alsine palustris (Arena- ria paludicola Rob.), named by Dr. Kellogg in 1862, was also listed as probably extinct. However, the record of a specimen in the Gray Herbarium collected by Henry N. Bolander led to its rediscovery in a marshy place near Fort Point now covered with army barracks and before that by an airfield. The specimen in the California Academy of Sciences that I collected there in 1899 was destroyed in the 1906 disaster but a duplicate was preserved in the herbarium of Miss Evelina Cannon to whom I often gave duplicates. Her herbarium was presented to the academy after the great fire by her niece, Miss Carrington, and in it are her specimens collected in San Francisco in the 1890's. In the 1890's, the open country everywhere around San Fran- cisco was a beautiful wild flower garden in the spring. In the region near Lake Merced the wild flowers were so thick that it was impossible to avoid stepping on them. There were California poppies, nemophilas, violets, cream-cups, owls-clover, mouse-ear 1 These plants were collected near the Russ Garden, along the border of extensive marshes. The garden was owned by the Russ family and was situ- ated on the south corner of Sixth and Harrison streets. It was opened in eh = a popular resort where many nationalities gathered on Sundays and anced. Leaflets of Western Botany, Vol. IV, pp. 153-176, May 21, 1945. 154 LEAFLETS OF WESTERN BOTANY _ [VOL. IV; NO. 6 chickweed, Indian paintbrush, clovers, etc. The yellow violet, Viola pedunculata, was especially common, known to the children as Johnny-jump-up. Today, new roads, golf links, vegetable fields, and human habitations have driven them away and it is doubtful if a single native flower persists. A solitary madrone grew in a gulch leading to the lake, the sides of which were cov- ered with dense chaparral. Ceanothus incanus formed a thicket together with hazel, manzanitas, low oaks, and other shrubs. Climbing over them was the Dutchman’s-pipe, Aristolochia californica. | The Bay View Hills near South San Francisco was the home of some flowers not found elsewhere in the city. These were the climbing nemophila, Nemophila aurita; the white flax-flowered gilia, Linanthus liniflorus; islay cherry, Prunus ilicifohia; the tall larkspur, Delphinium californicum; the stinging phacelia, Pha- celia malvaefolia. On the rocky summit, now known as Bay View Park, were rock cress, Arabis blepharophylla; the downy-leaved paintbrush, Castilleja foliolosa; pennyroyal, Monardella villosa, and later, several composites and eriogonums. In the meadow below was a garden similar to that at Lake Merced. In some places near Visitacion Valley the ground was white with the pelican-flower, Orthocarpus versicolor, and in other places the areas were pink-tinged from owls-clover, Orthocarpus densiflorus. These common names were given to these flowers by Dr. Kellogg, the latter because each flower formed the face of an owl. Every- where golden poppies and amsinckias glorified the landscape. On the hills back of Ingleside, then known as Sunset Heights, and on the hills rising from the Mission, a great many flowers could be found. The Potrero Hills, too, were then covered with flowers. The white Fritillaria liliacea grew there and in a wet place Miss Cannon discovered a long-lost species, Sanicula mari- tima. At Hunters Point, now entirely covered by-a housing proj- ect, this fritillaria was found a few years ago by Lewis S. Rose, and I found the fragrant Dodecatheon bernalinum there in 1916. This was still on Bernal Heights a few years ago. Twin Peaks still has the low blue-flowered lupine, Lupinus collinus, a few clumps of Iris Douglasiana and I. longipetala, but in the 1890’s, the tiny yellow-flowered Beria gracilis monopolized the ground in places and the golden slopes could be seen from the windows of the California Academy of Sciences on Market Street MAY, 1945] WILD FLOWER GARDENS 155 before the erection of the Emporium building. Some other relics of the 1890’s may yet be on those hills. The Mission Bells are occasionally found there, and other species, such as footsteps-of- spring, Sanicula arctopoides, other species of the Parsley Family, the pink-flowered Sidalcea, and, of course, buttercups, Ranun- culus californicus, everywhere. The San Bruno Hills south of Visitacion Valley were noted for the great beds of Jris longipetala. Homes are replacing them now. Some plants are still left but unless a tract is preserved by some lover of this beautiful flower, soon all will be gone. Colma Canyon was a favorite haunt of the Botanical Club. The white sessile-flowered trillium grew there, besides colum- bine, meadow rue, fringe-cups (Tellima), California saxifrage, baby-blue-eyes (Nemophila insignis), cream cups (Platystemon), bleeding heart (Dicentra formosa), and the tiny white-flowered Meconella. In some places the sluggish creek was covered with the dainty water-fern, Azolla. On the adjacent hills, the humming- bird sage, Salvia spathacea, was common. These hills were clothed with native shrubs. Besides the oaks, hazel? and poison oak, there were manzanitas, shad-brush (Amelanchier), twin-berry (Loni- cera Ledebouri), ocean spray (Holodiscus), oso-berry (Osma- ronia), the two thimble berries (Rubus parviflorus and R. spec- tabilis), blackberry vines, and other shrubs, but no ceanothus. These may still be there as it is a dairy ranch with no trespassers permitted. In the early 1890’s, the Presidio was as beautiful a wild flower garden as Lake Merced but the pines, cypresses, and eucalyptus have destroyed most of the flowers and any that might have survived have probably been smothered by the fig marigold, Mesembryanthemum. In olden times a low form of Zigadenus Fremont whitened the ground, the blue violet as well as the yellow was common, as were also two species of Orthocarpus not found in any other part of San Francisco; and a dark red onion and clumps of the blue Douglas iris loved rocky ridges. Besides there were the flowers common elsewhere. However, the yellow-flowered bush lupine, the blue-flowered Chamisso lupine, ‘and the broad clumps of Ceanothus thyrsiflorus still hold their own. Not long ago, Lewis S. Rose even added a species of the fern allies known as quillwort, [soetes Nuttallu. This had never 156 LEAFLETS OF WESTERN BOTANY [VOL. IV, NO. 6 before been found in the San Francisco area and is not a common plant. The wax myrtle too, may yet be growing along Lobos Creek or Mountain Lake. Another attractive excursion was along the railroad track that once went to Sutro Heights via Lands End. Except for the wild currant which was abundant in the spring, this area was better for summer and fall. The pink and yellow sand verbenas, the San Francisco willow-herb, Angelica Hendersom, cow-parsnip (Her- acleum lanatum), golden rod, lizard leaf (Eriophyllum), Bac- charis, and the heather-leaved Ericameria, all were common. The sands, forming dunes in many places, are still extensive _as always, though they too are being replaced by rows and rows of homes. However, the franserias, artemisias, seaside daisies, beach morning glory, cakile, wallflower, gilias, eriogonums, cas- tilleja, and many less conspicuous are still to be found. A survey of these and other unoccupied areas in San Francisco may dis- cover many unsuspected relicts. Open spots amid the trees in Sutro Woods back of the University of California Hospital may shelter some and on the upper slopes of Mt. Davidson a few may still persist. MARIN COUNTY MISCELLAN Y—III BY JOHN THOMAS HOWELL In these days when transportation to more distant places is nec- essarily so much curtailed, the hills and valleys of Marin County, California, continue to furnish fruitful fields for botanical ex- cursions near home. Trips to Inverness Ridge on Point Reyes Peninsula, to the hills northeast of San Rafael, to the marshes bordering San Francisco Bay, and to Mt. Tamalpais, have dis- closed a number of interesting plants, several of which will be noted here. First we shall record the discovery of several alien plants which, although they are perhaps as yet too narrowly estab- _ lished to be admitted to the naturalized flora of California, occur spontaneously and may become increasingly aggressive. m 7? uf 7 STENOTAPHRUM SECUNDATUM (Walt.) Kuntze. I have found St. Augustine Grass, which apparently has not been reported for California heretofore, at two widely separated stations in Marin County : as a sidewalk weed at Fairfax where undoubtedly it has MAY, 1945] MARIN COUNTY MISCELLANY—III 157 escaped from cultivation as a lawn grass ; and on the ridge west of Mill Valley along the road to Muir Woods. Although it is indig- enous to broad subtropical and tropical regions from southeastern United States to Argentina, this distinctive grass was flowering in temperate Marin County, and, from the note communicated by Mrs. M. K. Bellue, California seed-analyst, it may be expected in other parts of California. Mrs. Bellue writes: “We pass on vege- tative material of Stenotaphrum secundatum for propagation quite frequently and it is now offered in flats extensively in south- ern California. It has been used in Elysian Park, Los Angeles, since 1918 or before, and it seems to seed at Huntington, Los Angeles County, where it covers acres.’ Outside of Marin County, I have seen it only in Strawberry Canyon, Alameda County, where a small patch flourishes below the University of California Botanic Garden. CypERUs DIFFORMIS L. This sedge, which is a common Old World weed in California rice fields (cf. Leafl. West. Bot. 1: 104, —1934), has been found in a springy spot on the south side of Mt. Tamalpais above Muir Woods. It may be noted that, from the point of view of plant distribution, it is much easier to remark on the migration of this weed from Asia to California over thou- sands of miles of ocean than to speculate on the conveyance of the plant from the interior valleys of California to an obscure spring on Mt. Tamalpais. PoLyGoNuM PATULUM Bieb. P. Bellardi Boiss. Fl. Orient. 4: 1034, not Allioni; P. exsertum J. T. Howell, Leafl. West. Bot. 1:38, not Small. Plants of this erect-stemmed knotweed have been found in the salt marshes bordering San Francisco Bay below Kentfield. For many years I realized that the plants which I had reported as P. exsertum from other salt marshes in the bay region were misdetermined and I am grateful to Dr..J. F. Brenckle who verified in 1941 my tentative redetermination. I originally reported the plant in California from Solano, Napa, and Santa Clara counties, so it is not surprising that it-should turn up in Marin County. The plant is native to the Mediterranean region of the Old World and apparently has not been reported as P. patu- lum for western North America. Lotus uLicinosus Schkuhr. This attractive European pea was found growing vigorously on flats between Mill Valley and 158 LEAFLETS OF WESTERN BOTANY [VOL. Iv; NO. 6 the Almonte salt marshes. Concerning it, Mrs. Bellue has sent me an interesting note which she is allowing me to publish here: “Seed of Lotus uliginosus has been imported for several years and the plant utilized to a considerable extent for forage. It came through the customs under the names of L. major and L. villosus, and has been tried out in a number of soil conservation nursery stations. Some seedsmen also carried it.” By more conservative European botanists the ‘plant has been called L. corniculatus L. var. major Ser. or subsp. uliginosus Pers., names included in syn- onymy by Hegi (FI. Mit.-Eur. IV, 3: 1372). : Lotus uliginosus is a showier plant than L. corniculatus: Hegi describes the inflorescence as 2- to 5-flowered in the latter and at least 10-flowered in the former. Specimens of L. cornicu- latus have come to us from several stations in California, although for the Pacific states Abrams cites it only from Portland, Oregon (Ill. Fl. 2:553). The Californian collections of L. corniculatus have been seen from Siskiyou County (north of Yreka, Winblad in 1941), Solano County (Grizzley Island, Alexander & Kellogg in 1940), and Tulare County (near Springfield, Winblad in 1939). Lotus corniculatus is apparently widespread in the eastern United States but I have seen no reference to the American occurrence of L. uliginosus. Vicia LUTEA L. In May, 1944, this Old World vetch was ~ found locally but abundantly naturalized in the canyon of Salmon Creek by Dr. Hans Leschke and me. Although the corolla is gen- erally yellow, it is sometimes bluish or purplish and in our plant the banner was creamy-white veined with rose. In spite of varia- tion in corolla color the plant is unmistakable because of the pecu- liar pustulate-based hairs of the legumes. Mrs. Bellue has written that seed of “V. lutea is one of the incidental seeds commonly found with the commercial seed of vetch. We usually think of it in imported seed, but it also occurs in domestic. It is probably distributed here where vetch and oats are used for hay.’ As in the preceding, no American reference has been seen for V. lutea, a plant which may well become widely naturalized. 7 7 7 Quercus pumosa Nutt. In Marin County. The discovery of the California Scrub Oak at two stations in Marin County is of more than usual interest because of the rarity of the species in MAY, 1945] MARIN COUNTY MISCELLANY—III 159 the Coast Ranges north of San Francisco Bay. Although writers. have generally remarked on the abundance of the species in the chaparral belt in southern California, they have not specially noted its occasional occurrence to the northward. Out of 167 col- lections of Quercus dumosa in the herbaria at the University of California and the California Academy of Sciences, only 17 col- lections were from north of San Francisco Bay and an analysis of this number indicates the Napa-Lake County area as the center of North Coast Range distribution. The numbers of specimens seen from each of the North Coast Range counties are: Lake 8, Napa 4, Marin 2, Sonoma 1, Glenn 1, and Mendocino 1. In Marin County, I found Quercus dumosa as small trees in the Tan Bark Oak-Madrono forest on the north side of Mt. Tamalpais above Ross and as shrubs in the chaparral near the head of Black Can- yon in the San Rafael Hills. — ARCTOSTAPHYLOS VISCIDA PARRY IN Marin County. Of even greater phytogeographic significance than the occurrence of the California Scrub Oak in Marin County is the discovery of the White Manzanita in the San Rafael Hills northeast of San Ra- fael. Since the nearest locality where it has been reported hereto- fore is on the east side of Howell Mt. in the Napa Range (cf. Jepson, Fl. Calif. 3:45), this record represents not only a note- worthy extension of range in miles but also a critical floristic occurrence. In the San Rafael Hills, Arctostaphylos viscida is _ associated with other trees and shrubs of the Upper Sonoran Zone not commonly recorded from Marin County, namely, Quercus dumosa Nutt., Q. Douglasi H. & A., and Arctostaphylos Man- zanita Parry. While the occurrence of these four plants on the Marin County Peninsula is indicative of the unusual floristic di- versity found in this relatively restricted region, the fact that three of the species have not been found on Mt. Tamalpais immediately to the southwest indicates the critical distributional differences frequently found between closely adjacent areas in the county. A New Form or ARCTOSTAPHYLOS CUSHINGIANA EasT- woop. Whereas it may be suspected that geologic rearrangements more than ecologic differences are responsible for the isolated occurrence of Arctostaphylos viscida Parry in Marin County in the San Rafael Hills, I should judge from field observations that ecologic conditions alone are responsible for the segregation and 160 LEAFLETS OF WESTERN BOTANY _[VOL. IV, NO. 6 preservation of a form of the Cushing Manzanita in localized situations on Inverness Ridge. There, restricted to abrupt ex- posed rocky slopes of granitic hogbacks, a low-growing small- leaved form of A. Cushingiana is found, a form so different from the usual woodland or chaparral form of the species as to be en- tirely misleading. But since in these exposed surroundings even the Bishop Pine (Pinus muricata D. Don) is dwarfed to a tenth its normal size, it may be presumed that the conditions are too severe for the survival of the erect broad-leaved form of A. Cushingiana and that the habitat has “‘selected” that form which actually thrives under such rigorous conditions. In less exposed situations in the vicinity, the normal form of A. Cushingiana grows and between it and the subprostrate dwarf are numerous intermediates in habit and foliage. Although the extreme in variation is very distinctive, it can scarcely be considered more than a form botanically for the reason that it is so closely related to the usual type both geo- graphically and morphologically. It is of interest to speculate on the origin of this form which exhibits a habit so different from that found in the species. The fact that prostrate A. Uva-ursi (L.) Spreng. grows less than a mile away suggests the possibility of hybridization’; but there is nothing more than the unusual habit to prompt such a thought, for in every other character the subprostrate manzanita seems entirely like A. Cushingiana. The low manzanita with the associated pygmy pine was found in March, 1944, by Dr. Hans Leschke and me near the head of Second Valley west of Inverness as we were making a traverse of Inverness Ridge southward from the Point Reyes Lighthouse Road. It had been known several years earlier from a station to the south near the head of First Valley, where, I believe, it was discovered by Miss Beryl Schreiber. Plants from the latter sta- tion have been cultivated successfully by Mr. Louis L. Edmunds as the Huckleberry Manzanita at Danville, Contra Costa County, 1 An interesting plant, undoubtedly a hybrid, was found at one of the stations where the subprostrate form of A. Cushingiana grew. At this place, the pygmy manzanita was associated with A. virgata Eastw. as well as A. Cushingiana. The suspected hybrid was low but suberect, about 3 dm. tall, with stems, leaves, and inflorescences densely hirsutulous-glandular (J. T. Howell No. 20768). The indument was that of A. virgata but the small leaves and the stump-sprouting base corresponded to these features in the low form of A. Cushingiana. The little plant was like a pygmy A. glandulosa Eastw., a@ species which, as it occurs typically and extensively on Mt. Tamalpais, might possibly be a long-established hybrid-derivative of A. virgata and A. Cushingiana, both of which are also typical on the mountain. MAY, 1945] CARTHAMUS IN CALIFORNIA 161 and it promises to be valuable horticulturally. In recognition of its distinctive habit, the formal botanical name for the plant can appropriately be repens.” CARTHAMUS IN CALIFORNIA BY JOHN THOMAS HOWELL The earliest record of the spontaneous occurrence of a species of Carthamus in California is given by Greene in 1897 in Flora Franciscana, p. 475, where he reports C. lanatus L. (under the generic name Centrophyllum) from South San Francisco. Be- sides C. lanatus, two other species have been reported as weeds in California : in 1931 I reported the weedy occurrence of C. tincto- rius L. near Bethany in San Joaquin County (Madrofio 2: 22), and, in 1940 Robbins reported the occurrence of C. nitidus Boiss. in San Luis Obispo and Tuolumne counties (Univ. Calif. Agric. Exp. Sta. Bull. No. 637, p. 90). The safflower, C. tinctorius, is distinct enough with its entire or merely spinulose leaves and no taxonomic problem is concerned with, its identity ; but the identity of the distaff thistles in California is more critical. Two rather distinctive types of plants are represented by the California distaff thistles, the one in which the upper leaves and involucral bracts are broad and relatively short, the other in which the leaves and involucral bracts are narrow and elongate. To the former belongs the plant originally collected near San Francisco that was reported as C. lanatus, to the latter the plant collected in other California stations that has been generally reported as C. nitidus. The San Francisco plant corresponds well with the rather considerable suite of Eurasian specimens of C. Janatus in the herbarium of the California Academy of Sciences and the plant is undoubtedly correctly referred to that species. Since no Old World specimens of C. nitidus have been available for compari- son, the California plant that has been referred to that species has been compared with Boissier’s original description (FI. Orient. 2 Arctostaphylos Cushingiana Eastw. fma. repens J. T. Howell, fma. nov., a forma typica per caules subprostratos radicantes graciliores et folia minora abludens ; haud glandulosa. Types collected on granitic ridge west of Inverness, Inverness Ridge, Marin County, Mar. 26, 1944, J. T. Howell No. 19353 (flowers, Herb. Calif. Acad. Sci. No. 324000), and June 4, 1944, J. T. Howell No. 19537 (fruit, Herb. Calif. Acad. Sci. No. 323998). Also in the Academy herbarium are the fol- lowing: specimen from plants cultivated by Louis L. Edmunds; specimen from plant intermediate between A. Cushingiana and fma. repens, J. T. ewer No. 19354. As in typical A, Cushingiana, fma. repens stump-sprouts er fires. 