7 A e+ wong et plop ery we re Peas pope et weet qh rept Rewari Poi ed tg reer hed Averes boty Eas ditoeas soci tl site ae peinaicaa pr roa no vhahttartat 4 stan pemeked ‘ “ td wie CD aes Aisi ba Apoby PE Wi lhe el hh A el Ulsan ucks Apes byreerats whee? RAP ch AA sheharsos Sins farted wat LA edn, oh pag 4 a Fel PA dca caoha yeh Aphe. " ee) pein e maa epereee enrol eve. St yearend fete bene ell, is se yo wae hg A beh he hee § SO SgNae bp 4d poe, + irene chew athe bey rrr bee 4¢ m bf He roe Beet A. 9 Noten ne) we ryprevsle mp rst MED anne Hane ? 7 a ae bbe Adererdtarriny cig ytete Ceara carat se eanteese® betnt rere beams. vy ve dpabdonen oth hide ane sauatn ee 1 tobe) stapieior pee (hate — hella beeen 4 Preih ins Het vy sageret t dn Sib | 4 pethonte ret aaron § FN nes paten, Bs Fry tans fae Wine ARG bent SO Sah he re aes home. seem we ym ON AA Ae ae tg ’ 1 Cott ‘ ey ‘ Stow 64 by aly Eerie , wor herhebablverh ite Beant mci. Ter raat ie Ew i ede bee rh behnersreert satan ts Hrcbyideiogt seta nes ate gd WHE ott ad tm aie geld ‘Hfr ; eee howe speedier) bale geg oe vot orapeebearsteeetae eee > ad : r spovens her Sek LOM. WReOryE 4 OY be 99 sete, Aer ol Paneth ging ‘ abel photel cee beh maar) Aue? Vo ety abe reine 4 11odemm ares ooabin eee MaPedel heen : » wreaeen yer abe tbe bo ¢ Wy 4-808 D Serta mydens 4 Ableton www wtyeevinw roel ore brug te arlttee phn 14 St Aeheee rer bearene epee eare: 40 FY Wetqes Gite war hel + hee rye ered te rr) putts oy baa +17 HE ne ret: IF WI Note oe Naneed hip ‘ ‘ ' r i , tree be “+ temte | afethieaeatate rad of wnt» ) 04.4 , . ‘ + a rT wre SL tt to weewariee iC ‘ . ' ‘ ’ , ib deny shh sete hq berrhege we . - ' ' , W90-> ihe rs On Whe Res Feta e tome nd . ” “ ’ " hm Sn D0 O61 oer Ri eh totes, pekeiene YP Metronet ‘ ’ View . ’ #9 OEE Phe We Oo trmenel rte awl HAPTER Re ep Rees ey fo eB Ante eee te ee lows bevels rare belie PAF 0 obs 14 eg tom Reta he, +74) eM Dy ot begahepmirn ib sptendtg te V wie ebete nee — + Ne parte On w 100) he sOEe omy Fwy Ly eeoeeaey wt "Ol petra $4 Bienen § a0 Bribe = “ eaesen =i Law ete: : beat ene 1 0 f dourpe duke reps 06 an re Vw eee Sh mels oiebeetes A ' ny ' “ vel oe ae , . 1's HEH Vee Ww pi oer aneerc ate : p : } ; aa , PL he eg bis tb ’ oy OR bores 4 r WIP i . ” ‘ , ‘ te hivwrys : e% ’ 1 inne ap bet copeer - r : ¥ ’ ‘ oui * hieig ‘ OF fee Fey tng . ity " ' “4 ’ aaa A tea ‘ oh fate me bed wa . ’ 4 sj * we *, - wi > * Pes i. mA tis) y vib re toe i ‘ #10 ba eT) 0: de ya pow jplowy te Bab tet et . f , ’ ‘ . eee Wit le ’ " , 4 6 sme NORTES a penertge vat ine wer oes t . oe hie hy ce oon “ , "i ‘ owe 9 ' , ” Pp howe ' re ’ ele en ne ) ot ies we Ae ' A . We ye J . ‘” hur aod : i e™ ive) tien’ 2m) trans ae ‘ ’ } } 3 ¢ ' ' re ony F vs . ‘ . ro aphasia Lhe be Py ladnd ne y ' . > or cere ? » ¢ 12 A vr — + mein gvdbled-sehelaawine we 3 g ; ® PP Wied be ~—T ee ee igus tek : / ores ta . We svebt sehr ee ow *h m eh y : <2 het beter rs ; . “ Ve tp , ’ j he ‘ f we +> ree lope » Yeh bets we tun 2 - \ se if ’ . tare Pr et pe ‘ + otewe tenet ie! hey ' ‘4 oer sie Da hiieee led Kataheakat 4 orks , peeps , n 7 a ony erie Ye ‘ ‘ my 2 “ Vee) hehe bade pean we : ‘ +) thet err f , . . er +12 oom be eS Nn hele tym | ——- : U " , py A ‘ . ¥ feidiedlel ot Bere { 5 P . 