162 LEAFLETS OF WESTERN BOTANY _ [VOL. IV, NO. 6 3: 708) and it does not agree. In C. nitidus the stems are white and shining, the innermost involucral bracts are pectinate-ciliate above, and the flowers are purplish, whereas in the Californian plant the characters of these parts are different and indicate a close relation to C. lanatus. A survey of all the specimens of Carthamus in Herb. Calif. Acad. Sci. furnished a clue which, it would seem, has led to the identity of the second Californian type. A specimen in Herb. A. Prager. from Almeria in southern Spain that is determined as C. beticus (Boiss. & Reut.) Nym. seems exactly like the Califor- nian material; and, since the Spanish specimen comes not far from the type-locality of the species, it may be accepted as repre- sentative of that plant. The Spanish plant, which was originally described by Boissier and Reuter as Kentrophyllum beticum (Pugill. Pl. Nov. 65), was said by its authors to be very closely related to K. lanatum, differing in the more divaricate branches, the narrower few-flowered head, the strongly spreading elongate involucral bracts, and the nigrescent achenes. Since the Califor- nian weed is distinguished by these very characters it is undoubt- edly referable to the Spanish distaff thistle. Because of the close relationship between C. lanatus and C. beticus, I have sought to determine how the latter species is ac- cepted by more recent workers on the flora of Spain and of Algeria where the plant is indigenous. Whereas in Index Kewensis C. beticus is given as a synonym of C. lanatus, Knoche in Flora Balearica (2: 502,—1922) in synonymy under C. lanatus indi- cates indirectly that the two are separable. Probably the proper systematic position of the plant is given by Battandier in Flore de l’Algérie where he treats the entity as Kentrophyllum lanatum B beticum. (1888-90, p. 508). I have not been able to find this combination in Carthamus ; but since this Californian study is not the place to propose a nomenclatural change in so critical a taxo- nomic problem belonging to an exotic flora, I shall apply to the Spanish distaff thistle in California the specific name, C. beticus (Boiss. & Reut.) Nym. In summary, the species of Carthamus adventive or natural- ized in California, as represented in the California Department of Agriculture Herbarium (CDA) and in the California Academy of Sciences (CAS), are: MAY, 1945] SPARTINA ALTERNIFLORA 163 C. sxticus (Boiss. & Reut.) Nym. Sonora, Tuolumne Co., Shirrard in 1927, 1928 (CDA) ; Jackson, Tuolumne Co., J. T. Howell in 1945 (CAS) ; San Luis Obispo, San Luis Obispo Co., Saunders in 1932 (CDA), Fetters in 1937 (CDA) ; San Diego, San Diego Co., Fehlman in 1938 (CDA). C. ranatus L. South San Francisco, San Mateo Co., Cannon in 1894 (CAS); Larkin and Francisco streets, San Francisco, Eastwood in 1920 (CAS). ; C. tinctortus L. Bethany, San Joaquin Co., Keithley in 1930 (CAS); Antelope Valley (in 1929 acc. Munz, Man. S. Calif. Bot. 584) ; also in Kings, Tulare, San Benito, and Sacra- mento counties acc. Robbins (l.c.). In this study I wish to acknowledge the interest and codpera- tion of Mrs. Margaret K. Bellue of the California Department of Agriculture who loaned me the department specimens for study and who allowed me to examine the correspondence concerning Carthamus in the department files. THE INTRODUCTION OF SPARTINA ALTERNIFLORA TO WASHINGTON WITH OYSTER CULTURE BY THEO. H. SCHEFFER It will be of interest to biologists in general to consider the exotic species of plants and animals that may be introduced into our Pacific coastal bays and inlets with the recent extension of oyster culture. From limited knowledge of the broader subject, we shall not attempt here any list or classification but shall deal only with one of the higher plants which apparently has come with the oyster spat into Willapa Bay, Washington. We refer to a species of cordgrass, Spartina alterniflora Lois., native on the Atlantic and Gulf coasts from Newfoundland to Texas. The occurrence of this plant in Willapa Bay was first brought to our attention about seven years ago by an oyster grower living on the Long Beach peninsula, who had noticed the little “green islands” in the curve of a small bay partly encircled by a spit. In this situation the patches of Spartina were all but covered by water at the higher tides, giving the novel appearance that had attracted the oysterman’s attention. Our investigation at that time 164 LEAFLETS OF WESTERN BOTANY [VOL. IV, NO. 6 and during the following two or three years discovered no flower- ing plants but the species was spreading well by rootstocks. There were now several acres in the stand which had developed from the small clumps first noticed, as reported, “about 1911.” The eastern oyster was introduced into this locality about 1907 and there were subsequent plantings, the spat having been procured near Providence, R. I. Early endeavors to determine the species from the sterile plants failed, so it was not certain whether the plants represented S. leiantha Benth. of the Pacific coast but not reported north of San Francisco Bay or S. alterniflora of the Atlantic coast. In the fall of 1941, the refuge manager at Willapa National Wildlife Refuge collected some flowering plants and the determination was readily made as S. alterniflora. So far we have not located other stands of this plant on the Washington coast, but there are two or three likely spots to be investigated. Not only is the field open to all beach combers who have the inspiration to follow up these matters but lasting honors await those who would track the more recent introductions of the Japanese oyster spat. A GENUS NEW TO NORTH AMERICA ® BY C. V. MORTON AND J.T. HOWELL In the summer of 1944, we independently studied and identi- fied material of an unusual aquatic weed which was collected by Mr. Loren L. Davis of the U. S. Department of Agriculture at the Biggs Rice Field Station, Butte County, California. The plant strongly suggested the genus Elatine in habit but it differed in seed structure and other characters. It has proved to be Dopatrium junceum (Roxb.) Buchanan-Hamilton (ex Benth. Scroph. Indice 31,—1835), which was based on Gratiola juncea Roxb. (Pl. Coromand. 2: 16, pl. 129,—1798). The small genus Dopatrium Buch.-Ham., of the Scrophula- riace@, tribe Gratiolee, is strictly Old World in distribution. It is related to Gratiola from which it is distinguished technically by the dehiscence of the capsule, which is loculicidal only, rather than loculicidal and septicidal. The species D. junceum was described from India, but it is widespread through Asiatic and Malaysian MAY, 1945 | DOPATRIUM 165 regions from Japan and China to Australia. In Jepson’s Manual of the Flowering Plants of California it will key to Gratiola ebrac- teata Benth., from which it is at once distinguished by its dimor- phic flowers (the lower subsessile, the upper slender-pedicellate), the very short calyx-lobes (these about half as long as the capsule, rather than exceeding the capsule as in Gratiola), the violet corol- las of the uppermost flowers, and by the reduced, bract-like stem- leaves. An interesting and noteworthy detail exhibited by the material which we studied is the dimorphism of the flowers, those in the submerged lower axils being cleistogamous with abortive corollas, those in the emergent upper axils being open with small pale violet corollas. The former flowers are subsessile or nearly so, while the latter are borne on filiform pedicels that usually exceed the flower in length. In spite of the fact that Asiatic material examined by us discloses the cleistogamous flowers in the lower axils, we have seen no mention of the floral dimorphism in the floras of those regions where the plant is indigenous. Concerning the occurrence of this interesting weed, Mr. Davis has written as follows: I have observed this weed growing in the rice section for about five years but gave it no special notice since it does not appear to cause any trouble. The practice here is to submerge the dry land, after seeding the rice, to a depth of about 4 to 6 inches and the water is held continuously at this depth until harvest about September 15. The weed germinates and makes a rapid growth under this continuous submergence. On land that has been flooded on May 1, I have seen the flower by the latter part of June and plants with flowers can be found until the water is drained in September. I have not observed this plant growing in a thick stand of rice. It grows where the rice is thin, and here at the station is very thick in the alley-ways between plots. Apparently sunlight is necessary for its growth. Although all of the material on which our observations and determination rest was collected by Mr. Davis, the specimen re- ceived in Washington was from Mrs. M. K. Bellue of the Cali- fornia Department of Agriculture and the specimen received in San Francisco was from Dr. Katherine Esau, University of Cali- fornia at Davis. To Mr. Davis, Mrs. Bellue, and Dr. Esau we wish to express our appreciation for the specimens and their help- ful interest. 166 LEAFLETS OF WESTERN BOTANY _ [VOL. IV; NO. 6 NEW VARIETIES OF TWO WELL-KNOWN CALIFORNIAN PLANTS BY ALICE EASTWOOD The first is a variation in the common California Laurel or Bay, Umbellularia californica (H. & A.) Nutt. This species was named from specimens collected by Archibald Menzies who ac- companied Vancouver as physician and botanist.on his voyage of discovery in 1792. In the fall of that year he was in San Francisco and Monterey so that it is uncertain where the collection was made. However, it might have been San Francisco as Dr. H. H. Behr, who came to California in 1852, records large trees of laurel fringing the shores of the Presidio from Fort Point inland (Zoe 2: 2-6,—1892). In the list of plants indigenous to San Francisco, published in the same volume of Zoe by Katharine Brandegee, the reference is to Dr. Behr’s reminiscences only. From Oregon in the north to San Diego County in the south and extending into the Sierra Nevada, the trees show a similarity in every way _ except for differences in habit due to exposure. The color of the fruit is either yellow or purple in trees growing side by side. .In this new variety, which seems to be local in that area in Fresno County where the rare shrub Carpenteria flourishes, the lower surface of the leaves and the branches of the panicle are clothed with a fine white down which is entirely absent from trees elsewhere. No difference is apparent in the flowers, fruits, or shape of leaves, but this difference in indument is at once notice- able in the bicolored leaves. As it seems a geographical variation only, the name Umbellu- laria californica var. fresnensis’ is appropriate. The following specimens from Fresno County are in the herbarium of the Cali- fornia Academy of Sciences: No. 281134 from above Trimmer Springs, collected by Inez Whilton Winblad, May 16, 1937; No. 136199 collected at Big Sandy Creek near Prather by Julia McDonald, April 5, 1915. Miss McDonald’s great-niece, Mrs. Loraine Byrd Prather, is now living in Miss McDonald’s old 1 Umbellularia californica (H. & A.) Nutt. var. fresnensis Eastwood, var. nov., differt a U. californica typica: pagina inferiore foliorum et ramis pani- cule tenuiter tomentosis. Type in flower, Feb. 18, 1945, Herb. Calif. Acad. Sci. No. 323837; in fruit, Sept.-25, 1944, No, 323838; both collected at Big Sandy Creek, Prather, Fresno County, California, by Mrs. Loraine Byrd Prather. MAY, 1945] TWO CALIFORNIAN PLANTS 167 home at Prather. From her, fine specimens in both flower and fruit have been recently received and will be taken as the types of the variety. Mrs. Prather was eager to help and our thanks are due to her. 5 7 t The second is a peculiar variation in Linanthus dichotomus Benth. This species, by the way, is the type of the genus Linan- thus and was collected by David Douglas “in California,” actual locality unknown. As it is a common species in California and its characters are well known the peculiarity of the plants growing in Napa and Lake counties has been noted by collectors in the field only as it is not evident in herbarium specimens. During the day, in typical plants, the flowers are almost invisible, the dull pink color of the edge of the corolla-lobes convolutely folded in the closed flower completely conceals what is within. In the late afternoon these lobes unfold and the flowers spring open so that - soon their pure radiant beauty covers the earth like new-fallen snow, suggesting the common name Evening Snow, which came to me-when I first saw the plants in Kern County in 1893. This was near the San Emigdio Ranch where, through the kindness of William S. Tevis, I made my collecting headquarters. The odor was so disagreeable that the common name among the chil- dren was “pig-pens.” Some time ago we received from Mrs. Rodney Hazel of Lake- port, California, a specimen of Linanthus dichotomus which, she wrote, was open at mid-day and exhaled a pleasing fragrance. The characteristics were so different from those given in Jepson’s Manual and Miss Parson’s Wild Flowers of California that she was uncertain and sent it to me to settle the doubt. An inspection of the specimens in the herbarium of the California Academy of Sciences revealed two other collections noted as open at mid-day: specimens collected by J. T: Howell on upper St. Helena Creek, Napa County, May, 1940, No. 15459; Eastwood & Howell No. 5541, near Putah Creek north of Middletown, Lake County, May 15, 1938. The varietal name meridianus seems to be appropriate.? diftert corolla patente ad meridion ef odore suave. ‘Type: No. 380901, Herb. Calif. Acad. Sci., collected on upper St. Helena Creek, May 19, 1940, by J. T. Howell, No. 15459. . 168 LEAFLETS OF WESTERN BOTANY _ [VOL. IV, No. 6 SOME NOTEWORTHY CALIFORNIAN PLANT COLLEGTIONS BY JOHN THOMAS HOWELL Among the most interesting plants of the southern Sierra Nevada are those which recur at high elevations in the mountains of southern California and which are more or less restricted to the two regions. An addition to the considerable list of these plants . is Calyptridium Parryi Gray which has been known from the sierras of southern California (cf. Leafl. West. Bot. 3: 265) but which may now be reported from the Sierra Nevada: end of the Mt. Whitney road, Inyo County, Mark Kerr in 1937. A new California station for a rare but widely distributed western plant, Calyptridium roseum Wats., is indicated by the _ following collection: in volcanic sand, Smoke Creek northeast of Viewland, Lassen County, Ripley & Barneby No. 5767. (Cf. Leafl. West. Bot. 3: 265). Sedum Purdyi Jepson, a rare and distinctive species confined to the Californian part of the Klamath Mountains, has recently been treated in a taxonomic and genetic study by Clausen and Uhl (Madrofio 7: 175-177) who state that the highest altitude at which the plant is known to them is 617 meters (about 2000 feet). I have made two collections of the species at considerably higher elevations and these I cite here, not only to record the higher eleva- tions but also to indicate two additional stations for a species known perhaps from not more than a dozen collections: between Big Flat and Caribou Gulch in canyon of the South Fork of Salmon River, el. 5000 feet, Salmon-Trinity Alps, Siskiyou County, J. T. Howell No. 13565; between Sky High and Marble valleys, el. 5600 feet, Marble Mts., Siskiyou County, J. T. Howell No. 15142. In 1937, Lester Rowntree collected Hemieva ranunculifolia (Hook.) Raf. in Caribou Basin in the Salmon-Trinity Alps, Siskiyou County, California. The only other Californian col- lection of this plant recorded in the literature is one by Mrs. R. M. Austin on Spanish Peak, Plumas County, in the northern Sierra Nevada in 1878. These two Californian collections indicate a floristic relation between the Klamath Mts. and the northern Sierra Nevada that antedates the origin of the Cascade Range MAY, 1945 | CALIFORNIAN PLANT COLLECTIONS 169 in the late Tertiary. North of California, this plant ranges to Alberta and British Columbia, but, according to Peck (Man. Higher Pl. Ore., p. 372), its only occurrence in Oregon is in the Wallowa Mts. As will be noted in my remarks on Veronica Cusickii a couple of paragraphs below, this pattern of distribution from California to northeastern Oregon is probably floristically significant. Cryptantha circumscissa (H. & A.) Jtn. is a plant widespread through high arid valleys and on Basin slopes of western North America and only rarely does the range of the species or its variety hispida (Macbr.) Jtn. extend to the Pacific slope of the Sierra- Cascade crest. A noteworthy occurrence of var. hispida on the west slope of the Sierra Nevada, however, is to be recorded from Kaiser Peak near Huntington Lake in northern Fresno County where the plant was found at an elevation of about 10,000 ft. by Mr. Henry M. Pollard on July 31, 1943. This collection not only extends the known range of the variety northward along the west slope of the Sierra from southern Tulare County but the elevation at which it was found approaches the altitudinal limit for the species. The only specimen of either the species or the variety that I know from a higher station is one collected by Mr. Frank W: Peirson in the Rock Creek Lake Basin in northeastern Inyo County at 10,400 ft. (Peirson No. 9485). In August, 1944, Mr. Allan L. Chickering collected specimens of Veronica Cusickii Gray near Needle Peak, Placer County, at an elevation of 8500 ft. This is the first record of the Cusick Speedwell from the Sierra Nevada to come to my attention* and it is especially interesting since otherwise the plant has been known in California only from the Klamath Mountains in Siski- you County (cf. Jepson, Man. Fl. Pl. Calif. 933). To the north- ward I’. Cusickii is not uncommon in the Cascade Range in Washington, but in Oregon it is known only from the moun- tains in the northeastern part (cf. Peck, Man. Higher Pl. Ore. 660) and Rydberg (FI. Rocky Mts. & Adj. Plains 782) reports it from Idaho. The distribution of an occasional species or of a closely related group of species from the Sierra Nevada and Klamath Mts. in California to the mountains of northeastern * Shortly after writing this I learned that Veronica Cusickii had been col- lected in the Sierra Nevada in 1937 by Mr. Frank W. Peirson: along edge of open plateau southwest of Woods Lake, Alpine County, July 28, 1937, Peirson No, 12100. Mr. Peirson writes that “the plants grow in sandy or gravelly areas and not in wet locations by streams as does V. alpina.” 170 LEAFLETS OF WESTERN BOTANY _ [VOL. IV; NO. 6 Oregon and Idaho is not a usual distributional pattern, but it is notably exemplified in the distribution of the Phacelia Pringlet complex (cf. Howell, a revision of species related to P. Douglasi, P. linearis, and P. Pringlei, ms.). Moreover, the western distri- bution of Stellaria obtusa Engelm. which has recently been re- ported from the southern part of the Klamath Province in Cali- fornia (Leafl. West. Bot. 3:192) may also be related to this pattern, since apparently the stellaria is not known to occur other- wise south of the mountains of northeastern Oregon. I believe that the occurrence of the chickweed in the northern Sierra Nevada is to be expected. One of the most remarkable members of the Composite in western America, Dimeresia Howellii Gray, is to be reported from California on the basis of plants collected by Marion Ownbey and Fred G. Meyer (No. 2136) in a dry gravelly streambed, Cedar Pass, Warner Mts., Modoc County, July 1, 1940. This small an- nual herb is not only congested in habit but its heads are reduced to only two involucral bracts and two flowers. The genus, which is monotypic, was referred by Gray to the Inuloidee@ (Syn. Fl. 1, pt. 2: 448), but Peck has recently treated it as a distinct tribe, Dimeresie (Man. Higher Pl. Oreg., 702, 1941). According to Peck (op. cit., 743), the plant is distributed through eastern Ore- gon from “northeastern Baker Co. to the Steen, Pueblo and War- ner Mts.,” but I have seen no report of it from the southern part of the Warner Mts. in California. Dr. S. F. Blake has written that he collected Micropus amphib- olus Gray “on rocks near Petrified Forest, Sonoma County, in 1927.” This distinctive species, which has been known from rela- tively few restricted collections, has recently been reported from Lake County, California (Leafl. West. Bot. 4: 16). SCIRPUS TUBEROSUS IN CALIFORNIA. Scirpus tuberosus Desf., a plant of North Africa and southern Eurasia from the Medi- terranean to India, was first reported introduced in North America in New York and Alabama (Beetle, Am. Jour. Bot. 29 : 85—86,—1942). California may now be added with the fol- lowing localities reported: south of Willows, Glenn Co., East- wood & Howell No. 2736 and Beetle No. 3060; north of Gridley, Butte Co., Cloyd No. 3A.