4 ’ fe er mee * ve +) 4 bere) en OR. oa ; alos . Add ~ 1 Ob eee tpehet > A ae. : ' wor Frnats - “ ; - - se we . wlorsp >a Vgh i : f / P ; ' ' "” ee : ¢ ‘ ye ae, Cs! : pian 7; - ‘ ee “aelee A z erie i sete ‘ - ay * ‘ e ‘ +4 > ‘ ‘ ‘ “oan Lae. 4 ' Lae) “eu * 1OVMUS Wem, * hee wn Pruairaspean bs u f vy ’ ad Oa ‘ >, pelos . bh Ligh suo AAD + Ont) TR REE ee ‘ ‘ a v ’ ny j ee ed ORD PU 9 0 om oe wgtahene r t be ¥ 7 re ARR S P-ONH SE PR ee if * % ; ' pire mhry ion ae 108 60 yr ON — het me. , ' 10 Gen p avy: eqs rhb » Ae . * z LU f ‘ ’ peomeny bye ” '” ‘pg . fare “ owe ; , ’ 4 , renee . we on PE ote ' . per ’ ay e* ' rf me ; Ps Wier Oty Mine ne ' i r J ¥ ? if 1) epee - use + Were ge anew tre sry ‘ 7 ‘ ' ‘ Pa ee has ie hw ‘ eeremny ‘ PORTO BAT ip mage ¥ . ‘ , . ‘ w ’ vere Pre He om ' pris : - eye wae: ' +88 9A. pigeon ee ‘ * wy ‘ ' , ' v werns rermsens ; A omy vere weeny ’ ‘ ape aie ‘ Wo We heap ins . re ery ‘ ne trey te , , , A We . + Gi 44 wens wmarn porereberues ‘ae hat ‘ Dek) eet nee ae +) mere kt , : = ; TY Oe mareee wee OF PGP rea: & anh be renee ed t ’ wee a) vio eee ee WOE OSE ree Wgewe AA arm a TT BHPT om Syed), Chee oe ’ tyes hg’ ve wipe ' wen ree PU erie bak Ht hette ao ’ ‘ ‘ ore rer a et ery Se , oe tee ae ’ ' ‘ ) ' elatde é J i. fs i x | 28 LIBRA | [P>. THE NEW YORK BO 3 2 ae ¥ ¥ ‘ 0 ‘ ML 5% ’ une ive: ; : ae ee re , ~~ a > =, '¢ rr phe # od ut . ¥ aN r = aide tl af y Pg ae 7 ABS wa LER —- 4 y ee ty &) ko , A ' gi t} x i} Se ee a) Vee i mY ne i (any ork WY Aa Nes TANG Uy he LEAFLETS OF WESTERN BOTANY VoLuME II SAN FRANCISCO, CALIFORNIA 1937-1940 Owned and published by A.icEe Eastwoop and JOHN THomas HoweELL Printed by THE JAMES H. BARRY COMPANY SAN FRANCISCO Vot. II No. 1 LEAFLETS of WESTERN BOTANY Y CONTENTS PAGE New or Imperfectly Known Californian Species of Downingia | Rosert F. Hoover mee muecies..of. Western Plants, 6). a te ses Atice Eastwoop Three Species of Gnaphalium Adventive in California . . 10 Joun THomas Howe Lv RN Cee V LLP. je dt te br RS J. W. Stacey SAN FRANCISCO, CALIFORNIA JANvuARY 27, 1937 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. WEstT. Bor. POA UU RS PORE ALA ET CT Owned and published by A.icE Eastwoop and JOHN THomas Howe. ae ele ee ee ae, ae < 2 28. Ges, ee Seb Re) | Lae > 5 eal Sd ry & VORK. Wy OY URS t a4 yf AS GA @ai DEN NEW OR IMPERFECTLY KNOWN CALIFORNIAN SPECIES OF DOWNINGIA BY ROBERT F. HOOVER University of California, Berkeley The author of this article has had many excellent oppor- tunities to observe plants of the genus Downingia in the field. The results of these observations are given here as far as they apply to forms heretofore undescribed and to range extensions of previously named species. It is a generally accepted principle that two very closely related species, if actually distinct, seldom grow together. Specific limits as recognized in this article are in accord with that principle. In most cases the colonies of Downingia consist of one species only, even when conditions seem equally suitable for others. The herbaria in which specimens cited here are located are indicated as follows: Herbarium of W. L. Jepson (J), Her- barium of the California Academy of Sciences (CA), Dudley Herbarium of Stanford University (S), Herbarium of the University of California (UC). The types of the new species and varieties are deposited in the Herbarium of W. L. Jepson, with duplicates in the California Academy of Sciences and the University of California. DoWNINGIA PULCHELLA (Lindl.) Torr. In alkaline beds of vernal pools near the sand dunes 10 miles west of Merced, Merced Co., May 18, 1936, Hoover No. 1126. This locality is farther south than any previously reported for this species in interior California. The flowers had all the distinctive charac- ters of the species, such as the long divergent lobes of the upper corolla-lip and the fully exserted anther-tube. Downingia pallida