—A. A. Beetle, Division of Agronomy, University of California, Davis. a MAY, 1945] ‘ STUDIES IN ROSACE® 171 STUDIES IN ROSACE2, TRIBE POTENTILLEE BY JOHN THOMAS HOWELL 1. A RECONSIDERATION OF THE GENUS PURPUSIA Purpusia saxosa, representing a new genus and species, was described by T. S. Brandegee in 1899 (Bot. Gaz. 27: 446, 447) from a white-flowered, Potentilla-like plant collected by C. A. Purpus in southern Nevada. Brandegee related the genus to Potentilla and Chamerhodos: from the former it differed in its “long calyx tube and no bractlets,” from the latter “in its terminal style and position of the stamens,” and from both “‘in its peculiar stipitate receptacle.” Certainly it would appear that Purpusia need not be confused with Chamerhodos, but from Potentilla in its broad sense Purpusia can scarcely be excluded in view of fur- ther studies on recent collections, although as a genus it has con- tinued to be recognized by all workers who have had occasion to deal with it. Although I was convinced of the generic distinctness of Pur- pusia in 1930 when I published the description of P. arizonica Eastw., I began to have doubts as to its validity in 1938, not only when I mistook Potentilla (Ivesia) Baileyi (Wats.) Greene for Purpusia in the course of Nevada field work, but also when I detected there was no essential difference between the receptacle of Purpusia and that of certain species in Potentilla subgen. Ivesia. In 1940 in a specimen of Purpusia arizonica received from Miss Annie M. Alexander and Miss Louise Kellogg, I discovered a number of flowers with 1, 2, or even 3 bractlets, and, on examin- ing other specimens of both Purpusia arizonica and P. saxosa, I have found that nearly all plants (including those of the type of P. saxosa) have an occasional flower 1- to 3-bracteolate. The hypanthium of Purpusia saxosa is usually distinctive both in its length and in its vase-like shape, but the hypanthium of P. ari- zonica is morphologically indistinguishable from the broadly campanulate or hemispheric hypanthia found in certain poten- tillas. And now in a second remarkable plant collected by Miss Alexander and Miss Kellogg in the Kingston Range in southern California in 1941, we have a plant in which the hypanthium is merely patelliform to almost disciform, a hypanthium as shallow as the shallowest in Potentilla. P) LEAFLETS OF WESTERN BOTANY [VOL. IV; NO. 6 While in the past the elongate ebracteolate hypanthium has furnished Purpusia with important useful characters of secondary value, the status of the genus has rested chiefly on the character of the stalked receptacle, although an entirely comparable struc- ture has also been known in a few species of Potentilla subgen. Ivesia, notably in Potentilla nubigena Greene (Ivesia pygmea Gray). However, in the Alexander and Kellogg plant from the Kingston Range even this supposedly diagnostic character of Purpusia disappears, for in that plant the pistilliferous tissue is scarcely elevated above the subdisciform hypanthium. The only criteria by which the plant can be regarded a Purpusia are its ebracteolate hypanthia and Purpusia-like habit, a habit, however, that is also shared by Potentilla (Ivesia) Baileyi. This striking resemblance between Potentilla Baileyi and Purpusia would ap- pear to be too fundamental in details of foliage as well as habit to be interpreted merely as parallel responses induced by the same rigorous environmental conditions in two diverse groups, but rather I regard that the similarity indicates a direct relationship that is phyletic in character. Some may regard this connection between Purpusia and Potentilla subgen. Ivesia too close for sepa- ration, but I believe that the partial or complete suppression of bractlets demarks in Purpusia a group as distinct as other sub- genera in Potentilla. This opinion may find expression in the following new name: Potentilla subgen. Purpusia (T. S. Brandg.) J. T. Howell, stat. nov. Purpusia T. S. Brandg., Bot. Gaz. 27: 446 (1899) ; Rydb., N. Amer. Fl. 22: 291 (1908) ; Tidestr., U. S. Nat. Herb. Contrib, 25: 269 (1925) ; Lemée Dict. Gen. Pl. Phanérog. 5 : 688 (1934) ; Kearney & Peebles Flower. Pl. Ariz. 403 (1942) ; Mc- Vaugh, Contrib. toward Fl. Nev. no. 22:57 (1942). 2. SYNOPSIS OF POTENTILLA SUBGEN. PURPUSIA Key A. Hypanthium patelliform or nearly flat, densely pubescent within; receptacle not elevated ; stamens 6—9...._.................-- 1. P. patellifera A. Hypanthium campanulate to tubular, glabrous within; receptacle stipitates stamens oon: 3° ork ton aes oe 2. P. Osterhoutii B. Receptacle and stalk generally 1—3 times as long as broad; petals yellawase)!) io Ve 2a. P. Osterhoutii var. typica B. Receptacle and stalk generally 4—5 times as long as broad; petals white or occasionally yellow....u.....2.-......---cceecececeneeeeee eos ok. tun OS Een hd a Bene aoe se 2b. P. Osterhoutii var. saxosa MAY, 1945] STUDIES IN ROSACEE 173 1. Potentilla (Purrusia) patellifera J. T. Howell, spec. nov. Herba perennis, caulibus foliisque tenuiter velutino-villosis et paulum glandulosis, confertis in caudice lignoso basibus persistentibus foliorum veterium ves- tito; caulibus 0.5—2 dm. longis, gracilibus, patentibus, quam folia basalia plerumque paulum longioribus ; foliis basalibus (laminis petiolisque) 3.5—10 cm. longis, foliolis 1—4-jugis, obovatis vel rotundis, sessilibus vel breviter petiolulatis, palmato-lobatis, petiolis 1—5 cm. longis, foliis caulinis minori- bus, foliis in inflorescentia ad bracteas foliaceas reductis; floribus paucis in panicula cymosa foliacea, pedicellis gracilibus, villosis, 0.5—2 cm. longis; hypanthio patelliformi vel subdisciformi, 15—2 mm. diametro, villosis extus, intus dense breviter villosis; bracteolis 0; sepalis anguste vel late ovatis, 3—4 mm. longis; petalis flavis, angustis, oblanceolatis vel ellipticis, 3 mm. longis, 0.5—1 mm. latis, acutis; staminibus 6—9, filamentis circa 1 mm. longis, antheris circa 1 mm. longis; pistillis 6—9, stylis ovariisque glabris ; receptaculo humili, planiusculo, villoso, haud stipitato; acheniis ignotis. Type: Herb. Univ. Calif. No. 667183, collected by Annie M. Alexander and Louise Kellogg, No. 2355a, June 20, 1941, in crevices of a rocky wall near summit of main ridge, Kingston Range, San Bernardino County, California. There is a duplicate in the Dudley Herbarium. Only one other collection has been seen, Alexander & Kellogg No. 2355 (DS, UC, US),* collected from crevices of a north-facing cliff in a spur canyon southeast of Tecopa Pass, Kingston Range. As I have pointed out in the preceding section of this paper, Potentilla patellifera entirely breaks down the principal differ- ences by which the genus Purpusia has been separated from Potentilla, but, with its ebracteolate flowers and distinctive habit, it is definitely referable to this relationship. From the only other species in the subgenus that I am recognizing, the present species differs not only in the shallow, hairy hypanthium and sessile receptacle, but also in the greater, if somewhat indefinite, number of stamens. In the numerous flowers that have been examined none has had less than 6 stamens and generally the number has been 7 to 9. 2. Potentilla Osterhoutii (A. Nels.) J. T. Howell, comb. nov. Purpusia Osterhoutii A. Nels., Amer. Jour. Bot. 21: 574 (1934). Stems more or less spreading from the simple or few-branched woody caudex, 0.5—2 dm. long, the stems equaling or a little exceeding the basal leaves; leaflets in 2—5 pairs, roundish to elliptical or oblanceolate, crenate to palmately lobed or divided; flowers few to numerous in a leafy-bracted cymose panicle; hy- *Institutions from which material has been examined are: California Academy of Sciences, CAS; Dudley Herbarium, Stanford University, DS; LORS National Arboretum, NA; Rocky Mountain Herbarium, University of Wyoming, RM; University of California, UC; U. S. National Herbarium, US. 174 LEAFLETS OF WESTERN BOTANY _ [VOL. IV; NO. 6 panthium campanulate to tubular, glabrous within; bractlets generally lack- ing, or occasionally 1—3; receptacle stipitate, 1—5 times as long as thick, villous-pubescent throughout or glabrous at base; petals yellow or white; stamens 5; pistils 6—13. Crevices and ledges of limestone rocks and cliffs in desert mountains, 4000 to 10,000 feet, southern Nevada, northern Arizona, and southeastern California. 2a. Var. typica J. T. Howell, nom. nov. Purpusia Osterhoutti A. Nels., 1. c. Purpusia arizonica Eastw. ex J. T. Howell, Madrofio 2:12 (1930). Non Potentilla Osterhoutiana Th. Wolf, Biblioth. Bot. 71 :200 (1908). Non Potentilla arizonica Greene, Pitt. 1: 104 (1887). Mostly finely glandular- pubescent ; leaflets of basal leaves in 3—5 pairs, usually rather deeply lobed or divided; hypanthium with length equal to, or less than, the width, cam- panulate or hemispheric ; receptacle 0.3—1.5 mm. long, 0.3—1 mm. in diame- . ter; petals yellow; anthers roundish-ovate, about 1 mm. long. Collections. Arizona, all from the Grand Canyon, Coconino County: Bright Angel Trail, Osterhout No. 7103.(RM, type), Goldman No. 2205 (US) ; on rim near Bright Angel, Eastwood No. 5662 (CAS, type of Pur- pusia arizonica) ; El Tovar, Eastwood No. 3778 (CAS) ; Grand View, L. F. Ward in 1901 (US); Cliff Spring near Cape Royal, Peirson No. 7427 (CAS), Collom in 1940 (US) ; limestone ledges at Cape Royal, Eastwood & Howell No. 954 (CAS) ; Pt. Sublime, Collom No. 1318 (US). Nevapa: cliffs at head of Frog Canyon, Grant Range, el. about 10,000 feet, Nye Co., McVaugh No. 6068 (NA) ; Dead Man Springs, Desert Game Range, Clark Co., J. C. Allen No. 38 (NA); Deadmans Canyon, Sheep Mts., Clark Co., Alexander & Kellogg No. 1772 (CAS, DS, UC, US). CALIFORNIA: head of Titus Canyon, Grapevine Mts., Inyo Co., Gilman No. 1812 (CAS, US), No. 1813 (US). Although in the original description of Purpusia Osterhoutu Prof. Nelson noted particularly “the absence of hairs on the re- ceptacle,” the examination of the type disclosed the receptacle to be conspicuously and characteristically pilose. 2b. Var. saxosa (T. S. Brandg.) J. T. Howell, comb. nov. Purpusia saxosa T. S. Brandg., Bot. Gaz. 27: 447 (1899). Non Potentilla saxosa Lemmon ex Greene, Pitt. 1: 171 (1888). Velutinous-glandular with numer- ous elongate nonglandular as well as glandular hairs ; leaflets of basal leaves usually in 2 or 3 pairs; hypanthium tubular-campanulate or vase-shaped, as long as wide or longer, in age more or less constricted above the swollen, truncate or rounded base; receptacle 1.25—2 mm. long, 0.3—0.5 mm. in diameter ; petals white or yellow; anthers elliptic-ovate, 1—1.5 mm. long. Collections. NrvapA: on rocks, Pahroc Range, 4000 to 5000 feet, Purpus No. 6305 (UC, type; DS, US) ; crevices of dry calcareous boulders, south end of Pahroc Range, 15 miles east of Crystal Springs, Lincoln ~ County, Ripley & Barneby No. 4972 (CAS, “petals white’) ; crevices of canyon walls, 4 miles north of Caliente on road to Panaca, Lincoln Co., Train No. 2476 (NA, “flowers yellow’’). MAY, 1945] ; STUDIES IN ROSACE 175 Although Brandegee clearly states that the type of Purpusia saxosa is Purpus No. 6134 collected in the Sheep Mountains, Nevada, I have been forced to conclude that these published data are erroneous and that the type is Purpus No. 6305 collected in the Pahroc Range, Nevada. Of the latter number there are two speci- mens at the University of California, the one in the Brandegee Herbarium indicated as type by an annotation label, the other noted in the hand of Mrs. Brandegee “n. gen. & sp.,” and in the pocket of one of these sheets is a ticket bearing in Purpus’ hand the words ‘“Pahroc Range.’ Although Rydberg states in the North American Flora that Purpusia saxosa is known only from the type locality, the Sheep Mountains, he stamped the Pahroc specimen of Purpus No. 6305 in the U. S. National Herbarium “Examined for the North American Flora.”” A communication from Dr. Maguire states that the only Purpus collection of Pur- pusia at the New York Botanical Garden is No. 6305 and Dr. Lyman B. Smith has written that No. 6305 is represented in the Gray Herbarium but not No. 6134." From all this I have con- cluded that Purpus No. 6305 from the Pahroc Range should be regarded the type-collection of Purpusia saxosa. Plants of the subgenus Purpusia as they grow in the-Pahroc Range and around the Grand Canyon, respectively, exhibit suf- ficient divergence in habit, vestiture, and flowers to be differen- tiated as two closely related but distinct species. Other collections from southern Nevada, however, present intermediate. phases between the extremes and Gilman’s collections from the Death Valley region in California.seem almost as near the Pahroc plants as any others. Although it has been necessary to treat the ex- tremes of variation varietally in the light of present collections | and knowledge, I cannot escape the feeling that, with increased - field knowledge of the group, the type-species of the genus Pur- pusia (which because of nomenclatural priorities must herein be relegated to varietal status) will eventually be restored to full specific standing. Certainly as the matter now rests, var. saxosa is to me an excellent example of a true subspecies and our present plant might properly be treated as such except that I am yet to be persuaded of the general taxonomic usefulness of that category. 176 LEAFLETS OF WESTERN BOTANY _ [VOL. IV; NO. 6 3. NOMENCLATURAL TRANSFERS TO POTENTILLA In order to make available in Potentilla certain taxonomic propositions presented by Dr. D. D. Keck in his recent revision of Horkelia and Ivesia (Lloydia 1 : 75—142,—1938), the follow- ing transfers are listed. In most instances only the name-bringing synonym is given; fuller synonymy may be had by consulting Dr. Keck’s paper. a. PoTENTILLA subgen. HoRKELIA Potentilla congesta (Dougl.) Baill. var. nemorosa (Keck) J. T. Howell, comb. nov. Horkelia congesta Doug]. subsp. nemorosa Keck, op. cit., 108. Potentilla daucifolia Greene var. latior (Keck) J. T. Howell, comb. nov. Horkelia daucifolia subsp. latior Keck, op. cit., 106. Potentilla Douglasii Greene var. capitata (Lindl.) J. T. Howell, comb. nov. Horkelia capitata Lindl., Bot. Reg. pl. 1997 (1837). Potentilla Douglasii Greene var. parviflora (Nutt.) J. T. Howell, comb. nov. Horkelia parviflora Nutt. ex Hook. & Arn., Bot. Beech. Suppl. 338 (1840). Potentilla Hendersonii (Howell) J. T. Howell, comb. nov. Horkelia Hendersonti Howell, Pacif. Coast Pl. Coll. 1887, 2 (1887). Potentilla Lindleyi Greene var. sericea (Gray) J. T. Howell, comb. nov. Horkelia californica C. & S. var. sericea Gray, Proc. Amer. Acad. ,6: 529 (1865). Potentilla marinensis (Elmer) J. T. Howell, comb. nov. Horkelia Bolanderi Gray var. marinensis Elmer, Bot. Gaz. 41: 321 (1906). Potentilla Tilingii (Reg.) Greene var. flavescens (Rydb.) J. T. Howell, comb. nov. Horkelia flavescens Rydb., Mem. Dept. Bot. Columbia Univ. 2: 138 (1898). b. PorENTILLA subgen. IvESIA Potentilla Baileyi (Wats.) Greene var. setosa (Wats.) J. T. Howell, comb. nov. Ivesia Baileyi Wats. var. setosa Wats., Bot. King’s Explor. 91 (1871). Potentilla lycopodioides (Gray) Baill. var. megalopetala (Rydb.) J. T. Howell, comb. nov. Horkelia Gordonii Hook. var. megalopetala Rydb., op. cit., 152. Potentilla lycopodioides (Gray) Baill. var. scandularis (Rydb.) J. T. Howell, comb. nov. Horkelia scandularis Rydb., op. cit., 150. Potentilla sericoleuca (Rydb.) J. T. Howell, comb. nov. Horkelia sericoleuca Rydb., op. cit., 144. Potentilla Tweedyi (Rydb.) J. T. Howell, comb. nov. Ivesia Tweedyi Rydb., N. Amer. Fl. 22: 288 (1908). LEAFLETS of WESTERN BOTANY Available A limited number of copies of volumes I, I, and III at $5.00 each. Single copies of most numbers in these volumes are available at $.40 each. LINDLEY’s GLOSSOLOGY An excerpt from this most useful work, reissued with an introduction by Alice Eastwood. $.50. ALIcEe Eastwoop’s Key to the Common Families of Flowering Plants in California and Guide for the Analysis and Description of Flowering Plants. $.25. Address: Joun THomas HowELi California Academy of Sciences, Golden Gate Park San Francisco, California Vot. IV No. 7 LEAFLETS of WESTERN BOTANY CONTENTS PAGE Some Interesting Plants of Malheur County, Oregon. . . 177 Morton E. Peck Tentative Key and Discussion of the Lupines eRene coy Te Calyx Y). ello Wily he.) ore Nar dele ol he Auicr Eastwoop peeivew Carex From Colorado . « .. se ie ke we wt 194 F. J. HERMANN SAN Francisco, CALIFORNIA Avcust 13, 1945 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. ee UU AL OU A ad Raat pvngaaypouaptaaguoanpeagyaeagteag engages ascot cay Owned and published by Autce Eastwoop and JoHN THomas Howe ee ae ee AUGUST, 1945] | SOME INTERESTING OREGON PLANTS 177 SOME INTERESTING PLANTS OF MALHEUR COUNTY, OREGON BY MORTON E. PECK Willamette University, Salem, Oregon The observations upon which the following notes are based were made chiefly on four collecting trips into Malheur County, namely, July 8-12, 1938, May 4-9, 1940, June 10-20, 1942, and June 1-11, 1943. A few are from earlier notes. The flora of Malheur County is probably less known than that of any other section of Oregon. The southern half and much of the northern are mainly desert with only a few widely scattered ranches. Most of the population is in the northeastern part where there is extensive irrigation. The proximity to parts of Idaho and Nevada that have not been very thoroughly explored botani- cally should lend additional interest to the observations. Many of the species here included do not occur very far on our side of the border, and many others are sure to be added in the future. A thorough exploration of the remarkable canyon of the Owyhee River cannot fail to yield most interesting results. A detailed account of the physical features of Malheur County would greatly enhance the value of these observations, but would call for more space than can here be allowed. It is hoped that the notes may at least stimulate further investigation of this neglected region. ALLium ANCcEPS Kell. In clayey soil 15 miles north of Mc- Dermitt and in a similar situation 15 miles south of Sheaville. Not otherwise known to us from Oregon except for specimens taken by Mrs. Lilla Leach on Hart Mt., Lake County, June 11, 1937. Allium nevadense Wats. var. macropetalum Peck, var. nov., scapo _robustiore plerumque breviore; folio scapo longiore; perianthio longiore 9—11 mm. longo; staminibus perianthium subzquantibus; cristis ovarii 1 mm. altis late rotundatis vel truncatis spe denticulatis. Scape stouter than in the species and mostly shorter ; leaf surpassing the scape ; perianth-parts longer, 9—11 mm. long. ; stamens nearly equaling the perianth ; crests of the ovary 1 mm. high, broadly rounded or truncate, often denticulate. Type: Peck No. 21704, in low sagebrush 17 miles north of McDermitt, Malheur County, Oregon, June 3, 1943. Differs Leaflets of Western Botany, Vol. IV, pp. 177-196, August 18, 1945. 