Hoover, spec. nov. Caulibus 5—10 cm. altis et rectis vel longioribus et diffuse ramosis; foliis oblongis ad linearibus, 5—10 mm. longis; lobis calycis equilongis tuba corollz, sub anthesi ascen- dentibus, patentibus et szepe accrescentibus in fructu; corolla czsia, labio inferiore loco albo centrali flavyo in medio, tuba corollz tenui, lobis inferi- oribus superioribusque in eodem plano, labio superiore semi-breviore quam inferius; sinibus inter labia corolle2 haud extendentibus trans limbum corollz; lobis labii superioris ovatis, subparallelis; labio inferiore plano, Leafl. West. Bot., Vol. II, pp. 1-16, January 27, 1937. 2 LEAFLETS OF WESTERN BOTANY [voL. II, NO. I profunde partito, basi sine rugis, gibbis vel maculis; tuba antherarum brevi et crassa, breviore vel paulum longiore quam tuba corolle, apice pilis duobus brevibus vel pilis nullis. Stems 5—10 cm. tall, erect, or longer and diffusely branched; leaves oblong to linear, 5—10 mm. long; calyx-lobes equalling corolla-tube, ascending in flower, in fruit becoming rotate and often accrescent; corolla pale blue, the lower lip with central white area with yellow center; corolla-tube slender; sinuses between lips of corolla not extending beyond plane of lower lip; lobes of upper lip ovate, subparallel, in same plane as lower lip, about half as long as lower lip; lower lip plane, deeply parted, no folds, projections, or spots at the base; anther-tube short and thick, included or partly exserted, with or without 2 short bristles at apex. Central Sierra Nevada foothills. Type collection: in stream beds, Warnerville, Stanislaus Co., April 29, 1936, Hoover No. 1042 (J, type, CA, UC). Other collections: Mokelumne Hill, Calaveras Co., F. E. Blatsdell (CA); bed of former pool, 4.7 miles from San Andreas on road to Valley Springs, Cala- veras Co., J.T. Howell No. 4702 (CA); Salt Springs Valley, Calaveras Co., Tracy No. 5649 (J); Sonora, Tuolumne Co., April 25, 1925, E. A. Green (S); bed of former winter pool, 1 mile north of Friant, Madera Co., Jepson No. 12903 (J). This species seems nearest to D. pulchella. There are differ- ences, however, which because they are constant are thought to indicate specific rank. In D. pallida the calyx-lobes are ascending until the corolla withers; in D. pulchella they are rotate from the beginning. The lobes of the upper corolla-lip in D. pallida are subparallel and shorter than the strongly diver- gent ones of D. pulchella. The lower lip of D. pallida lacks the two yellow folds and three purple spots at the base which are present in D. pulchella. The anther-tube of D. pallida is much shorter and often lacks the two bristles which are present at the apex in D. pulchella. Even the color of the corolla appears to be constant in this case, being a uniform shade of pale blue in all plants of D. pallida seen. At the Warnerville locality it was noted that the plants described here grew only in stream beds, while D. ornatissima and D. bicornuta grew together in adjacent vernal pools. In Calaveras County, however, D. pallida also is said to grow in beds of pools. Downingia bella Hoover, spec. nov. Caulibus paucis ex basi, crassis fistulosisque, decumbentibus vel ascendentibus, 4—15 cm. longis; foliis JANUARY, 1937] CALIFORNIAN SPECIES OF DOWNINGIA 3 oblongis ad linearibus, 5—12 mm. longis; lobis calycis rotatis; corolla cyanea, labio inferiore loco albo centrali flavo in medio; lobis labii superi- oris semi-ovatis vel lanceolatis, subparallelis, circa semi-brevioribus quam labium inferius; labio inferiore plano, basi in ore faucis gibbis duobus flavis, gibbis