178 LEAFLETS OF WESTERN BOTANY _ [VOL. IV, NO. 7 from the species especially in the relatively longer leaf and larger flowers. The variety occurs infrequently from Jordan Creek southward at least to the Nevada line. SALIX SESSILIFOLIA Nutt. Specimens were collected in Owy- hee Canyon a short distance above Rome. The species has an erratic distribution. Its regular range seems to be from the Umpqua Valley northward to Vancouver Island, but we have it from southern Lake County, and now from this isolated station. Eriogonum novonudum Peck, spec. nov., suffrutescens ; ramis ligneis circa 1 dm. longis decumbentibus copiose ramosis; foliis spathulatis vel anguste obovatis utrinque dense albo-tomentosis 2—3 cm. longis in petiolis multo longioribus, pedunculis 2.5—3.5 dm. altis glaberrimis glaucissimis ; radiis umbelle plerumque tribus 8—15 cm. longis prope erectis; involucris capitatis capitibus compositis capitulis 2—5, involucris anguste turbinatis 3—5 mm. altis lobis 6—8 latis rotundatis scariosis; floribus viridescenti- albis 3—3.5 mm. longis glabris segmentis omnibus similibus. Plant suffrutescent, the main branches of the shrubby base about 1 dm. long, decumbent and copiously branched ; leaves crowded at the base of the scape-like flowering stems and on very short sterile shoots, spatulate or narrowly obovate, densely white-tomentose on both faces, 2—3 cm. long, on petioles often more than twice as long; peduncles 2.5—3.5 dm. high, slender, glabrous and very glaucous, each bearing a 3-rayed umbel sub- tended by 3—5 small, scarious, connate bracts, the rays 8—15 cm. long and closely ascending, the involucres sessile or nearly so in 2—5 small, nearly sessile heads, forming the capitate cluster of the ray; involucre narrowly turbinate, 3—5 mm. high, with 6—8 broad, rounded, largely scarious, erect teeth, glabrous and distinctly ribbed; flowers dull greenish-white; perianth 3—3.5 mm. high, glabrous, the outer and inner segments essentially alike; stamens long-exserted; ovary and fruit glabrous. Type: Peck No. 21216, on a sandy clayey slope 10 miles south of Adrian, Malheur County, Oregon, June 13, 1942. This species is perhaps related to E. nudum Dougl., but differs in many respects, most conspicuously in the much more woody | stem and the uniformly 3-rayed umbel with very long, closely ascending rays capped with the small, dense, compound heads. Several collections were made, all in the same general locality. Eriogonum ochrocephalum Wats. var. calcareum (Stokes) Peck, comb. nov. E. ochrocephalum subsp. calcareum Stokes, Gen. Eriog. p. 92. A well-marked local variety. Our material was taken probably very close to the type-locality. The name, perhaps suggested by the original collector’s field notes, is an unfortunate choice. The plant grows a few miles north of Harper on a series of rather high, steep hills of a pale gray, very soft, AUGUST, 1945] | SOME INTERESTING OREGON PLANTS 179 somewhat chalk-like material, not in the least calcareous. A sample submitted to Professor Herman Clark, Department of Geology, Willamette University, was identified as tufaceous and diatomaceous, apparently from a deposit in an ancient lake bed. Mirasitis BrceLtovir Gray var. RETRORSA (Hel.) Munz. Plentiful locally on the nearly vertical walls of Owyhee Canyon, 10 to 15 miles below the great dam. We have no other Oregon record of this plant. ARENARIA BurkEI Howell. Found sparingly on the white tufaceous hills 5 miles north of Harper. Rare in Oregon. CERATOCEPHALUS FALCATUS (L.) Moench. It is worth noting that this Old-World species has become well established about McDermitt. It is plentiful in places and is steadily in- creasing from Wheeler County northward and along the Columbia River. BEcKWITHIA ANDERSONII (Gray) Jeps. A large colony was discovered on a high clayey slope 15 miles north of McDermitt. Drasa Dovuctasit Gray. Plentiful locally on sterile stony ridges 15 to 20 miles north of Jordan Valley. It also occurs on the eastern slope of the Steens Mts. and on Stinking Water Mt. in eastern Harney County. : Puysarta CHAmBeErsII Rollins. A good-sized colony was found on a sandy knoll just within the Idaho-Oregon line 20 miles north of Jordan Valley. LEPIDIUM MONTANUM Nutt. var. CANESCENS (Thell.) C. L. Hitche. One of the characteristic species of the low alkaline flats along McDermitt Creek just west of McDermitt. It was col- lected also along Crooked Creek 25 miles southwest of Rome. We have it from Narrows, Harney County, and from the south- eastern part of the same county. DESCURAINIA PINNATA ( Walt.) Britt. vars. FrLIPEs (Gray) Peck, HALICTORUM (Cocker.) Peck, pARADISA (Nels. & Ken.) Peck, and netsonir (Rydb.) Peck. From Vale and Ontario southward abundant intergradations between these four entities occur. They are distinct and not difficult of recognition when typi- cally developed, but their boundaries are here quite lost. This fact was noted by Detling in his excellent treatment of the genus. STREPTANTHUS PILosus Wats. From Harper and Adrian southward this species occurs rather infrequently, usually in 180 LEAFLETS OF WESTERN BOTANY __ [VOL. IV, NO. 7 dwarf sagebrush, where it is often conspicuous by its tall, much- branched habit. STANLEYA PINNATA (Pursh) Britt. One large colony was found on stony, clayey ground 5 miles south of Rome. It has been collected also in southeastern Harney County, but is rare in Oregon. CLEOMELLA HiLi~MANI Nels. From Adrian southward to Rome, but especially in the Sucker Creek country, during May, many of the pale ochraceous-gray slopes of the high broken hills and ridges are mantled in bright yellow, often distinguishable at a distance of two or three miles. These fine displays of color are due mostly to a dense growth of Cleome platycarpa and Cleomella Hillimani, the former usually, but not always the more abundant. These peculiar deposits of Eocene clay, very viscid and slippery when wet, have usually a rather sparse covering of vegetation, but offer excellent collecting ground, as they are frequently the favorite habitat of rare and local plants. CLEOMELLA PLOCASPERMA Wats. Moderately plentiful on the low alkaline flats west of McDermitt. It occurs in similar situ- ations along the eastern base of the Steens Mts. and southward in Harney County. Rises Goopprneir Peck. Found sparingly on the nearly vertical walls of Sucker Creek Canyon. This is the southernmost point known for the species. POTENTILLA PARADOXA Nutt. On wet shores of Snake River in the extreme northeast corner of the county. Lupinus unciALIs Wats. A good-sized colony of this rare and most diminutive of lupines was found on a high gravelly slope 15 miles north of McDermitt. Evidently this is the first record of the species for Oregon. Astragalus Jessiz Peck, spec. nov., perennis humilis a basi ramosis- sima ; caudice brevi ligneo; caulibus crassis rigidis 6—12 cm. longis adscen- dentibus et patentibus dense albo-strigosis; foliis 1—2 cm. longis dense utrinque strigosis, stipulis scariosis modo ad basin conjunctis ad apices longas acutas induratas abrupte contractis valde patentibus, foliolis 5—7 coactis 8—13 mm. longis anguste lanceolatis involutis rigidis in apices longas acutas induratas flavas excurrentibus ; pedunculis brevissimis, floribus ple- rumque 3, pedicellis 1 mm. longis; calyce 4 mm. longo dentibus subulatis tubum paulo superantibus ; corolla 5 mm. longa albida, carina apicem versus purpurea; legumine (immaturo) lanceolato-oblongo compresso dense stri- goso, seminibus 2. AUGUST, 1945] | SOME INTERESTING OREGON PLANTS 181 Stems several or many from a short woody caudex, stout, rigid, ascend- ing and spreading, 6—12 cm. high, densely white-strigose ; leaves 1—2 cm. long, pubescent like the stem, the scarious stipules slightly connate at base, abruptly contracted to strongly spreading, slender and pungent tips often 2—3 times as long as the body, the leaflets 5 or occasionally 7, crowded, 8—13 mm. long, surpassing the petiole, narrowly lanceolate, strongly invo- lute, very stiff, drawn out to long, rigid, very pungent, yellow tips ; peduncles mostly 3-flowered, very short, the pedicels about 1 mm. long; calyx 4 mm. long, the subulate lobes a little longer than the tube; corolla 5 mm. long, whitish, the keel purple-tipped ; pod (immature) lance-oblong, compressed, densely strigose, 2-seeded. Type: Peck No. 21220, from a high gravelly slope 10 miles south of Adrian, Malheur County, Oregon, June 13, 1942. This new member of the subgenus Kentrophyta differs from any other known to us in the stout, rigid, brittle stems, the bushy habit, and coarse, very pungent foliage. It inhabits an imperfectly known territory from which additional collections may be expected. It is named for Mrs. Jessie G. Peck, who discovered the colony from which the type was taken. Astragalus solitarius Peck, spec. nov., perennis gracilis 3—5 dm. altus, caulibus solitariis vel paucis debilibus plantis vicinis sustentis sparse ramosis flexuosis pilis minutis incurvatis canescentibus; foliis cum petiolis 5—10 cm. longis, foliolis 7—13 utrimque canescentibus anguste linearibus obtusis remotis 1—2 cm. longis; pedunculis aliquantum robustis foliis multo longi- oribus ; racemis laxe 10—15-floris ; calyce nigro-crinito cylindrico-campanu- lato 3—4 mm. longo dentibus triangularibus brevissimis; corolla 7—8 mm. longa luteo-viridi purpureo-tincta; legumine angusto-oblongo valde com- presso suturis utrisque exsertis dense puberulo prope recto vel valde arcuato deorsum curvato cum stipo 2—2.5 cm. longo, stipo 6—8 mm. longo. Stems arising singly or in small clusters from a deep taproot, 3—5 cm. high, slender and weak, mostly supported by other vegetation, sparingly branched, flexuous, canescent with minute incurved hairs; leaves, including the petiole, 5—10 cm. long, the leaflets 7—13, canescent like the stem on both surfaces, narrowly linear, obtuse. remote, 1—2 cm. long; peduncles relatively stout, up to 2 dm. long including the raceme; racemes loosely 10—15-flowered ; calyx black-hairy, cylindric-campanulate, 3—4 mm. long, the triangular lobes less than a third as long as the tube; corolla 7—8 mm. long, yellowish-green tinged with purple; pod narrowly oblong, strongly compressed, including the stipe 2—2.5 cm. long, the latter 3 times as long as the calyx, acute at both ends, densely puberulent with incurved hairs, nearly straight to decidedly arcuate and turned downward on the hori- zontally spreading pedicel. Type: Peck No. 21726, among sagebrush, 15 miles north of McDermitt, Malheur County, Oregon, June 4, 1943. Plants widely scattered, very inconspicuous, growing close to the base 182 LEAFLETS OF WESTERN BOTANY __ [VOL. IV, NO. 7 of the sagebrush clumps and supported by the branches. It be- longs to the subgenus Homalobus, perhaps most nearly related to A. stenophyllus T. & G. from which it differs, however, in numerous characters. ASTRAGALUS ALVORDENSIS Jones. On high sandy slopes near Rome and at Sheaville. Very local. . ASTRAGALUS NupIsILiguus Nels. Collected near Nyssa and also at Harper. The caudex, stems, and peduncles are sometimes elongated, the stems up to 2 dm., the peduncles up to 1 dm.; the flowers are often strongly purple-tinged and the pods up to 3 cm. long. The species, however, seems sufficiently well marked. ASTRAGALUS TETRAPTERUS Wats. Ona high dry slope near Rome and at Blue Mountain Pass, 25 miles north of McDermitt. Though far out of the usual range of the species, our specimens appear quite typical. ASTRAGALUS CINERASCENS (Rydb.) Tidest. On sandy hills near Jordan Valley; otherwise apparently known only from the type-locality, near Harney Lake, Harney County. AstracALus GEyert Gray. On stony clayey slopes near Rome. Not noted elsewhere. ASTRAGALUS WEISERENSIS (Jones) Abrams. Rather plentiful over about the northeastern fourth of the county. ASTRAGALUS MALHEURENSIS Hel. Apparently rare. Two collections were secured, one near Vale and one in the extreme northeastern corner of the county. ASTRAGALUS MULFoRD# Jones. A single specimen was col- lected on a shady slope in Owyhee Canyon 10 miles below the Owyhee dam. TRIBULUS TERRESTRIS L. This evil weed was first recorded from Oregon at Robinette, Baker County, in 1938. It has since appeared at Umatilla, Umatilla County, and at Ontario, Malheur County. It seems in a fair way to spread over much of eastern Oregon. : SPH#RALCEA COCCINEA (Nutt.) Rydb. This occurs fre- quently from Jordan Valley southward. In the extreme southern part of the county, near McDermitt, numerous examples of what are apparently hybrids between this species and S. Munroana were found. Both were locally plentiful. AUGUST, 1945] | SOME INTERESTING OREGON PLANTS 183 Mentzelia mollis Peck, spec. noy., annua humilis erecta 5—10 cm. alta; caule crasso ramoso pilis minutis vix scabris dense puberulo; foliis ad basin confertis oblongo-lanceolatis vel oblanceolatis obtusis plerumque integris vel obscure undulatis 3—5 cm. longis caule minus dense puberulis; floribus subsessilibus ad terminos ramorum foliis reductis subtentis ; segmentis caly- cis lanceolatis acutis 4—6 mm. longis; petalis anguste obovatis flavis 10—12 mm. longis; staminibus 40—50, filamentis filiformi-subulatis ; stylo stamini- bus multo longiore, stigmate capitato; capsula (immatura) quadrilatera striata circa 2.5 cm. longa. A low annual 5—10 cm. high; stem erect, stout, freely branched, dark colored, densely puberulent with fine glochidiate but only slightly scabrous hairs; leaves crowded below, oblong-lanceolate to oblanceolate, obtuse, nearly entire or obscurely undulate, rarely with 1 or 2 low teeth, sessile or on broad short petioles, 3—5 cm. long, less densely puberulent than the stem; flowers nearly sessile, mostly in small clusters at the ends of the branches, subtended but not at all concealed by reduced leaves; calyx- segments lanceolate or lance-ovate, acutish, 4—6 mm. long ; petals narrowly obovate, clear yellow, 10—12 mm. long; stamens 40—50, a little shorter than the petals, the filaments filiform-subulate; style much surpassing the stamens, the stigma capitate; capsule (immature) 4-sided, longitudinally striate, about 2.5 cm. long. Type: Peck No. 20631, on a clayey slope 1 mile north of Rockville, Malheur County, Oregon, May 7, 1940. An interest- ing addition to the section Trachyphytum, clearly distinguished by its low stout stem, nearly entire leaves, fine, hardly scabrous pubescence, and relatively large flowers. It is probably very local. EL#AGNUS uTILIs Nels. Not uncommon on low alkaline flats west of McDermitt. In Harney County it is found sparingly near Alvord Lake, along the east base of the Steens Mts., and south- ward to Nevada. CENOTHERA PALMERI Wats. Rather plentiful in sandy soil a few miles west of Vale; apparently rare elsewhere. CENOTHERA MULTIJUGA Wats. An abundant and very showy ‘evening primrose, especially in the southern half of the county. It occurs as far north as Harper, and also near Fields, Harney County. CENOTHERA ALyssoIpEs H. & A. var. viLLosa Wats. A densely villous form of this variety was found in some abun- dance about 12 miles southwest of Vale. Its blooming period is considerably later than that of the species, from which it differs strikingly in appearance. PTERYXIA PETREA (Jones) C. & R. Found in considerable abundance on the dry, nearly vertical walls of Owyhee Canyon 6 miles above Rome. 184 LEAFLETS OF WESTERN BOTANY __ [VOL. IV, NO. 7 CyMOPTERUS ACAULIS (Pursh) Raf. Frequent on clay slopes of the lower Owyhee and Sucker Creek section. Cymopterus Watsoni (C. & R.) Jones. Collected in two localities, at Blue Mountain Pass and on a stony slope 15 miles north of McDermitt. The species has been known hitherto only from Nevada and Utah. ASCLEPIAS CRYPTOCERAS Wats. Occurs sparingly on high dry hills from northern Malheur County southward. There is a record from Robinette, Baker County. PHLOX AUSTROMONTANA Cov. Specimens were taken on high stony slopes a few miles north of McDermitt. The species is apparently rare and local in Oregon. GYMNOSTERIS NUDICAULIS (T. & G.) Greene. Found in the valley of Malheur River near Harper, northward to Westfall and eastward to Vale. A few miles west of Vale it is very abundant and presents remarkable color variations, from nearly white to deep lavender and bright yellow. The smaller species, G. parvula (Rydb.) Hel., has a much wider range in Oregon, occurring as far west as northwestern Harney County. LANGLOISIA SETOSISSIMA (Gray) Greene. Small colonies were found in various places on dry stony slopes in the lower Owyhee Canyon and the Sucker Creek section. CoLLOMIA MACROCALYX Eastw. Collected on a high clayey slope about 15 miles north of McDermitt. So far as we are aware this is the first record of this very distinct species since the dis- covery of the type. PuHaceLiA Ratrani Gray. This plant has a remarkably erratic distribution in Oregon. We have it from the southwestern part of the state (Josephine County), from the canyon of the: Deschutes River, Wasco County, and from Owyhee Canyon a few miles above Rome, but from no intermediate stations.. It also occurs in Idaho and California. Salvia carnosa Dougl. var. gracilior Peck, var. nov., erecta ramosissima 3 dm. alta ramulis ultimis tenuissimis; foliis permultis 7—15 mm. longis; bracteis 3—5 mm. longis apicem versus plumbeis vix ciliolatis. Erect, copiously branched, about 3 dm. high, the ultimate branches almost filiform; leaves very numerous, including the petiole under 15 mm. long; floral bracts 5 mm. long or less, plumbeous toward the tip, scarcely ciliate. AUGUST, 1945] | SOME INTERESTING OREGON PLANTS 185 Type: Peck No. 21810, on a dry rocky bank of Owyhee Canyon 6 miles above Rome, Malheur County, Oregon, July 9, 1943. Strikingly different from the ordinary form of the species, which occurs in the same neighborhood. The very slender branches, small leaves, and very small bracts are distinctive. VERONICA ANAGALLIs L. Established along irrigation ditches about Ontario. APLOPAPPUS INTEGRIFOLIUS Porter. A considerable colony of this large and conspicuous Aplopappus was found on a slightly moist slope 2 miles west of Brogan. A characteristic species of the Rocky Mountains, this is far out of its normal range. TowNnsenpi1a Wartsont Gray. This is the common Town- sendia throughout Malheur County. The characters on which it is separated from T. florifer are mainly the more strongly spatu- late cauline leaves, the shorter rays, and the very short outer pappus. The best of these, so far as our material shows, is the character of the pappus, though T. florifer often has the outer pappus not more than half as long as the inner. Specimens with typical T. Vatsoni pappus sometimes have very long rays and the cauline leaves may approach those of otherwise typical T. florifer. Stylocline psilocarphoides Peck, spec. nov., planta omnino laxe tomen- tosa erecta vel depressa parce vel copiose ramosa 3—5 cm. alta; foliis 4—8 mm. longis linearibus vel oblongis maxima ex parte sub capitulis congestis ; capitulis sessilibus in capitibus globosis; floribus fertilibus 8—12; paleis amplectentibus exterius albo-tomentosis 3 mm. longis apicibus hyalinis brevibus acutis marginibus involutis supra medium late hyalinis; floribus staminiferis 2 vel 3 paleis hyalinis; pappo nullo. A small annual herb, erect and proliferously branched with few filiform, naked branches, sometimes more or less depressed and more copiously branched, gray-tomentose throughout, 3—5 cm. high; leaves alternate or a few of the lower opposite, 4—8 mm. long, the lower cauline few, linear, the upper closely rosulate about the clusters of heads, mainly linear-oblong, acute or slightly acuminate, not mucronate; heads in small, dense, globose clusters; pistillate flowers 8—12 to a head, the enfolding scales white- woolly dorsally, about 3 mm. long, the acutish, scarious tips one-fourth as long as the body but continued as a hyaline wing down the margins to the middle ; staminate flowers 2 or 3; often with as many minute neutral flowers, each subtended by an ovate or lanceolate scale equaling or shorter than the flower ; pappus wholly wanting. Type: Peck No. 21730, on a dry slope about 15 miles north of McDermitt, Malheur County, Oregon, June 4, 1943. This obscure little plant seems most closely related to S. micropoides, 186 LEAFLETS OF WESTERN BOTANY _ [VOL. IV, NO. 7 from which it differs in the smaller size, the acute rather than mucronate leaves, the fewer flowers to a head, the smaller, less densely woolly pistillate scales with much shorter scarious tips that continue downward as a marginal wing, and in the complete absence of pappus. Two other collections have been made by the writer, both from Malheur County, namely, near Brogan, June, 1910 (Peck No. 865), and 8 miles west of Vale, May, 1941 (Peck No. 20677). Both were at first mistaken for a species of Psilo- carphus, which the new species closely resembles superficially. DimereEsiA Hower Gray. This remarkable little plant is more widely distributed in Oregon than has generally been sup- posed. We have it from near Prairie City, Grant County, from Stinking Water Mountain and Steens Mts., Harney County, from the vicinity of Lakeview and of Adel, Lake County, and in 1943 we collected it on a high sandy slope 15 miles north of McDermitt, Malheur County. CH2#NACTIS MACRANTHA D. C. Eat. Collected once, near Adrian. It has also been found near Fields, Harney County. Cuanactis Cusicxit Gray. From near Adrian well to the south of Rome this species is to be met with, sometimes in immense abundance, on soft clay slopes. CREPIS RUNCINATA T. & G. In Owyhee Canyon a few miles above Rome. The typical form of the species does not seem to have been previously found in Oregon. TENTATIVE KEY /AND DISCUSSION OF THE LUPINESS WITH SPURRED CALYX BY ALICE EASTWOOD This most interesting group of lupines with spurred calyx extends from the Pacific Coast to the Rocky Mountains and from beyond the northern boundary of the United States to the southern. My aim has been to attempt a key to the species that have been definitely named as species and that have been vari- ously treated as species or varieties of these species. It is not in any sense a revision, but notes are added in explanation and enti- Explanation of Plate. Sheet in Herb. Lindley., Cambridge University, Eng- land, on which are mounted the type-specimens of Lupinus lepidus Doug}. (left) and L. laxiflorus Dougl. (right). In the lower left hand corner are the data: ‘“‘N. W. Coast Amer. Douglas 1826.” iris michele sone oP are i EM RA a AUGUST, 1945] LUPINES WITH SPURRED CALYX 187 ties that seem distinct are described as new species. It is to be expected that over so great an area, many species may occur and probably many more will be added with greater field knowledge and more ample collections. That taxonomists will ever agree is not to be expected. The species fall into two groups, one in which the upper lip of the calyx is very short, is not concealed by the banner, and in the mature flower stands out separate from the banner though rarely the tips may. be concealed. The spur is generally con- spicuous. In the other, the flower has a different shape since the upper lip of the calyx is decidedly concealed by the banner. In this group, the spur is always short. TENTATIVE KEY 1. Upper lip of calyx short not concealed by the banner, the tip some- Ca 71M aC (G10) ee nS Re eT came see eo LE Ae amen ees car eee aE a 1S 2 1. Upper lip of calyx definitely concealed by the banner......................---- 9 2. Foliage bright green, flowers small, blue (rarely white) -......0.00.00...... 3 2. Foliage not bright green but greenish, indument silvery or silky........ 4 3. Both leaf surfaces lightly appressed-sericeous; upper part of keel CENT ME ANNs NS 2 ON et Se LO Sino aD acters Ree EC L. silvicola 3. Upper leaf surface glabrous, lower slightly appressed-sericeous ; lower-part ot keel ciliate) S.A. 2 oe L. pseudoparviflorus eoibiowers: biuecor purplish; sput small. et eee 5 4. Flowers white or varicolored; spur prominent.........2....020.....c0-0-ssesese--e 6 5. Racemes loosely flowered, flowers about 10 mm. long; keel slightly eihtates leatlets) Narrows. scoot eee L. laxiflorus 5. Racemes more densely flowered, flowers 15 mm. or more long; keel long-ciliate about the middle; leaflets wider.................... L. arbustus SEI EGE le) |: oy teat an AUR na RIGA re ORO AS ei ORE 7 ReEIRIVICES, VATICOLONEU 05-84 cue, ndocchocns cesta nae nee eatin on een 8 7. Racemes densely flowered, flowers about 15 mm. long, 4 mm. wide, upper part of banner and upper margin of wings sericeous............ Fe Oa nae OEE DY PELTED RR EEN Wty SCM Die. OO er AE L. calcaratus 7. Racemes loosely flowered, flowers about 12 mm. long, wings glabrous near upper edge; calyx 2 mm. wide; leaflets about 3 mm. wide ae Leen Sate ee epee RTS eee OAS, iat ABUL Cd eS eine ete ae ok L. Noldeke 8. Flowers about 15 mm. long, similar to those of L. calcaratus except InteOlors WarittionS Okt. sare tees ea ees L. calcaratus 8. Flowers 10 mm. long, with yellow blotch on standard turning purplish or rose; tips of the short teeth of the upper lip of the calyx often eqneealed, by -the bannet. ie oUt nea L. proteanus Stenis apparently: not branching. ee ee 10 Uresreras: rater cs ts Sod ale PN i ie eee 11 © 188 LEAFLETS OF WESTERN BOTANY __ [VOL. IV, NO. 7 10. Stems decumbent and leafy, with incipient branches; foliage with silvery-appressed pubescence ; flowers small, blue.....:......L. aduncus 10. Stems erect ; foliage greenish; flowers small, purplish, soon becoming Ein) 115 Sa Oe ARN MAE ke ABN OBA. pt RateelMee ENG 0 ar aeC L. wenatchensis 11. Plants white with dense, appressed, silvery-silky indument.................. 12 11. Plants greenish, indument less silvery... ..2.0.c...2o.---eeeececeesceeeceeseee-eoee 14 12. Branches corymbose; flowers about 15 mm. long; banner glabrous ss RRR Ss ML At eat BA coat cee oe Se RSS Tee a Si L. Hendersom 12. Branches spreading; flowers smaller ; banner sericeous..................---.- 13 13." Lower ‘lip of calyx faleately spreading: uso OR ee L. caudatus 13. Lower lip of calyx appressed to the standard....................-.-- L. inyoensis 14. Stems densely leafy, leaflets acute..............2--c20c2ecceceeceseeeeeoeeoee L. Helleri 14. Stems not densely leafy, leaflets obtuse.........000..0c2ceceeeeceeeeeee es L. Cutleri LUPINUS LAXIFLoRUS Dougl., Bot. Reg. vol. 14, pl. 1140. This was described and illustrated by Lindley, who states that it was “found by Mr. Douglas in dry, open, gravelly plains, about the great rapids of the River Columbia, where it is very common, forming patches of considerable extent, occasionally acquiring a suffruticose habit. The flowers are blue mixed with pink.” The specimen named and collected by Douglas at the herbarium at Kew is labelled as “very common at Fort Vancouver along with L. arbustus.” A photograph of the Douglas specimen in Lindley’s herbarium, now at the Herbarium of Cambridge University, England, shows a plant quite different from the illustration. In this photograph, reproduced in the accompanying plate, the spur is small but dis- tinct and the upper lip of the calyx is plainly visible apart from the banner. In Lindley’s illustration, the calyx is not spurred and the upper lip of the calyx is completely hidden by the banner. In that illustration spreading hairs are depicted, while in the speci- men, the pubescence is appressed except in the inflorescence. There are other differences not apparent in the illustrated plant. Agardh, in his Synopsis Generis Lupini, page 27, evidently ex- amined Douglas’ specimen, as his description agrees with my examination of the same. He describes the keel as ciliate, said to be naked in Lindley’s description. In Flora Boreali-Americana (1: 164), Sir William Hooker was so impressed by the differ- ences that he named as var. a the specimen collected by Douglas and as var. £8 the illustration in the Botanical Register. LUPINUS ARBUSTUS Dougl., Bot. Reg. vol. 15, pl. 1230. The illustration was made from a plant in the garden of the Royal i lad AUGUST, 1945] LUPINES WITH SPURRED CALYX 189 Horticultural Society but no specimen was preserved in Lindley’s herbarium. The label on Douglas’ specimen at Kew reads “under the shade of solitary pine and oak trees‘near Fort Vancouver and in the plains of.the Multnomah” (Willamette River). The original description is as follows: L. arbustus; perennis, floribus alternis pedicellatis bracteolatis, calycis labio superiore bifido: inferiore integro acuto, foliolis 7—13 obovato- oblongis utrinque sericeis, leguminibus 3—5-spermis. Douglas journ. ined. Caulis teres, albus, glaber, subdecumbens, pedalis v. sesquipedalis. Foli- ola 7—13, oblonga, utrinque parce sericea; stipulis parvis, subulatis. Flores alterni, v. obscure verticillati. Pedicelli breves. Calyx villosus; labio superiore leviter bifido, obtuso, inferiore integerrimo, acuto. Bracteole minim, deciduze. Vexillum obcordatum, ceruleum, medio purpureum, majusculum. Alz et carina rosee ; hac ciliata. Legumen apice latius, 3—5- spermum ; semina parva, alba. Douglas. I made the following notes on the Douglas type at Kew: flowers about 18 mm. long ; upper calyx-lip 5 mm. from spur to tip, lower 7 mm., entire calyx appressed-silky ; keel long, ciliate about the middle; banner somewhat sericeous about the middle; petioles about 15 cm. long; leaflets 5 cm. long. Lupinus elegantulus Eastwood, Leafl. West. Bot. 3:20. This is too near L. arbustus. Type-locality: Fandango Pass, Warner Mts., Modoc Co., California. Lupinus sILvicota Heller, Muhl. 6:81. Type-locality : Summit, Placer Co., California. This species extends from Placer, Eldorado, and Nevada counties to Sierra, Plumas, Shasta, and Modoc counties. Occasionally a white-flowered plant is seen amid the blue-flowered ones. It is the common lupine of middle elevations with spurred calyx, small flowers, and bright green leaves. LUPINUS PSEUDOPARVIFLORUS Rydb., Mem. N. Y. Bot. Garden 1:232. Type-locality: Bridger Mts., Montana, in open woods at 2000 to 2300 m. Specimens in Herb. Calif. Acad. Sci. from Idaho, Montana, Wyoming, and the Wallowa Mts. in Ore- gon, represent this species. Although some are simple-stemmed and some are branching and the flowers vary somewhat in size, all agree in the characters in the key. LUPINUS CALCARATUS Kell., Proc. Calif. Acad. Sci., 2: 195, fig. 60. This was probably collected somewhere in Washoe Co., Nevada. A specimen in Herb. Calif. Acad. Sci. labelled by 190 LEAFLETS OF WESTERN BOTANY _ [VOL. IV, NO. 7 Kellogg ‘Andrew Veatch Washoe lupine” may be the type as it agrees with the description. Specimens from Virginia City, Ne- vada, collected by Mary K. Curran (Katharine Brandegee) and identified as L. calcaratus, agree with the Andrew Veatch speci- men. WM. FE. Jones No. 8815 from Franktown, Nevada, is the same. I am considering these specimens as typical. Lupinus variegatus Heller is similar except in the variegated color of the flowers. Specimens from Oregon and Washington, called “harle- quin lupines” by L. F. Henderson, are probably the same. The flowers vary in color through yellow, white, purple, or blue, generally bicolored. These may be the same as L. multitinctus Nelson, the type of which I have not seen and the description of which is inadequate. In all these, the calyx is similar with the large spur, the banner is sericeous on the back, the wings sericeous near the upper edge, the flowers 12 to 15 mm. long, the keel ciliate, broad up to the top where it abruptly curves to the apex. Heller describes the banner and keel of L. variegatus as smooth but in an isotype in Herb. Calif. Acad. Sci. they are as sericeous as in L. calcaratus. Typical L. calcaratus has simple stems. A minor variation occurs in Mono and Inyo counties, Calif., with branching stems and white flowers similar to those of the type. Lupinus NoLpEK# Eastwood, Leafl. West. Bot. 4: 149. Type-locality: near Hot Creek, Mono Co., California. This dainty white-flowered lupine is related to L. calcaratus in the shape and color of the flowers, but in this the flower is more slender and the inflorescence more open. The very narrow leaflets, slender branching habit, and denser indument mark it as distinct. Miss Noldeke noted the fragrance of the flowers. Lupinus proteanus Eastwood, spec. nov. Caules czspitosi ex radice lignea, 45 dm. alti, leviter sericei, supra ramosi et foliacei, basi foliis paucis vel nullis longe petiolatis; foliolis viridescentibus adpresse sericeis supra et infra, anguste lanceolatis, 2—6 cm. longis, 3—6 mm. latis; petiolis foliorum superiorum circa equilongis vel brevioribus foliolis; stipulis fili- formibus, discretis; floribus circa 10 mm. longis, verticillatis in racemis aliquando densifloris; pedicellis brevioribus calycibus; bracteis plerumque persistentibus filiformibus, brevioribus calycibus; labio superiore calycis 2 mm. longo latoque, breviter bidentato, labio inferiore 3—4 mm. longo, in senectute reflexo, calcare 1 mm. longo; vexillo superante alas, primo aureo- maculato, in senectute cesio-purpureo, roseo vel purpureo, circa 10 mm. lato, sericeo infra; alis leviter sericeis prope apicem; carina tecta allis, ciliata ; leguminibus circa 3 cm. longis, 8 mm. latis, laxe et adpresse sericeis. AUGUST, 1945] LUPINES WITH SPURRED CALYX I9I Type: Herb. Calif. Acad. Sci. No. 216883, collected June 10, 1933, on the Victory Highway at Emigrant Pass, Eureka Co., Nevada, Eastwood & Howell No. 231. Other specimens from the same locality are Ripley & Barneby No. 4669 (Herb. Calif. _ Acad. Sci. No. 302705) and Lyman Benson No. 4682 (Herb. Pomona College). The species has also been collected by East- wood and Howell near the junipers east of Wells, Nevada, No. 334, and on the Egan Range, 21 miles east of Ely, White Pine Co., Nevada, by Ripley and Barneby, No. 3605. On all the labels the coloring of the corolla with its conspicuous yellow spot that changes to a rosy or purplish shade is noted. The changes in the color of the flowers suggest the specific name. LuPINUS ADUNCUS Greene, Pitt. 4:132. Type-locality: dry ravines among the sandy hills at Aztec, New Mexico; C. F. Baker No. 433. This is described as a low perennial with terminal race- mes ; flowers small in distinct whorls; keel of the corolla naked. C. F. Baker’s specimen in the Pomona College Herbarium has been taken as typical as well as one with the same number in the Brandegee Herbarium at the University of California. A close examination of the flowers of these typical specimens shows the keel of the corolla to be finely ciliate, not naked as described. The low decumbent habit, small flowers, and finely ciliate keel of the corolla differentiate it from L. Helleri. Inthe Dudley Herbarium at Stanford University the following specimens collected by C. P. Smith agree in general with typical L. aduncus: Aztec, N. Mex., Nos. 3931, 3932; Mancos, Colo., Nos. 3934—3939. LUPINUS WENATCHENSIS Eastwood, Leafl. West. Bot. 3: 174. Type-locality : alpine slopes of Wenatchee Mt., alt. 6800 ft. In this lupine several simple stems arise from a perennial woody root. The small, dark purplish flowers are in densely flowered racemes and soon become tawny. Lupinus HENDERSONI Eastwood, Leafl. West. Bot. 2: 266. Type-locality: Alvord Ranch, east Steens Mts., Harney Co., Oregon. This is similar to L. caudatus Kell. but the flowers are much larger, the spur of the calyx more conspicuous, and the standard smooth on the back. Lupinus cAuDATUS Kell., Proc. Calif. Acad. Sci. 2: 197, fig.61. An unnamed specimen in Herb. Calif. Acad. Sci. labelled with Dr. Kellogg’s large label, bearing only the words “Carson 192 LEAFLETS OF WESTERN BOTANY __ [VOL. IV, NO. 7 Valley,” may be the type. It is a poor specimen. However, several specimens collected by Mary K. Curran (Katharine Brandegee) from Virginia City, Carson, and Franktown, and others from the same county collected by Walter Bryant, M. E. Jones, A. A. Heller, and P. B. Kennedy agree and are undoubt- edly typical. All are white with the dense, appressed, silvery-silky indument, branching stems, standard satiny-pubescent along the middle line of the back, and ciliate keel. Lupinus INYOoENSIS Heller, Muhl. 2% 211. Type-locality in the Sierra foothills west of Bishop, Inyo Co., California, at McGee’s Meadows. Type: Herb. Calif. Acad. Sci. No. 707, Heller No. 8312. This is very much like L. caudatus, the most noticeable distinction being in the lower lip of the calyx. In L. caudatus it is falcately spreading from the flower, in L. m- _ yoensis it is appressed to the flower. Lupinus HELtEr! Greene, Pitt. 4: 134. Type-locality : a can- yon one mile southeast of Santa Fe, New Mexico ; Heller No.3557. This has been included under L. aduncus Greene, from which it differs in erect branching stems, larger flowers, and conspicuously ciliate keel of the corolla. A duplicate of Heller’s collection in Herb. Calif. Acad. Sci. agrees with Greene’s description. It is without distinct branches, though showing incipient ones, and has the flowers as described. Several specimens in the Dudley Her- barium collected in the region adjacent to Santa Fe by C. P. Smith (Nos. 4056—4060, 4062, 4063) show flowers which agree with the type but the stems all branch above. Besides these specimens from New Mexico, the following from Colorado in the Dudley Herbarium agree with L. Helleri in branching stems and similar flowers : Dolores, Montezuma Co., C. P. Smith Nos. 3940, 3941; Ignacio, La Plata Co., C. P. Smith No. 3927 ; Durango, Silverton road, C. P. Smith Nos. 2928, 2929; also doubtfully, from Wagon Wheel Gap, Mineral Co., Hitchcock et al. No. 4125; highway near Poncho Pass, Saguache Co., C. P. Smith Nos. 3904—3906. Lupinus Cutleri Eastwood, spec. nov. Caules perennes alti et sub- robusti, supra ramosi, villosi villis densis brevibus diffusis; foliolis 6—9 zquilongis petiolis, oblanceolatis, obtusis, mucronulatis, 4—5 cm. longis, 5—7 mm. latis, viridibus, supra prope glabris, infra densissime adpresse sericeis, petiolis et foliis junioribus densissime sericeis; floribus 12 mm. longis, 3 mm. inter vexillum et alas, verticillatis in racemis 1—3 dm. longis, equilongis pedunculis; bracteis caducis, equilongis alabastris; pedicellis AUGUST, 1945] LUPINES WITH SPURRED CALYX 193 divaricatis, circa 3 mm. longis; tubo calycis 3 mm. longo, calcare brevis- simo, obtuso; labio superiore ovato, 4 mm. lato longoque, bidentato, tecto vexillo; labio inferiore ovato acuto, paulo longiore; vexillo reflexo, circa 11 mm. lato, medio dorsi sericeo ; alis 7 mm. latis, 11 mm. longis cum stipite, sericeis prope apicem et infra; carina medio 4 mm. lata, abrupte acuminata, apice purpurea, laxe ciliata ex basi ad acumen, tecta alis. Type: Herb. Calif. Acad. Sci. No. 271410, collected June 12, 1938, 18 miles north-northwest of Fort Defiance, Apache Co., Arizona, by Hugh C. Cutler, No. 2141, in whose honor it is named. The species has been collected also in the Lukachukai Mts., Apache Co., Arizona, at an elevation of 7250 ft., Peebles No. 14384. The type was distributed as L. aduncus Greene, but the spe- cies is nearer L. Helleri Greene. The flowers are much larger than those of L. aduncus and the pubescence and habit are differ- ent. The broad oblanceolate leaflets, obtuse at apex, with a different indument, conspicuously differentiate it from L. Helleri. The type-specimen seems to be a branch of a large plant. > £ 7, The following have been described as variants: LUPINUS LAXIFLORUS var. COGNATUS C. P. Sm. in Jepson Man. FI. Pl. Calif., p. 527. Type-locality : Wallowa Mts., Oregon; Cusick No. 3187. LUPINUS LAXIFLORUS f. THEIOCHROUS Robinson in Piper Fl. Wash., p. 358. Type-locality : Rattlesnake Mts., Yakima Co., Washington ; Cotton. Spur prominent. LUPINUS LAXIFLORUS var. LyLEIANUS C. P. Sm. Spec. Lup. sig. 7, p. 105. Type-locality : head of Sheep Creek, Wallowa Co., Oregon; Lyle. Calyx not spurred. LuPINUS LAXIFLORUS var. ELMERIANUs C. P. Sm., ibid., p. 106. Type- locality: Paradise, Wallowa Co., Oregon; Applegate No. 6483. Calyx unknown. LUPINUS LAXIFLORUS var. DURABILIS C. P. Sm., Amer. Journ. Bot. 13:529. Type-locality: Priest River Range, Bonne Co., Idaho; Leiberg No. 2731. Keel naked; calyx scarcely spurred. LUPINUS LAXIFLORUS var. VILLosULUS C. P. Sm., ibid., p. 530. Type- locality: Clover Mts. near Deeth, Elko Co., Nevada; Heller No. 9098. Keel ciliate; calyx spurred. LupINUS CAUDATUS var. SUBTENELLUS C. P. Sm., Bull. Torr. Bot. Club 51: 304. Type-locality: Paulina Lake, Deschutes Co., Oregon; Leiberg No. 591. Calyx not spurred. LuPINUS CAUDATUS var. RUBRICAULIS (Greene) C. P. Sm., Contrib. Dudley Herb. 1:29. (L. rubricaulis Greene Pl. Baker. 3:35.) Type- locality: Crested Butte, Gunnison Co., Colorado; C. F. Baker No. 342. Calyx not spurred. . 194 LEAFLETS OF WESTERN BOTANY _ [VOL. IV, NO. 7 LuPINUS CAUDATUS var. SUBMANENS C. P. Sm. Spec. Lup. sig. 7, p. 106. Type-locality: Antone Creek, 2 miles east of Anthony Lake, Baker Co., Oregon; Lyle. Spur short. LUPINUS INYOENSIS var. ERIOCALYX C. P. Sm. in Jepson Man. FI. Pl. Calif., p. 527. Type-locality : east of the Minarets, Madera Co., California; Congdon. Calyx villous, spurred. LUPINUS INYOENSIS var. DEMISsUS C. P. Sm., Bull. Torr. Bot. Club 51: 304. Type-locality: Wallowa Mts., Baker Co., Oregon; M. E. Peck No. 5329. A NEW CAREX FROM COLORADO BY F. J. HERMANN Bureau of Plant Industry, Soils and Agricultural Engineering, U. S. Department of Agriculture Two collections of a sedge from the Uncompaghre Plateau of western Colorado, referred to the writer for determination by Professor H. D. Harrington of the Colorado A. and M. College, prove to represent a new species of the complex section Ovales. Carex stenoptila Hermann, spec. nov. Dense cespitosa e rhizomatibus crassis ; culmi rigidi, 1.7—7.5 dm. alti, foliis bis longiores basi obtuse superne acute triangulares, sub capitulis scabridi; folia frondosa 2—6 ad basim culmi aggregata, laminis 5—25 cm. longis, 1—3.5 mm. latis, vaginis per- teneris non artis; spice 7—10 gynzcandre oblongo-ovoidee vel ovoider, 6—10 mm. longz, 3—5 mm. late, sessiles, in capitulum terminale 12—19 mm. longum, 8—13 mm. latum conferte, bracteis squamiformibus ; squamz cuneato-oblongz vel oblongo-lanceolate, 3.5 mm. longe perigynia haud occultantes ; perigynia plano-convexa, oblongo-lanceolata vel anguste ovato- lanceolata, 4.5—5 mm. longa, 1—1.5 mm. lata, membranacea, straminea, prominenter sed tenuiter multistriata utrinque, peranguste marginata, ad basim sessilem valde spongiosam contracta, in rostrum tenue serrulatum circa 1.5 mm. longum apice ferrugineum teres haud hyalinum sensim at- tentuata; achzenia oblonga, 1.5 mm. longa, 0.8 mm. lata, substipitata. Densely cespitose from stout, brown rootstocks, the culms 1.7—7.5 dm. high, stiff, about twice the length of the leaves, obtusely triangular and smooth below, sharply triangular and scabridous above; leaves with well- developed blades 2 to 4 to a fertile culm, clustered toward the base, the blades flat or somewhat canaliculate, erect, usually 5—25 cm. long, 1—3.5 mm. wide, scabridulous toward the attenuate apex, the sheaths loose, ven- trally white-hyaline, very thin and early ruptured, prolonged 1—2 mm. at the throat beyond the base of the blade; spikes 7—10, usually 8 or 9, aggre- gated into an oblong to suborbicular head 12—19 mm. long, 8—13 mm. wide, obtuse to acutish at the apex, rounded at the base, the perigynia 10—25, ascending; bracts scale-like; scales cuneate-oblong to oblong-lanceolate, pale to dark chestnut-brown, the pale yellowish-brown midvein conspicuous, obtuse to acute, 3.5 mm. long, not concealing the perigynia; perigynia plano- AUGUST, 1945 | A NEW CAREX 195 convex, oblong-lanceolate to narrowly ovate-lanceolate, 4.5—5 mm. long, 1—1.5 mm. wide, membranaceous, stramineous, with 5—7 slender raised nerves on both faces, very slightly if at all dilated over the achene, very narrowly wing-margined, narrowed toward the rounded, sessile base, very spongy at the base and along the sides of the achene, the body serrulate on the margins above, gradually tapering into a slender beak about 1.5 mm. long, obliquely cut dorsally, dark brown, very shallowly bidentate, serru- late below, smooth and terete but not at all white-hyaline at the tip ; achenes lenticular, oblong, 1.5 mm. long, 0.8 mm. wide, tapering to the substipitate base, glossy, yellowish-brown; style straight, slender, jointed 0.3 mm. above the achene, at length deciduous; stigmas two. CoLtorapo: dry ground, Sneezeweed Camp, alt. 9500 ft., Un- compahgre National Forest, Montrose County, August 16, 1940, David Costello No. 4500 (type, Herbarium of Colorado A. and M. College), and David Costello No. 4501 (Herbarium of Colo- rado A. and M. College, U. S. National Arboretum Herbarium), same data except plants from moist habitat. Carex stenoptila is perhaps most closely related to C. abrupta Mack., a montane species of Washington, Oregon, Nevada, and California, from which it is readily distinguished by its larger (4.5 to 5 mm., rather than 3.75 to 4 mm. long), conspicuously spongy, very narrowly margined, and non-hyaline-tipped peri- gynia. The somewhat boat-shaped, almost wingless perigynia suggest those of C. leporinella Mack., another species of approxi- mately the same geographic range as C. abrupta, but from that it differs not only in its non-hyaline-tipped and much longer beak but also in the very spongy base of the perigynia and in the short pistillate scales which do not conceal the perigynia. The two collections made by Dr. Costello are of markedly different aspect, the type-collection (No. 4500), having few culms only 17 to 24 cm. high with comparatively stiff leaf-blades 5 to 11 cm. long, representing the dry soil phase of the species. In the specimens from a moist habitat (No. 4501), the culms are much more numerous, 53 to 75 cm. long with lax leaf-blades 15 to 25 em. long. The beaks of the perigynia are not conspicuous in the somewhat over-mature heads of this collection as they are in the type. Another plant showing affinity with species of the Pacific States rather than Rocky Mountain relatives has recently been described from the Uncompaghre Plateau. This is Arctostaphylos coloradensis Rollins (Rhodora 39 : 463,—1937 ), allied to A. neva- densis Gray, ranging from middle California northward. 