et tribus parvis purpureis maculis alternantibus ; sinibus inter labia corolle parallelis cum axi tube corolle et extendentibus paulo trans planum labii inferioris; columna staminali breviore vel paulum longiore quam tuba corollz, apice pilis duobus retrorso-curvatis; capsulis late divaricatis. Stems few from the base, stout and fistulous, decumbent or ascending, 4—15 cm. long; leaves oblong to linear, 5—12 mm. long; calyx-lobes rotate; corolla deep bright blue, the lower lip with central white area with yellow center; lobes of upper lip semi-ovate or lanceolate, nearly parallel, about half as long as lower lip; lower lip plane, with 2 yellow projections at the base in the mouth of the throat, with a small purple spot between and on either side of the projections ; sinuses between corolla- lips parallel to axis of tube, extending a short distance beyond plane of lower lip; stamen-column included or slightly exserted, with 2 short downward curving bristles at the apex; capsules spreading at a wide angle from the stem. Vernal pools of alkaline plains on the east side of the San Joaquin Valley. Type collection: near San Joaquin River southwest of Modesto, April 1, 1936, Hoover No. 837 (J, type, CA, UC). Other collections seen are as follows. Stanislaus Co.: 10 miles west of Modesto, Hoover No. 386; from type locality, Hoover No. 563. Merced Co.: Merced, J. T. Howell No. 4168 (CA); 6 miles south of Merced, Hoover No. 953. Tulare Co.: 5 miles east of Traver, Hoover No. 1015; 4 miles north of Visalia, Hoover No. 921; near Pixley, April 3, 1917, Eastwood (CA). This is the only Downingia known at present from the valley plains of Tulare County, where it is locally abundant. Farther north it is less common and is limited to the low plains near the trough of the valley, while other species of Downingia are found on the higher plains and low foothills to the east. Downingia bella comes into flower about two weeks earlier than the other species of Downingia of this region (the last week in March, or perhaps even earlier in dry years). So far as I have observed, the soil where it grows is always somewhat alkaline. While the corolla is generally as described, some variations can be found. The central one of the three purple spots at the base of the lower lip may be lacking. The lobes of the upper 4 LEAFLETS OF WESTERN BOTANY [ VOL. II, NO. I lip, which usually are in the same plane as the lower lip, fre- quently curve backward somewhat, but are always nearly parallel, never curving to the sides of the corolla as in D. ornatissima. The bristles of the anther-tube, suggesting walrus tusks, help to distinguish this species, but sometimes they are very short or apparently absent. Dowmingia bella is about equally different from several species, but may be most closely related to D. pulchella, with which it is identical in the color-pattern of the corolla, differing in having the main sinuses of the corolla cut below the platform, in the short approximate lobes of the upper lip, and in the included anther-tube. It resembles D. concolor Greene in the shape of the corolla, but differs in color-markings and in its stout stems and spreading habit. Downingia bicornuta Gray var. picta Hoover, var. nov. Corolla cesia; tuba brevissima et lata, superne ochracea; labio inferiore valde concavo, basi cornibus duobus prominentissimis atro-purpureis curvatis oblique; lobis labii superioris albis vel czesiis, sepe cyaneis apicibus, lobis haud divergentibus immo transversariis inter se, acute declinatis et ap- pressis ad tubam corolle; pilis apice