196 LEAFLETS OF WESTERN BOTANY _ [VOL. IV, NO. 7 ' Note on THE DistRIBUTION OF CHAM#SARACHA NANA. According to Jepson (Man. Fl. Pl. Calif., p. 893), the distri- bution of Chamesaracha nana (Gray) Gray is given as the eastern slope of the Sierra Nevada in California from Nevada County northward to southern Oregon at elevations of 5000 to 6000 feet. In June, 1938, Miss Anita Noldeke collected speci- mens of this species at the south end of Long Valley, Mono County, California, at. 7000 feet elevation. The nearest station to this that I know of is over 100 miles to the north in the Lake Tahoe district of western Nevada. I collected the plant in that region in 1938 on the north side of Slide Mt. at an elevation of 7500 feet; and Smiley reports a collection by Kennedy from Mt. Rose at 9000 feet (U. C. Publ. Bot. 9:321). Although several collections have been seen from the west slope of the northern Sierra Nevada, it would seem that farther south the plant is found mostly to the eastward. In July, 1943, however, I found it locally abundant on the west side of Donner Pass in Placer County, California, perhaps not far from the place where Kellogg originally found it (cf. Gray, Proc. Amer. Acad. 10:62).—J. T. Howell. A CorrEcTION. Ina contribution to the literature of Astrag- alus (Leafl. West. Bot. 3 :98,—1942), the writer had occasion to refer to the species first described as Diholcos wucranthus Rydb., and, in accordance with the recent monographic review of Astragalus subgen. Diholcos of Prof. Porter (Amer. Journ. Bot. 26 :690, sequ.,—1939), the name A. haydenioides Porter was used. It should be remarked that this combination is antedated by Astragalus Diholcos Tidestr., Proc, Biol. Wash. 50:20 (1937), also proposed to replace D. micranthus.—R. C. Barneby. CULBERTSON PAINTBRUSH AT FroG LAKE. One of the at- tractive Indian paintbrushes of the Sierra Nevada in California is Castilleja Culbertsonii Greene, which has apparently been reported only as far north as southern Tuolumne County. In July, 1943, I found this subalpine species at Frog Lake on the southeastern slope of Castle Peak in Nevada County (No. 18787), a northward extension of range of about 110 miles.—J. T. Howell. | | | 5 | LEAFLETS of WESTERN BOTANY Available A limited number of copies of volumes I, II, and IIT at $5.00 each. Single copies of most numbers in these volumes are available at $.40 each. LINDLEY’s GLOSSOLOGY An excerpt from this most useful work, reissued with an introduction by Alice Eastwood. $.50. AticEe Eastwoop’s Key to the Common Families of Flowering Plants in California and Guide for the Analysis and Description of Flowering Plants. $.25. Address: Joun THomas HowELi California Academy of Sciences, Golden Gate Park San Francisco, California es Vot. IV LEAFLETS of No. 8 WESTERN BOTANY CONTENTS Noteworthy Plants of Idaho . C. L. Hrrcucock anp J. W. riiaineaeen Studies in Carex—I JoHN THomas Howe. Nomenclatorial Recombinations in Trifolium and Opuntia . LyMAN BENSON A New Species of Asclepias from Nevada . R. C. BARNEBY Additions to the Flora of Utah FRANK W. GouLp Some Weedy Species of Lathyrus Joun THomas Howe. New Varieties of Calyptridium Parryi Joun Tuomas Howe Ly A New Namen Astragalus . Tuomas H. KEARNEY SAN FRANCISCO, CALIFORNIA Novemeser 6, 1945 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. Bor. APU PO INCHES pevngacnyvanaeoaggara tenga cageey acs agge tap eg ecueyoe aceaggeety | Owned and published by Auicr Eastwoop and Joun THomas HoweEtyi NOVEMBER, 1945] NOTEWORTHY PLANTS OF IDAHO 197 NOTEWORTHY PLANTS OF IDAHO BY C. L. HITCHCOCK AND J. W. THOMPSON University of Washington, Seattle Among the collections made in Idaho in recent years by Hitchcock and Muhlick (cited herein as H. & M.) and by Thompson, occasional specimens are to be found that merit no- tation. There follow a few of these: 2” Poa ruPIcota Nash. This species, previously once reported for the state, was collected several times, as shown by the follow- ing collections, the determination of several of which has been verified by Mrs. Agnes Chase: Hitchcock & Martin No. 5809, 5 miles nw. of Mt. Borah, Custer Co.; H. & M. No. 10151, Middle Spangle Lake, Blaine Co.; H. & M. No. 10884, near Castle Peak, White Cloud Range, Custer Co.; H. & M. No. 11201, Leather- man Pass, near Mt. Borah, Custer Co. Three of these four col- lections were made on limestone talus. H. & M. No. 10242, Bald Mt., Elmore Co., was identified as P. rupicola by Mrs. Chase. In our opinion it is intermediate to P. Leibergu Scribn. RM Poa Patrersonir Vasey. J. W. Thompson No. 14083, Boulder Creek Canyon, Blaine Co.; H. & M. No. 11146, Leather- man Pass, near Mt. Borah, Custer Co. These are the first two collections of the species from Idaho according to Mrs. Chase. q** Lewista Kettocei K. Brandg. In 1937, Mr. Thompson made two collections of a white-flowered Lewisia in the Sawtooth Mts., No. 13995, from alpine granitic slopes of high ridge near Stanley Lake, Custer Co., July 28, 1937, and No. 13750, from granitic slopes of Gold Fork Lookout, Valley Co., July 9, 1937. Since no such plants were known to have been reported from Idaho, it was thought the collections represented a new species, but the material was so difficult to study that an effort was made to collect fresh material. In 1944, the plant was collected in abun- dance on the summit of the divide to the north of Dog Creek, about 4 miles n. of Pine, Elmore Co., June 4, H. & M. 8690. In these specimens the flowers are solitary, with 2 sepals and 2 den- ticulate bracts juxtaposed, the petals mostly 6 or 7 and about 15 mm. long, the stamens 10—16 (mostly 14), the styles 4 (or 3), united about 1% length, the capsule thin and membranous Leaflets of Western Botany, Vol. IV, pp. 197-216, November 6, 1945. 198 LEAFLETS OF WESTERN BOTANY © [VOL. Iv, No. 8 and splitting upward by 4 valves, and seeds about 2 mm. long, black and shining, and finely and evenly reticulate-tuberculate. Through the kindness of J. T. Howell this material has been compared with material of L. Kelloggu from California (Howell No. 15551, from El Capitan Trail, Mariposa Co., and East- wood No. 14655, near Long Valley, Plumas Co.). The California and Idaho plants are identical. Thus we have a species known through collections only from central Idaho and the middle Sierra Nevada of California: either L. Kelloggii is an old species on its way to extermination, or (and this we suspect) it is such an evanescent plant that collectors have missed it in the northern Sierra and the desert ranges of Oregon and northern Nevada. iN LEWISIA PYGM#A (Gray) Robinson var. AaRIDORUM Bartlett. In describing this variety Bartlett stressed not only such charac- ters as the glandular calyx and bracts and the glandular-apiculate petals, but also the flower-color, namely, white as compared with red in the species. In 1944 Hitchcock and Muhlick made several collections of the plant in Idaho, e. g., Ryan Peak, H. & M. No. 10649, and Castle Peak area, H. & M. No. 10916, noting that wherever the plant grew, individuals with red, white, and greenish-white flowers occurred together, in that order of abun- dance. Surely, then, the type of var. aridorum, Suksdorf No. 5725, from Mt. Adams, Washington, cannot be considered unique because of the flower color. A collection from the Devil’s Bed- stead, Blaine Co., Idaho, July 28, 1936, Thompson No. 13579, has stipitate-glandular sepals and bracts, the glands being dark red to yellow and the petals are stipitate-glandular at tip. Since there is but one hair at the tip, they might more correctly be called. glandular-apiculate. Thompson’s collection, therefore, is refer- able to var. aridorum, and it would appear that this case more or less parallels L. Kelloggti. However, we feel that var. aridorum is merely the extreme glandular phase of the species and doubt that it is a significant variation. We believe additional plants of such a nature will be found if specimens are examined in the field where their characteristics can more easily be observed. yy CaLttHa CHELIpont Greene vel aff. H. & M. 10942, cirque at nw. base of Mt. Borah, Custer Co. Although it would seem to match Greene’s type in many particulars, there is some doubt as to the identity of this peculiar plant which may be described as follows: NOVEMBER, 1945] NOTEWORTHY PLANTS OF IDAHO 199 Plants 3—10 cm. tall; blades 1.5—3 cm. long, cordate-ovate to ovate- lanceolate, bases truncate to sagittate or auriculate, rather thick and fleshy, bright green; flowering stems 1 or 2, usually bearing a single leaf and two flowers, but the pedicels of flowers and the leaf attached so close to the caudex that they are concealed by the sheathing stipules of the basal leaves, the pedicels simulating scapes 3—10 cm. long; flowers about 2 cm. broad, the sepals about 1 cm. long, linear-oblong, rounded to slightly emarginate, about 8 in number, bright canary-yellow, slightly purplish on back, with 3 principal nerves; stamens about 30; pistils about 12, the fairly mature follicles slightly over 1 cm. long; ovules about 4. RMS. THALICTRUM ALPINUM L. H. & M. No. 11187, wet meadow (but not a bog) at head of West Fork Pahsimeroi River, e. base of Mt. Borah, Lost River Mts., Custer Co., at 10,000 ft. elev. We believe this to be the first report of the species for Idaho. (\*\ Drapa orEIBATA Macbr. & Pays. This plant was previously represented with certainty only by the type-collection. The plants of that collection, Macbride & Payson No. 3135, were immature and (as now known) fragmentary. In 1944, the species was found in abundance in crevices of limestone about 10 miles n. of Leslie, Idaho, in Pass Creek Gorge, Butte Co., at about 6501 ft. elev., H. & M. No. 8840. The plants formed dense cushions several inches in diameter. In habit the plant differs materially from D. subalpina Goodman & Hitchcock, its nearest ally, since that species, occurring on sandy soil or shale, is not nearly so heavily matted. There are other significant differences in leaf and fruit characters, so there can be no doubt that D. subalpina and D. oreibata are distinct species. ®\ Drapa arcyr@a Rydb, and DraBa sPH#ROCARPA Macbr. & Pays. When Hitchcock treated these two species in 1941 (Univ. Wash. Pub. Biol. 11:80), there were but eleven collections avail- able in herbaria. Seven additional collections were made in 1944, and in all cases D. argyrea was found to be in rock crevices only, D. spherocarpa always on sand or gravelly talus. The two spe- cies were sometimes growing side by side, but more often one species only would be found in a given locality. Draba argyrea: H. & M. No. 9796, near Elk Lake, S. Fork Payette River near head, Boise Co.; H. & M. No. 10207, 6 miles s. of Atlanta, Elmore Co.; H. & M. No. 10506, 7 miles nw. of Alturas Lake, Blaine Co. Draba spherocarpa: H. & M. No. 9633, on peak 3.5 miles sw. of Stanley Lake, Custer Co.; H. & M. No. 9685, inlet to Stanley Lake; H. & M. No. 9798, near Elk Lake, Boise Co. ; con XK Va ae RS RA Yd 200 LEAFLETS OF WESTERN BOTANY _ [VOL. IV, No. 8 H.& M. No. 9883, bank of Baron Creek, 3 miles s. of Sacajawea Hot Springs, Boise Co. DRABA LANCEOLATA Royle. This white-flowered species was collected twice in the vicinity of Mt. Borah, H. & M. No. 10948 and No. 11157. First collected in Idaho by Macbride and Pay- son, it is very rare in the state where it is known only from the Lost River Mts. LESQUERELLA MONTANA (Gray) Wats. A collection from summit of grade between Dickey and Challis, Custer Co., Idaho, June 3, 1944, H. & M. No. 8922, is apparently closest to this species. The silicles are oblong, about 8 mm. long, very decidedly wing-margined (more so than in any other collection seen) and about 16-seeded ; the styles are scarcely 2 mm. long (shorter than recorded for the species). : ‘“\ LESQUERELLA CONDENSATA A. Nels. Several collections of this species have been made in Idaho: 2 miles e. of Dickey, Custer Co., on gravelly limestone mesa, June 3, 1944, H. & M. No. 8903: gravelly wash, Salmon River bank, about 20 miles s. of Challis, Custer Co., June 15, 1944, H. & M. No. 8995; roadside 2 miles e. of Gilmore, Lemhi Co., June 25, 1944, H. & M. No. 9299; dry rocky point 6 miles n. of Gilmore, Lemhi Co., May 17, 1941, R. J. Davis No. 3130. In most cases there is an intergradation Sates the species proper and the var. levis Pays. ~~ Prysarta GEYERI (Hook.) Gray var. puRPUREA Rollins. Two collections of this plant were made in Idaho, one from sandy banks above Silver Creek, about 8 miles ne. of Myer’s Cove, Lemhi Co., July 1, 1944, H. & M. No. 9467. These plants had — yellow petals that turned rose or purplish as they aged, most silicles had 2 (but occasionally 3) ovules in each locule. The other collection, from hard, packed soil in canyon 3 miles directly w. of Challis, Custer Co., July 3, 1944, H. & M. No. 9523, was similar, but the petals were yellow and did not redden as they aged. It is doubted, therefore, that the distinction in color, purple as compared with yellow, will prove constant. "* GROSSULARIA NEGLECTA Berger, N. Y. Agr. Exp. Stat. Tech. Bull. 109: 106 (1924). Two collections made in 1944, namely, H. & M. No. 8842, from Pass Creek Gorge in Lost River Mts. about 10 miles n. of Leslie, Butte Co., and H. & M. No. 9309, from near head of Spring Canyon, about 8 miles se. of Gilmore, NOVEMBER, 1945] NOTEWORTHY PLANTS OF IDAHO 201 Lemhi Mts., Lemhi Co., as well as Hitchcock et al. No. 3785, Lost River Mts., 7 miles n. of Dickey, Custer Co., are referable to this species. Berger’s original description might well be augmented some- what by our observations of these plants. They are low spread- ing shrubs mostly about 1 foot (to 3 ft.) in height. The cream- colored petals are about 1 mm. long, broadly ovate in outline, but abruptly narrowed to a tiny claw, % to % length of the yellow- greenish (or pinkish-tinged) sepals. The ovary as well as the hypanthium and sepals are often rather conspicuously hairy with very soft straight slender hairs. The bracts are glandular-hairy and the pedicels are hairy. It is one of the most common goose- berries of the Lost River and Lemhi mountains, forming matted growths on many of the slopes. To one who ordinarily calls all species of this relationship by the generic name Ribes, Grossu- laria neglecta seems almost a misnomer. NS Horoprscus pumosus (Nutt.) Heller. H. & M. No. 10398, on lava flow, 8 miles nw. of Pine, Elmore Co. The leaves of this plant are deeply cut to the base and contrast sharply with those of H. & M. No. 10885, from S. Fork Salmon River, Custer Co., the latter collection having smaller leaves that are toothed at apex only. It would key to H. microphyllus Rydb. var. typicus Ley (Bull. Torr. Bot. Club 70 : 283,—1943), whereas the former col- lection is almost surely H. dumosus var. typicus Ley (ibid., p. 281), although that author did not report the species for Idaho. me PoTENTILLA OVINA J. M. Macoun. Although we can find no reference to its occurrence in the state, this species with finely dissected leaves is the most common subalpine Potentilla on most of the peaks in central Idaho and is represented by the following collections, all made on limestone: H. & M. No. 8844, Pass Creek Gorge, 10 miles n. of Leslie, Butte Co.; H. & M. No. 10962, cirque at nw. base of Mt. Borah, Custer Co.; and H. & M. No. 9308, 8 miles se. of Gilmore, Lemhi Mts., Lemhi Co. \)\ PorenTILLA VIRIDESCENS Rydb. The following collection is thought to represent the first record of this species for Idaho: H. & M. No. 10316, 1 mile ne. of Little Trinity Lake, Elmore Co. ASTRAGALUS TENUIFOLIUS Nutt., T. & G. Fl. N. Amer. 1: 352 (1838). H. & M. No. 10854, dry open ridge north of peak im- mediately east of Castle Peak, Custer Co., elev. about 11,000 ft. 202 LEAFLETS OF WESTERN BOTANY _ [VOL. Iv, No. 8 The plants of this collection are scarcely 10 cm. tall and fit well the description of the species. Two other collections, however, H. & M. No. 10986, 1 mile w. of Mt. Borah, Custer Co., elev. 10,600 ft., and H. & M. No. 9280, 3 miles west of Leadore, in the Lemhi Mts., Lemhi Co., will, in general, fit A. tenuifolius also, the plants being less than 10 cm. tall, but the leaves vary so much in width that some of the plants would have to be referred to A. divergens Blank., Mont. Agric. Coll. Stud. 1:73 (1905). Two additional collections, H. & M. No. 8997, 20 miles s. of Challis, along Salmon River, Custer Co., elev. about 5500 ft., and H. & M. No. 9457, 3 miles s. of Cabin Creek Ranger Station, Lemhi Co., elev. 5500 ft., include plants mostly over 2 dm. tall (corollas over 10 mm. long). They would key to Homalobus strigosus (Coulter & Fischer) Rydb. in the key in the North American Flora (24:268). There surely can be little basis for the maintenance of these phases as distinctive species, and from our field observations we believe Jones was correct (Rev. Astrag. 74,—1923) in treating these variants as subspecific in nature and in noting that “all the varieties intergrade.” ASTRAGALUS RECLINATUS Cronquist. H. & M. No. 11310, along ditches near hay fields at May, Custer Co., Aug. 20, 1944. Although there are minor discrepancies between our plants and the specimens which Cronquist collected, the two collections are surely conspecific. May and Dickey (type-locality) have similar elevations and are located in parallel valleys separated by the Lost River Mts. The soil in both places is rather alkaline, being chiefly of limestone origin. Our plants differ as follows, the comparable characters of the type are shown in parentheses: stems 2 mm. thick (1 mm.) ; leaflets mostly 3, rarely 5, the terminal one as much as 45 mm. long and 9 mm. broad (mostly 20-—30 mm. long and 3—5 mm. broad), the lateral ones mostly 20—30 mm. long (mostly 10—20 mm. long) ; flowers dirty-white, not at all purplish, only the buds showing any purple color (whitish with a trace of purple). This plant is one of several collected by us in Idaho which apparently will not fit any niche already carved for it by Rydberg or M. E. Jones. We Oxytropis Parryt Gray. H.& M. No. 10880, rocky summit of peak immediately to west of Castle Peak, White Cloud Range, Custer Co., elev. 11,000 ft. So far as we know, this species, like NOVEMBER, 1945] NOTEWORTHY PLANTS OF IDAHO 203 the next following, has not previously been reported for Idaho. pM Oxytroris naNA Nutt. H. & M. No. 11024, head of Ma- hogany Creek, about 2 miles nw. of Mt. Borah, Custer Co., elev. 10,000 ft. CoLLOMIA DEBILIS ( Wats.) Greene. Several collections of this interesting plant were made, all having flowers with included anthers. One collection, H. & M. No. 10586, from fine limestone talus slides on west base of Mt. Ryan, Boulder Mts., Custer Co., has practically all leaves trifid, and unquestionably belongs under the var. trifida Payson. Three other collections, H. & M. No. 10488, from above Alpine Lakes, Blaine Co., H. & M. No. 10985, from 1 mile w. of Mt. Borah, and H. & M. No. 11267, west base of Mt. Mogg, Lemhi Mts., have at least half of the leaves entire and would key to var. integra in Payson’s treatment (Univ. Wyo. Publ. Sci. Bot. 1 : 79—-87,—1924), but there is so much variation in the lobing of the leaves that we question whether the two vari- ties can be maintained other than arbitrarily. Of these three col- lections, No. 10985 is grayish-pubescent, the others are much less hairy. A fifth collection is not identifiable by means of Payson’s key and appears to merit varietal distinction as follows: Q\ Collomia debilis (Wats.) Greene var. dentata C. L. Hitchc., var. nov. Planta 2—5 dm. lata, decumbens, viscida; foliis seepe oblanceolatis, acutis, 2.5—3.5 cm. longis, 6—9 mm. latis, petiolatis, laminis 4- vel 5-serrato- dentatis, bidentatis, vel integris; corollis czesiis, infundibuliformibus, pauce glandulosis, tuba lobi circa duplo longiore; staminibus corolla 1—3 mm. brevioribus; stylis exsertis 2—3 mm. Plants forming dense mats as much as 6 dm. in diameter from thick deep tap-roots, viscid throughout, leaves mostly oblanceolate, acute 2.5— 3.5 cm. long, 6—9 mm. broad, narrowed to slightly winged petioles %4 to % length of blades, the lower usually distinctly serrate-dentate at apex with 3 to 5 teeth, the upper ones entire or 2-toothed near tip, narrower than lower leaves ; corollas 20—26 mm. long, pale lavender, funnelform, slightly glandular without, the lobes about % length of tube, very slightly erose; stamens 1—3 mm. shorter than corolla, inserted 10—12 mm. from base of corolla; styles exserted 2—3 mm. Type: from granitic talus about 10 miles s. of Gibbonsville, Lemhi Co., Idaho, June 17, 1944, Hitchcock & Muhlick No. 9058 (Univ. of Wash. Herb.). This plant would key to var. [pome@a in Payson’s key, were the stamens exserted. However, it differs from that variety not only in that particular, but also in the peculiar dentate leaves 204 LEAFLETS OF WESTERN BOTANY [voL. IV, NO. 8 and the lavender corolla (as contrasted with reddish-pink). Pay- son’s plant is an alpine, whereas var. dentata was collected in the Arid Transition Zone at an elevation much lower than any other specimen of this species that I have seen. MERTENSIA CAMPANULATA A. Nels. About 16 miles n. of Mountain Home, Elmore Co., in a meadow beside the Mountain Home-Dixie Road, several acres of a robust Mertensia were seen. These plants, collected as H. & M. No. 8647, have proved to be Dr. Nelson’s species which is therefore considerably more widespread than reported by Williams (Ann. Mo. Bot. Gard. 24 : 68,—1937) : “Blaine and Camas Counties, Idaho.” Our field notes slightly modify and somewhat augment previous descrip- tions of the plant: “plants in large clumps 2 to 3 ft. wide, corollas blue, reddish-blue, or the limb blue and the tube red; flowers markedly drooping.” RN PEDICULARIS HALiit Rydb. H. & M. No. 9179, gravelly, dry meadow land about 6 miles n. of Wisdom, Beaverhead Co., Mon- tana, June 21, 1944. The corollas are more nearly magenta than “purple” and the entire plant is arachnoid-pilose (“stem gla- brous to inflorescence” according to Rydberg’s description), although the calyx is decidedly more hairy than the foliage. This is apparently the first record of the plant for Montana, although it was one of the most noticeable and abundant plants along the ‘Big Hole River. It was also collected in Idaho, where it has not been reported before: H. & M. No. 11191, meadow land over- lying limestone at 10,000 ft. elev. at head of W. Fork Pahsimeroi River, Lost River Mts., Custer Co., August 16, 1944. The plants of this collection are so nearly glabrous, even to the calyx, that they appear to be varietally distinct from the Montana collection. AGOSERIS PUMILA (Nutt.) Rydb. H. & M. No. 10989, from open limestone knolls at head of Rock Creek, west base of Mt. Borah, Custer Co., elev. 10,600 ft. Although the involucres of these plants are only 12—15 mm. long and no more than 1 cm. broad, the specimens appear to be a very much dwarfed form of this species. 6. ‘\ TOWNSENDIA SPATHULATA Nutt. H. & M. No. 10977, in talus at summit of saddle between Rock and Mahogany creeks, 1 mile w. of Mt. Borah, Custer Co., at 10,000 ft. elev. Not only did Larsen (Ann. Mo. Bot. Gard. 14:22,—1927) not list this NOVEMBER, 1945] NOTEWORTHY PLANTS OF IDAHO 205 species for Idaho, but her description to include these plants must be emended as follows: Biennials or short-lived perennials from slender tap-roots, rather uni- formly grayish-hirsute or pilose with 3—7-celled hairs as much as 2 mm. long; caudex crowned with rosette of 25—100 spatulate, rounded leaves 2—3 cm. long and 2—4 mm. broad; cauline leaves linear-oblanceolate, rounded to acute, somewhat narrower than basal leaves, usually equaling lateral heads and surpassing the central one; plants with single, sessile, central head and 2—8 lateral heads on spreading peduncles 1—3 cm. long, the central head 2—3 cm. broad, exclusive of rays, its involucral bracts 12—15 mm. long, about 1 mm. broad, long-attenuate, the outer ones greenish, scarcely at all scarious-margined, the inner ones with narrow scarious and more or less pinkish margins, somewhat ciliate-lacerate, all with tips more or less curved and twisted, ligules 12—18 mm. long, 2—2.5 mm. broad, rounded and entire or cleft into 2—4 teeth about 1 mm. long, bluish-violet, uniformly short stipitate-glandular on back, the lateral heads similar but only 1—1.5 cm. broad; achenes about 3.5 mm. long, the faces grayish- _strigillose with short thick simple hairs about 0.4 mm. long, these neither glochidiate nor bidentate ; pappus of about 20 scabrid-barbellate awns nearly twice as long as achenes. Since the hairs of the achenes are simple, these plants will key nowhere in Larsen’s key. Because of this, material was sent to the New York Botanical Garden where Dr. Cronquist kindly confirmed the identification. 7 7 7 The following species were identified by others and reported as unusual : R\ Atrium stmittImum Henderson. H. & M. No. 8621, 6 miles s. of Lowman, Boise Co.; H. & M. No. 8738, summit of Dog Mt., 4 miles nw. of Pine, Elmore Co.; H. & M. No. 9640, high ridge sw. of Stanley Lake, Custer Co.; H. & M. No. 10156, Middle Spangle Lake, 17 miles n.-ne. of Atlanta, Elmore Co. ; H. & M. No. 10595, south base of Ryan Peak, Boulder Mts., Custer Co. These collections, determined by Marion Ownbey, represent great extension of range. The following two species of willows were identified by Dr. C. R. Ball. The quoted comments are his: SALIX BRACHYCARPA var. SANSONI Ball. H. & M. No. 11133, boggy meadow land at junction of East and West forks of Pahsimeroi River, north side of Lost River Mts., elev. about 8000 ft. “Rare in Idaho.” 206 LEAFLETS OF WESTERN BOTANY _ [VOL. IV, No. 8 ® Satrx Farr# Ball. H. & M. No. 11172, in wet meadow ‘land at head of West Fork of Pahsimeroi River, north side of Lost River Mts., elev. about 10,000 ft. Matted shrubs 0.5—2 ft. high, well browsed by deer. “Probably the first record for Idaho, although I have not checked all records. Occurs commonly in the mountains of southern Alberta and British Columbia and has been found in Wallowa Co., Washington, and in Glacier and Yellowstone National parks.” XPS: RANUNCULUS ORESTERUS L. Benson. H. & M. No. 8507, 2 miles w. of McCall, Valley Co. Dr. Benson, who identified the plant, reports that this is but the third collection of the species. VACCINIUM OCCIDENTALE Gray. H. & M. No. 9832, shores of Elk Lake, near source of S. Fork Payette River, Boise Co. According to J. H. Schultz, who made the determination, this is the first report of the species for the state. STUDIES IN CAREX—I BY JOHN THOMAS HOWELL 1. Species NEw To IDAHO In the course of general field work in the mountains of Idaho during the summer of 1944, C. L. Hitchcock and C. V. Muhlick obtained a large and varied collection of Carex which was re- ferred to me for determination. More than 130 numbers were collected representing 50 species, a truly remarkable collection since it represents well over a third of the Carex flora of the state. In the recent enumeration of Idaho carices by R. J. Davis (Contributions towards a Flora of Idaho, Leaflet No. 10, pp. 1—35), 121 species are listed for the state. Among the Hitch- cock and Muhlick collections are several which are not on Davis’ list and which represent new records for Idaho. For the deter- mination of several of these and for his valued opinion on critical species in the Atrate and Acute, I am very grateful to Dr. F. J. Hermann. C. VERNACULA Bailey (?). Determined by F. J. Hermann. Margin of Upper Norton Lake, elev. about 9000 ft., Smoky Mts., Blaine Co., No. 10713. The vegetative parts of this puzzling plant are reminiscent of species of the Vulpine but, as Dr. Her- NOVEMBER, 1945 | STUDIES IN CAREX 207 mann points out, the perigynia are quite like those of C. vernacula. Further collections will be needed to determine the exact identity of the plant. & YS _C. Eastwooprana Stacey. Dry ground at head of East Fork of Pahsimeroi River, elev. about 10,300 ft., Lost River Mts., Custer Co., No. 11076. This species was reported from Idaho at the time it was published (Leafl. West. Bot. 2: 122,—1938), but it was omitted by Dr. Davis from his list. C. scIRPOIDEA Michx. Cirque at northwest base of Mt. Borah and at head of Rock Creek, elev. about 11,500 ft., Lost River Mts., Custer Co., No. 10957. C. GaRBERI Fern. var. BIFARIA Fern. Determined by F. J. Hermann. Willow Creek about 4 miles from Boise-Lowman highway cut-off, Boise Co., No. 10044. C. rLava L. Boggy meadow bordering Elk Lake at head of South Fork of Payette River, Boise Co., No. 9792, 9793. In the Idaho plants the perigynia are only 4 mm. long although they are generally considerably longer (cf. 4.5—6 mm. as given by Mackenzie, N. Amer. Fl. 18: 307). C. PELOcARPA Hermann. Determined by F. J. Hermann. Moist places on talus, west base of Ryan Peak, elev. about 11,000 ft., Boulder Mts., Custer Co., No. 10581; cirque at northeast base of Mt. Borah and at head of Rock Creek, elev. about 11,500 ft., Lost River Mts., Custer Co., No. 10941, 10955; near seep- age on steep slide, head of East Fork of Pahsimeroi River, elev. about 10,300 ft., Lost River Mts., Custer Co., No. 11090. Because of their superficial resemblance to C. nova Bailey, I had referred these collections to that species before sending them to Dr. Hermann. His notes on the differentiation of C. nova and C. pelocarpa are as follows: “C. nova can generally be recog- nized by the stiffly erect culms which are roughened above and by the mainly pale stramineous perigynia which are muricate to almost ciliate-scabrous on the upper margin”; while in C. pelo- carpa “the flexuous culms are not roughened below the heads, the perigynia are not stramineous and are smooth on the upper margins, and the achenes are relatively long-stipitate.”” With these notes it is now clear that the Idaho collections referred to C. nova by Stacey (cf. Leafl. West. Bot. 2:15, 91) are really C. pelocarpa: Devil’s Bedstead, Sawtooth Mts., 8000 ft., Blaine 208 LEAFLETS OF WESTERN BOTANY __ [VOL. Iv, No. 8 Co., Thompson No. 13555; Boulder Creek Canyon, Sawtooth Mts., 9000 ft., Blaine Co., Thompson No. 14058. The Oregon collection referred by Stacey (ibid., p. 15) to C. nova (Peck No. 18493 from the Wallowa Mts.) needs further study; in it the culms are roughened and the perigynia are stramineous, but the upper margin of the perigynia is only granular-roughened instead of muriculate. C. CuimaPHILA Holm. Small meadow just below upper Frog Lake, below north base of Castle Peak, White Cloud Range, Custer Co., No. 10832. 2. Species NEw To NEVADA In 1944, H. D. Ripley and R. C. Barneby collected two carices in White Pine County in eastern Nevada that were not listed for the state by F. J. Hermann in his 1940 enumeration (The genus Carex, Contrib. toward a Fl. of Nev. No. 17). They grew in alkaline clay along warm water streams at-Monte Neva Hot Springs and were part of a remarkable localized flora that has been recently described by Ripley (Quart. Bull. Alp. Gard. — Soc. 13:34, 35,—1945). CaREX SCIRPIFORMIS Mkze. Ripley & Barneby No. 6287. This collection contains only staminate material, but from the relatively broad white-hyaline margin of the scales it cannot be referred to C. scirpoidea Michx., C. stenochlena (Holm) Mkze. or C. curatorum Stacey, species of Scirpine with culms aphyl- lopodic. The occurrence of C. scirpiformis in Nevada is of spe- cial interest not only because it represents a new station for a rare species, but also because it represents a notable extension of ‘range southward from southwestern Montana. CarEx GarBERI Fern. Ripley & Barneby No. 6286. The discovery of the typical form of this species in Nevada is even more remarkable than the discovery of its associate, C. scirpi- formis. In western America, C. Garberi has been represented only by var. bifaria Fern. which has been reported there from British Columbia, Alberta, Washington, and Idaho. The species, heretofore, has been known from regions adjacent to the Great Lakes, the westernmost station being on Isle Royale, Michigan (cf. Fernald, Rhodora 37: 254,—1935). I am grateful to Dr. Hermann for verifying my determination. NOVEMBER, 1945] TRIFOLIUM AND OPUNTIA 209 NOMENCLATORIAL RECOMBINATIONS IN TRIFOLIUM AND OPUNTIA BY LYMAN BENSON Pomona College, Claremont Alice Vaughn, Amer. Midl. Nat. 22:575-579 (1939), has shown that Trifolium involucratum Ort., Hort. Reg. Bot. Matrit. 33 (1797), is a Mexican species and that it does not occur in the western United States. This plant is T. Ortege@ Greene, Pittonia 3:186 (1897), based upon T. involucratum Ort., and the work of Vaughn confirms the earlier study of Greene. Joseph Ewan, Leafl. West. Bot. 3: 222-224 (1943), has pre- sented evidence that T. Wormskjoldu Lehm., Ind. Sem. Hort. Ham. 17 (1825), is a name of uncertain application. Trifolium Willdenoviu Spreng., Syst. 3:208 (1826), is the earliest specific name available for the species occurring on the Pacific Coast and in the southern Rocky Mountain region. Therefore the follow- ing recombination is proposed for a variety occurring in the mountains of Arizona and New Mexico: Trifolium Willdenovii Spreng. var. longicaule (Woot. & Standl.) L. Benson, comb. nov. 7. longicaule Woot. & Standl. Contr. U. S. Nat. Herb. 16: 141 (1913). T. Wormskjoldti Lehm. var. longicaule L. Benson, Amer. Jour. Bot. 28: 363 (1941). Leaflets mostly cuneate-obovate ; stipules entire or nearly so; peduncles pilose; involucre reduced and cleft nearly to the base, much shorter than the calyces. Meadows in the yellow pine belt of the Rocky Mountain forests (cf. L. Benson, Amer. Jour. Bot. 29 : 491—500,—1942) at 6500 to 8000 feet elevation or along streams at lower levels ; mountains of eastern Arizona and eastward to the White and Sacramento mountains, New Mexico; particularly well known from the Santa Catalina Mountains, Pima County, Arizona (cf. various collections including L. Benson No. 9009, Pomona, Univ. Ariz.). The corresponding characters of Trifolium Willdenovii var. typicum L. Benson, nom. nov. (based upon T. Willdenovii Spreng. Syst. 3: 208,—1826; in turn based upon T. involucratum Willd. in L. Sp. PL, ed. 4, 3: 1372,—1800, not Lam. in 1778, not Ort. in 1797) are as follows: leaflets approximately elliptic- obovate, but usually acute at both ends, rarely more than a few of them cuneate-obovate ; stipules laciniate; peduncles glabrous or rarely hairy; involucre cleft and parted, usually two-thirds to four-fifths as long as the calyces. Var. typicum occurs in the mountains of Arizona and New Mexico as well as on the Pacific Coast. Opuntia erinacea Engelm. & Bigel. var. xanthostema (K. 210 LEAFLETS OF WESTERN BOTANY _ [VOL. Iv, NO. 8 Schum.) L. Benson, comb. nov. O. rhodantha K. Schum., Mo- natschr. Kakteenk. 6: 111 (1896), nom. subnud. ; Gesamtbeschr. Kakt. 735 (1898), cf. Croizat, Cact. & Succ. Jour. 16:88, 89 (1944). O.«anthostema K. Schum., loc. cit. (spelling apparently intentional). O. rhodantha var. xanthostema Rehder, Jour. Arn. Arb. 7: 149 (1926). O. erinacea var. rhodantha L. Benson, Proc. Calif. Acad. Sci., ser. 4, 25:249 (1944). The single validly pub- lished varietal epithet among the many invalidly published ones (those in trade lists and popular articles) was overlooked in the previous publication. A NEW SPECIES OF ASCLEPIAS FROM NEVADA BY R. C. BARNEBY Asclepias Eastwoodiana Barneby, spec. nov. foliis tomentoso-margi- natis, corolle forma, totoque habitu A. involucrate Engelm. manifeste affinis, sed foliis multo latioribus, umbellis plerumque nudis exinvolucratis, corollz limbo saturate purpureo, coronaque antheras haud attingenti (nec eas superanti), cucullis abruptius truncatis necnon corniculo introflexo in- cluso diversa. Herba perennis humilis paucicaulis e caudice lignea tortuosa subterranea oriens, preter foliorum margines albo-ciliatas sparse puberula vel glabres- cens; caulibus plerumque simplicibus flexuosis adscendentibus 1—2 dm. altis, angulatis striatisque, inferne nudis, minute puberulis; foliis oppositis vel haud raro internodio brevissimo separatis subalternis, integerrimis, in- ferioribus suborbicularibus vel latissime ovatis, apice retusis truncatisve mucronatis, lamina 1—3 cm. longa secus petiolum circa dimidio breviorem decurrenti, in folia media ovato-acuminata 2.5—4.5 cm. longa brevius petio- lata ac summa anguste lanceolato-acuminata breviora subsessilia sensim abeuntibus, omnibus glaucescentibus pallidis, utraque facie glaberrimis vel secus nervum medianum subtus prominulum minute puberulis, margine hinc inde undulata villis crispulis incurvis albociliata ; umbellis 6—11-floris, unica terminali et 1—3 in foliorum summorum axillis sessilibus, vel pedunculo 0.5—2 cm. longo gracili gestis; pedicellis filiformibus rectis 1.5—2 cm. longis purpureo-tinctis puberulis, bractea minuta subulata decidua suffultis ; calycis laciniis anguste ovatis acutis 2—3 mm. longis, hirsutulis ciliatisque ; corolla purpureo-violascenti, in lacinias 5 ovato-acuminatas scarioso-margi- natas 4—5.5 mm. longas profunde fissa; columna 0.3—0.6 mm. longa; corone cucullis subquadratis, 1.5—2 mm. altis, a tergo obtusissime angu- lato-convexis, marginibus superioribus concavo-truncatis in auriculam del- toideo-acuminatam ad antheras accumbentem productis, primum albis demum purpurascentibus, corniculo subulato incurvo albo incluso; antheris 2—2.5 mm. longis coronam breviter sed manifeste superantibus, late scarioso-marginatis appendiculatisque; folliculis ignotis. Nevapa: low alkaline clay hills in the valley of the Reese River, 6 miles northwest of Austin, Lander Co., alt. 5800 ft., 11 NOVEMBER, 1945] NEW SPECIES OF ASCLEPIAS 211 June, fl. Ripley & Barneby No. 6196. Type in Herb. Calif. Acad. Sci. No. 320832. Toquima Range, between Belmont and Man- hattan, Nye Co., alt. 6900 ft., No. 3690. In desert, northeast of Goldfield on the Esmeralda-Nye county line, Rives & McKenzie No. 22 (Herb. Univ. Nevada, Reno). In general habit and in structure, as well as in the white- margined leaves, Asclepias Eastwoodiana is closely related to A. involucrata Engelm., differing at first sight in the purple color- ing of the corolla and in the broader leaves, the lowest being sub- orbicular or very broadly ovate, obtuse or mucronate. Further- more the hoods are smaller and more abruptly truncate, surpassed by (rather than, as in A. involucrata, surpassing) the anthers, while the horn which, in the latter, is curved centripetally towards the axis of the flower and exserted, is in ours curved backwards and included in the fold of the hood. In A. involucrata the umbel is commonly terminal and solitary, prominently involucrate by a whorl of 2—5 linear or narrowly lanceolate leaves ; a terminal sessile umbel subtended by a pair of leaves occurs occasionally in A. Eastwoodiana also, but the stems normally bear one or more lateral umbels in the upper axils, either sessile or pedunculate, and the inflorescence, in any case, is conspicuously exserted be- yond the upper leaves. It should be borne in mind that A. involu- crata has been reported to range no farther north or west than southern Utah, so that A. Eastwoodiana, known at present from a narrow strip of territory in central Nevada, appears to be geo- graphically isolated from its commoner relative. In the broad leaves and purple flowers A. Eastwoodiana somewhat resembles A. Ruthie Maguire of the Navajo Basin, but the form of the corolla and the pubescence are quite different. Asclepias involucrata and A. Ruthie, as I have observed them in Arizona, Utah, and New Mexico, occur normally, perhaps . exclusively, on dunes or in open valleys where the soil is deep and sandy. In contrast to this the Nevada species prefers barren clay hills and a stiff soil derived from the comminuted detritus of limestone mountains, and appears to require, or at least to toler- ate, some quantity of soluble alkali. In this respect it resembles A. cryptoceras Wats., with which, in the type locality, it was associated. It is a pleasure to name this species in honor of Miss Alice Eastwood of the California Academy of Sciences, to whom the author is indebted for much help and encouragement. 212 LEAFLETS OF WESTERN BOTANY [ VOL. IV, NO. 8 ADDITIONS TO THE FLORA OF UTAH BY FRANK W. GOULD University of Arizona, Tucson In 1941 and 1942 the writer had the privilege of botanizing extensively in southwestern Utah. Collections were made mostly in the St. George region and the nearby Pine Valley Mts. which rise to over 10,000 feet from the valley of the Virgin River. St. | George is well known to botanists through the collecting activi- ties of Marcus E. Jones in the latter part of the nineteenth cen- tury. One of Jones’s favorite localities, the flat-topped Black Hill on the western outskirts of the town, still possesses an interest- ingly diverse flora. Here steep, boulder-strewn slopes have dis- couraged extensive grazing, and the native species have been able to hold out against the encroachment of such weedy introductions as Bromus rubens L. and Erodium cicutarium L’Her. The native flora of the historic Red Hill, a sandstone outcrop that bounds St. George on the north, has not fared as well, for this area is not only heavily pastured but is also used extensively for picnic grounds. One collection of special interest from the St. George region is that of Erodium texanum Gray. A small stand of this species was located on Red Hill (Gould No. 1488), and a single de- pauperate plant was observed on Black Hill. There has been apparently no previously published record of E. texanum in Utah, -but Dr. Bassett Maguire, in personal communication, reported another collection made near the Colorado River in Garfield County (Parry in 1940). Kearney and Peebles (FI. Pl. and Ferns Ariz., p. 507) give the range of the species as “Texas to southeastern California.” Two species of Composite collected near Pine Valley, on the northern slope of the Pine Valley Mts., represent new state records according to Dr. S. F. Blake. Cirsium arizonicum (Gray) Petrak was found scattered but not infrequent in oak openings at 7,000 ft. (Gould No. 1883). Blake in Kearney and Peebles (op. cit., p. 1014) gives the range of this species as “Utah ( ?) and Arizona.” It is common throughout northern and south- western Arizona. er saligna L. was collected on roadsides in the Pine Valley Forest Recreational Camp (Gould No. 2011). Concerning this weed, Dr. Blake wrote, “I find that this is a European species which has been turning up in many places in NOVEMBER, 1945] WEEDY SPECIES OF LATHYRUS 213 the last fifteen years or so. There are specimens in the U. S. National Herbarium from Virginia, West Virginia, Kentucky, Ohio, Indiana, Michigan, Missouri, and California, and it no doubt occurs in other states.” SOME WEEDY SPECIES OF LATHYRUS BY JOHN THOMAS HOWELL Two Orecon Vetcuiincs. West of the Cascade Range in Oregon two annual species of Lathyrus from the Mediterranean Region of the Old World have been found growing without culti- vation, L. angulatus L. and L. sphericus Retz. Although they are quite alike in appearance with their unijugate leaves and with their uniflorous peduncles, there are a number of characters by which they may be distinguished. In L. sphericus the stipules are linear-semisagittate, all leaf-tendrils are simple, the peduncles are only about as long as the short petioles, the legumes are prominently longitudinally veined, and the seeds are smooth. In L. angulatus the stipules are lanceolate-semisagittate, the tendrils are branched on the upper leaves, the peduncles are much longer than the petioles, the legumes are finely net-veined, and the seeds are rough. Lathyrus angulatus as it occurs in Oregon has been confused with L. pusillus Ell., a species indigenous to the southeastern United States. Lathyrus pusillus can be readily distinguished from L. angulatus by the position of the flower: in the former it is at the top of the peduncle, in the latter it appears lateral since the peduncle is produced beyond it into an elongate slender tip. ~ The following Oregon specimens of L. angulatus have been loaned to me by Dr. Helen M. Gilkey from the Oregon State College Herbarium: Harrisburg, Linn Co., Gillett in 1941; Jefferson, Marion Co., Hyslop in 1943; Dallas, Polk Co., Beck in 1932. The only Oregon specimen of L. sphericus that I have seen is one collected by Miss Eastwood and me in waste ground along the Umpqua River at Winchester, Douglas Co., No. 1476. TANGIER Pea IN CALIFORNIA. Rocky bluffs at the south end of Tiburon Peninsula, Marin County, are covered in the spring with masses of Lathyrus tingitanus L., which, with its bright reddish-purple flowers, makes a brilliant display. For a number 7 214 LEAFLETS OF WESTERN BOTANY _ [VOL. Iv, No. 8 of years this rampant annual was passed as L. latifolius L. which is perennial and very different in many characters. In Marin County, the Tangier Pea may be reported not only from Tiburon (Howell in 1939, Howell No. 21260) but also from San Rafael (Howell No. 19639) and Inverness Park (Howell in 1944) where it has become established in waste places. Outside of California, L. tingitanus has been reported from the Willamette Valley, Oregon, the only record of the plant that I have seen for the Pacific states (cf. Gilkey Handb. of NW. Flow. Pl., 183; Abrams Ill. Fl. 2: 619). NEW VARIETIES OF CALYPTRIDIUM PARRYI BY JOHN THOMAS HOWELL } A further detailed study* of Calyptridium Parryi Gray has disclosed that forms of that species which have been found at a few isolated stations in the mountains of southern Arizona and in southwestern Nevada should be accorded taxonomic recog- nition. The species in typical form is found in California from the higher mountains of southern California northward to Mt. Pinos and the southern Sierra Nevada, with an outlying station in the Mt. Hamilton Range in the central western part of the state. The variants are distinguished by characters that I believe to be rather fundamental and perhaps they should be accorded specific recognition; but since the plants bear a close habital resemblance to each other that is probably an indication of a common ancestry and close relationship, I have preferred to treat them as divergent parts of a single species. The varieties may be distinguished by the following key: 1. Flowers in age persistent or not readily deciduous. 2. Pedicels apparently jointed with the rhachis; seeds more or less PODereulate ee AEs ae ese aes le ee 1. var. typicum 2. Pedicels not jointed with the rhachis; seeds smooth and shining PATE NERY ce, WE ANE aS AAO eRe EH Ente 2. var. arizonicum 1. Flowers in age readily deciduous. Seeds shining and generally smooth except for narrow tuberculate margin.................-..- 3. var. nevadense 1. C. Parryi Gray var. typicum J. T. Howell, nom. nov. C. Parryt Gray, Proc. Amer. Acad. 22: 285 (1887). Representative collections, all from California. San Jacinto Peak, San * See Leafl. West. Bot. 3: 262-266 (1943). The present study is based on specimens in the California Academy of Sciences (CAS), Dudley Herbarium (DS), Gray Herbarium (G), Sacaton Field Station (Sac), University of California (UC), and United States National Herbarium (US). The letters in parentheses are used in the customary manner in citing specimens. NOVEMBER, 1945 | CALYPTRIDIUM PARRYI 215 Jacinto Mts., W.W. Swarth in 1914 (CAS) ; Mohave slope of Bear Valley, San Bernardino Mts., Parry & Lemmon in 1876 (type, G; photograph, DS); Bear Valley, Parish No. 3725 (CAS, DS, G, UC) ; near summit of Mt. San Antonio, San Gabriel Mts., Abrams No. 1929 (DS) ; Griffins, Mt. Pinos region, Elmer No. 3948 (CAS, G, DS) ; end of Mt. Whitney Road, southern Sierra Nevada, Kerr in 1937 (CAS) ; Isabel Peak, Mt. Hamilton Range, H. K. Sharsmith No. 3381 (DS). The last-cited collection differs in having somewhat smaller seeds (0.5—0.6 mm. in diameter instead of 0.66 mm.) that are less conspicuously tuberculate and more shining. The fact that the plants are small perhaps indicates only an edaphic response, but the central Californian plant needs further study with ad- ditional collections. Because C. pygmeum Parish was referred to C. Parryi by Mrs. Ferris (in Abrams FI. Pac. St. 2: 130) the question of the specificity of that plant was again raised and I have restudied the type-collection (DS, type; G). This study has not only con- firmed my belief that the species is distinct from C. Parryi, but it has disclosed that the “notable” differences in the seeds of the plants from the San Bernardino Mts. and from the Sierra Nevada that were described in my earlier notes do not exist (cf. Leafl. West. Bot. 3:263). The pocket on the type-sheet of C. pyg- me@um contains capsules and seeds of C. monandrum Nutt. and C. Parryi as well as of C. pygmeum, and, since the seeds of C. Parryi are most numerous, they were erroneously believed to belong to C. pygmeum, not only by me, but also by Rydberg, who published Parish’s name (N. Amer. Fl. 21:320). Hence the smaller (0.4 mm.) smooth, shining seed characteristic of C. pyg- meum adds yet another structure to those already indicated for distinguishing this “good” species. 2. C. Parryi Gray var. arizonicum J. T. Howell, var. nov. A var. typico per semen omnino leve et nitens et a var. typico et var. nevadensi per pedi- cellos non articulatos abludens. Type: U. S. National Herbarium No. 497458, collected in hills above Rosemont, Pima Co., Arizona, between Mar. 13 and Apr. 23, 1903, by David Griffiths, No. 4125. Other Arizona collections are: Canoa to Arivaca, Pima Co., Griffiths No. 3556 (US); near Nogales, Santa Cruz Co., Peebles & Fulton No. 11454 (Sac). The smooth shining seed of the Arizona variant of C. Parryi was noted by Kearney and Peebles in their Fl. Pl. & Ferns Ariz., p. 300. In typical C. Parryi the jointed character of the pedicel is usually evident, but in var. arizonicum no articulation can be 216 LEAFLETS OF WESTERN BOTANY _ [VOL. Iv, NO. 8 seen. In this character var. arizonicum resembles C. pygmeum. In var. typicum the stems are prostrate but, at least at times, in var. arizonicum the stems are suberect. 3. _C. Parryi Gray var. nevadense J. T. Howell, var. nov. A var. typico differt : floribus in senectute a rhachide deciduis ; seminibus sublevibus niti- disque utrimque preter marginem leviter tuberculatum. Type: Dudley Herbarium No. 190847, collected on the north slope of Gold Mt., elev. 7000 ft., Esmeralda Co., Nevada, by D. D. Keck, No. 559, June 23, 1930. Duplicate in Herb. Univ. Calif. This form of C. Parryi is known only from the attractive plants of the type-collection which were distributed under the name C.roseum. Although the relationship of the plant as shown by the character of the capsule is with C. Parryi, the readily de- ciduous fruits and fruiting sepals are distinctly reminiscent of C. roseum, along the southern distributional limit of which var. nevadense has been found. A NEW NAME IN ASTRAGALUS BY THOMAS H. KEARNEY There is an interesting species, not uncommon in sandy areas from South Dakota and western Nebraska to Idaho, New Mexico, and northeastern Arizona, which apparently has no valid name in Astragalus. The plant in question is characterized by slender stems arising at intervals from long slender creeping rootstocks, the terminal leaflet represented by a prolongation of the rachis or the entire leaf reduced to the rachis, and the pods greatly in- flated, papery-walled, and conspicuously mottled. It is obviously related to A. ceramicus Sheld. (A. pictus Gray, not of Steud.), but differs in the erect leaves with fewer (or no) lateral leaflets, — fewer-flowered racemes, and broader pods, as well as in the great prolongation of the leaf-rachis. To remedy the deficiency, a new name is proposed, as follows: Astragalus mitophyllus Kearney, nom. nov. Psoralea longifolia Pursh, Fl. Amer. Sept. 741 (1814). Phaca longifolia Nutt. ex Torr. & Gray, Fl. North Amer. 1: 346 (1838). Astragalus filifolius Gray, Pacif. Railroad Rpt. 12: 42 (1860). Not A. filifolius Clos (1846). Astragalus ceramicus Sheld. var. imperfectus Sheld., Minn. Bot. Studies 1: 19 (1894). Astragalus longifolius Rydb., Fl. Nebr. 21:47 (1895). Not A. longi- folius Lam. (1783). LEAFLETS of WESTERN BOTANY Available A limited number of copies of volumes I, II, and III at $5.00 each. Single copies of most numbers in these volumes are available at $.40 each. LINDLEY’s GLOSSOLOGY An excerpt from this most useful work, reissued with an introduction by Alice Eastwood. $.50. ALIcE EastTwoop’s Key to the Common Families of Flowering Plants in California and Guide for the Analysis and Description of Flowering Plants. $.25. Address: Joun THomas HowE.Li California Academy of Sciences, Golden Gate Park San Francisco, California Vo. IV No. 9 LEAFLETS of WESTERN BOTANY CONTENTS PAGE A Tentative Key to the Small-flowered Lupines of tue Vv eatern United otatess 6) 2) sie a) eileen tha ALICE EAstwoop pemunuarte ae Fistragalye. 006)! see oe Peal bj ates eh eh bees Tuomas H. Kearney Bees Abtracaloruin’ Woes. sie ae bg ie Vk ae eee R. C. BARNEBY Interesting Western Plantt—VII ....... .. 239 Puiuie A. Munz SAN FRANCISCO, CALIFORNIA JANUARY 28, 1946 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. PUUPETP TTA PEPE) ETP TETP PERE UETPTyeTey INCHES eee ee ee UM DAU AU bd his aie ith | Owned and published by Aticrt Eastwoop and JoHN THomas Howktr JANUARY, 1946] SMALL-FLOWERED LUPINES 217 A TENTATIVE KEY TO THE SMALL-FLOWERED LUPINES OF THE WESTERN UNITED STATES BY ALICE EASTWOOD This key, as well as other keys to species of Lupinus that I have tried to make and have published in LEAFLETS oF WESTERN Botany, have been chiefly for my own guidance in attempting to resolve the intricacies of the numerous and perplexing entities in this genus. In each case some salient morphological character or environmental situation has been chosen for delimiting a group. So many lupines unlike in appearance have been aggre- gated as synonyms or varieties under some species collected by early explorers and therefore the first named, a species, perhaps, which is purely local or which is known from a part of the country still but little explored. As my guide, I have taken the original descriptions supplemented when possible by photographs of types taken at the Royal Herbarium at Kew, at the Herbarium of the British Museum at South Kensington where Nuttall’s specimens are, and at the Lindley Herbarium at the University of Cam- bridge, England, where those named by Agardh are, chiefly the collections of Douglas. I also took notes on these types, adding characteristics not mentioned in the descriptions. I call these keys tentative, since they are practically only tentative, and I hope that they may help others as they have already helped me. The present key includes the very small-flowered species with flowers less than a centimeter long and also generally as wide since many are nearly circular in outline. It excludes the small- flowered cespitose lupines related to L. lepidus for which I hope some time to prepare a key that will include those from western North America. The key to the species from California allied to L. Breweri has already been published (Leafl. West. Bot. 2 : 249— 253). The species treated here are divided into three groups and the keys to each, with notes and descriptions, will be published in two papers. The three groups may be distinguished as follows: 1. Leaves bright green, upper surface of leaves glabrous or sometimes with a few scattered fine appressed hairs Leaflets of Western Botany, Vol. IV, pp. 217-240, January 28, 1946. 218 LEAFLETS OF WESTERN BOTANY __ [VOL. IV, NO. 9 2. Leaves greenish, with indument evident on both surfaces ace sap tag BO Ee EES SE ois Sa cee ae Group 2 3. Leaves white, indument dense, silvery and silky........ Group 3 Key To Group 1 (Species with leaves bright green, glabrous ‘or subglabrous) 1.) Stems apparently simple.i52240 2 et ee eee 2 Stems branching: 2 2522 oss ca ee ee et oe ee 4 2. Stems low, decumbegt; leaves on long petioles, leaflets oblanceolate to narrowly spatulate, acute; flowers dark blue ; pods large............ BUTE Se cis, OSE nd Tet eae Oe Hie me Ne see SPE E, 1. L. onustus group 2: >Stemis- erect’ and enerally~ tall. .2.....c22. 2k ee 3 3. Leaflets narrowly linear-oblanceolate, petioles 3 cm. or more long; calyx saccate at base; corolla blue, 7—8 mm. long........ 2. L. tenellus L. foliosus var. stenophyllus 3. Leaflets broadly oblanceolate or spatulate; stems leafy, leaves erect on short petioles ; calyx gibbous at base; flowers blue, about 8 mm. —_ Notas a i Pee eee ep eRe ee ee pee, eee 3. L. spathulatus 4: Leafiets broadly. .oblanceolate.22.2 20 ee eee 5 4. Leaflets narrowly oblanceolate or linear-oblanceolate..................-...---.-- 6 5. Flowers blue, 5—6 mm. long, somewhat scattered, keel ciliate on upper jb eae Raa A PN 2 HANS) Heh Mie rae ee em eee 4. L. parviflorus 5. Flowers blue, rarely white, more or less verticillate, 8—9 mm. long, keel densely ciliate on lower pafrt..............---...----0-++- 5. L. viridifolius 6. Branches erect from basal decumbent stems; flowers 5—6 mm. long, white or sordid, in densely flowered narrow racemes....6. L. ingratus 6. Stems branching above; bracts surpassing flowers in bud; flowers violet, about 6—7 mm. long, keel delicately ciliolate......................- Be AGaE ch VAATED FYE cde Me tlen ee ee las eee. Seats an: 7. L. myrianthus 1. Lupinus onustus Watson and allies. These are low plants with decumbent stems, leaflets broadly oblanceolate, leaves chiefly basal and on long petioles, and flowers small, roundish in outline, violet or purple. See Leafl. West. Bot. 4:41—43 for a discussion of the group by the author. 2. Lupinus TENELLUS Dougl. ex G. Don Gen. Syst. 2: 367 (1832). Type-locality as given on Douglas’ specimens in Herb. Mus. Brit. : “Grand Rapids and on the Multonomah River 1825” ; type- locality on Douglas’ specimen in Herb. Lindley.: “between Fort © Explanation of Plate. Type-specimen of Lupinus parviflorus Nutt. in Herb. Nuttall. in Herb. Mus. Brit. The specific name is preceded by an asterisk, the peculiar mark with which Nuttall indicated a new species. “Columbia plains’’ is the locality given. § ‘ hie 4° ty y , ity Pat) ’ yee Wey 4, hy tots ; ‘ 4 RS f th’ | ee - A ARM Ay: a + north (aA itp om «| hoi, Wa he ie JANUARY, 1946] SMALL-FLOWERED LUPINES 219 Vancouver and the Rocky Mts.” A photograph of Douglas’ specimen in Herb. Lindley. at Cambridge, England, is in the Calif. Acad. Sci. and from that specimen the following description was made: Stems very slender, reddish, sparingly pubescent, branchlets ascending ; racemes short-peduncled, flowers subverticillate with whorls about 2 cm. apart, pedicels 5 mm. long, bracts shorter than buds, soon falling ; leaflets 5—8 with a few short hairs on upper surface, more densely pubescent on lower, folded and falcate, about 2 mm. wide when folded, 3—4 cm. long, linear-lanceolate, tapering at both ends; calyx saccate at base, but not spurred, distinctly bracteolate, lower lip entire, tapering, about 4 mm. long, _upper a little shorter; folded banner and wings broad at top, with a short claw at base; keel strongly falcate, slightly ciliate above the middle where it is 4 mm. wide. This and L. foliosus Nutt. var. stenophyllus Nutt. resemble the illustration of L. laxiflorus in Bot. Reg. pl. 1140, which, as before noted (Leafl. West. Bot. 4:188), Sir William Hooker referred to as L. laxiflorus var. B. In Torrey and Gray Flora of North America (1: 377), under L. laxiflorus var. tenellus, all three are included. From the description, L. rubricaulis Greene (PI. Baker. 3: 35; type-locality, Crested Butte, Colorado), seems to be very similar to L. tenellus. The leaflets are broader and of a different shape. 3. Lupinus spATHULATUS Rydb., Bull. Torr. Bot. Club 29: 244 (1902). (L. parviflorus of Watson in King’s Report, 1871.) Type-locality: Wasatch Mts., Utah, the type, Watson No. 225, collected in 1869. Stems erect and apparently simple, densely leafy with erect short-petioled leaves ; leaflets about 7, narrowly obovate, glabrous above, lightly pubescent beneath; flowers blue, somewhat verticillate, 8—9 mm. long, banner and » keel glabrous; calyx obtuse at base, lower lip twice as long as the upper. In Herb. Calif. Acad. Sci. specimens from the Wasatch Mts., Utah, agree with this description, also some from Wyoming and Colorado. 4. Lupinus PARVIFLORUS Nutt. ex H. & A. Bot. Beech. Voy. Suppl. 336 (1841). Type-region: “plains of the Rocky Mountains towards the Oregon,” the type collected by Nuttall. “A very distinct large and branching species with rather crowded racemes of small blue flowers and large smooth leaves, the leaflets often 1.5 inches in length and half an inch wide, broader upwards ; pedicels rather shorter than the flowers, keel small.” Nuttall Ms. At first the stems are 220 LEAFLETS OF WESTERN BOTANY [ VOL. IV, NO. 9 _ somewhat hairy becoming smooth. The racemes are long and slender, the flowers very small, the calyx silky, and the keel ciliate near the apex. These notes were made from Nuttall’s type in Herb. Mus. Brit. There is a photograph in Herb. Calif. Acad. Sci. which is reproduced here. Lupinus alsophilus Greene, Pitt. 4: 135 (type- locality: mountains above Cimarron, Colorado), from the de- scription, is very similar. 5. Lupinus viripIFoLius Heller, Muhl. 2:64 (1905). Type-locality : “on wooded slopes in rich soil at Dunsmuir,” Siskiyou County, California, the type, Heller No. 7928. Stems several from a woody root, often decumbent, branching and leafy ; leaflets broadly oblanceolate, glabrous above, paler and lightly pubescent beneath ; flowers violet-blue, about 8 mm. long, generally verticillate in long racemes; keel strongly curved, densely bearded near the base. Specimens in Herb. Calif. Acad. Sci. are from the type-region, and also north to Weed and south to Kennet. Specimens from Siskiyou and Trinity counties are also included but are not typical. 6. Lupinus INGRATUS Greene, Pitt. 4:133 (1900). Type-locality : low grassy lands at Chama, New Mexico, the type collected by C. F. Baker. Stems decumbent with many erect and simple branches ; leaves crowded, smooth above, minutely pubescent beneath, leaflets narrowly oblanceolate ; flowers white or sordid, 5—6 mm. long, in densely flowered, narrow racemes; banner pubescent on back; keel glabrous, covered by the wings. This description was made from a specimen collected by C. F. Baker at Chama, New Mexico, loaned from the Baker Herbarium at Pomona College, Claremont, through the kindness of P. A. Munz. A photograph is in Herb. Calif. Acad. Sci. 7. LUPINUS MYRIANTHUs Greene, Pitt. 4: 134 (1900). Type-locality: Gunnison, Colorado, the type collected by Greene in 1896. Stems freely branching above; leaflets 7—9, narrowly oblanceolate, acutish, glabrous above, finely strigulose beneath; flowers violet, about 5 mm. long, verticillate in densely flowered subsessile racemes, keel delicately ciliolate. Specimens collected July 20, 1945, in a meadow 4 miles west of Sapinero, Gunnison County, Colorado, alt. 8300 ft., by H. D. Ripley and R. C. Barneby (No. 7166), agree with the original JANUARY, 1946] SMALL-FLOWERED LUPINES 221 description and since they come from the type-region are un- doubtedly this species. : Key to Group 2 (Species with leaves greenish, pubescence evident on both sides) MRE PRIDE... 54 ai, Se AS Sie AN ee 2 Sumerians er amchitna: t=" 2 Sohal bs AT EO da a a Pe Ae 4 Seales with a shortspuriic2!. tl te ee 8. L. lassenensis SERRE A IITACSS 28 he 8 a al ot ee es 3 3. Keel ciliate about the middle, flowers lavender-blue, about 7 mm. long; pubescence of leaves SCanty.....<2c-ccn---