tube antherarum tortis inter se, multum curvatis, tuba antherarum longioribus. Corolla light blue; tube very short and broad, brownish-yellow on side next to upper lip; lower lip strongly concave, with 2 very prominent dark purple horns at the base which curve to the sides; lobes of upper lip white or pale blue, often tipped with darker blue, not divergent but directed toward each other so that the tips cross, sharply reflexed and appressed to corolla-tube; bristles at apex of anther-tube twisted together, strongly curved and longer than anther-tube. Type collection: gravelly stream bed 7 miles southeast of Le Grand, Merced Co., May 1, 1936, Hoover No. 1083 (J, type, CA, UC), the extreme form, quite different from D. bicornuta. Pleasant Grove, Sutter Co., Hoover No. 1142, similar to the type. The following are intermediate in some respects between this form and typical D. bicornuta, but all have the brownish spot in the corolla-tube: Amsterdam, Merced Co., Hoover No. 598; 2 miles northwest of Merced on Oakdale road, Merced Co., J. T. Howell No. 4174 (CA) ; 12 miles east of Merced on Yosemite highway, Hoover No. 615; Rio Linda, Sacramento Co., Jepson No. 16571 (J). Although this variety appears quite distinct from D. bi- cornuta in the extreme form, the intergrades prohibit its accept- JANUARY, 1937] CALIFORNIAN SPECIES OF DOWNINGIA 5 ance as a species. It appears to be localized in Sutter County and Merced County. Elsewhere in the Great Valley the species has been seen only in the typical form, though with some trivial variations. The characters of typical D. bicornuta contrasted with the above description are as follows: corolla usually deep blue or violet (but sometimes white); corolla-tube broad but not extremely short, with no brownish-yellow ; lower lip nearly plane, the horns at the base straight, prominent but less so than in var. picta; lobes of upper lip deep blue or violet, divergent, recurved but not appressed to the tube; bristles at apex of anther-tube (similarly twisted together) less strongly curved, shorter than the anther-tube. DowNINGIA ORNATISSIMA Greene. The typical form of this species, as indicated by Greene’s description of plants from Elmira, has pale blue or nearly white flowers. Similar plants are found on the east side of the Sacramento Valley. In the lower San Joaquin Valley most plants of this species have the corolla bright or dark blue, and the anther-tube is often fully exserted. My collections of the latter form are listed below. San Joaquin Co.: Farmington, Hoover No. 1054. Stanislaus Co.: Knight’s Ferry, Hoover No. 1033; Warnerville, Hoover No. 1044; 4 miles south of Oakdale, Hoover No. 488 and 1021 ; Modesto, Hoover No. 566; Gobin Ranch, Hoover No. 1101. Merced Co.: Ryer, Hoover No. 1075; 3 miles north of Snelling, Hoover No. 961; 7 miles southeast of Le Grand, Hoover No. 1084. DowNINGIA MIRABILIS J. T. Howell. Although this species was described from the type collection only, it is actually common on the east side of the San Joaquin Valley from Stanis- laus County to Fresno County. It has been seldom collected except by the present writer. The collections are listed as follows. Stanislaus Co.: Montpellier, Hoover No. 585. Merced Co.: Ryer, Hoover No. 1074; 6 miles southeast of Le Grand, Hoover No. 1081. Madera Co.: Berenda, Hoover No. 906; Madera, Hoover No. 1248; 6 miles north of Madera, Abrams No. 11638 (S); 6 miles east of Madera, April 24, 1927, A. G. Vestal (S). Fresno Co.: Pinedale, Hoover No. 984; Herndon Avenue east of Pinedale, April 17, 1932, Springer (J. A. Ewan Herb.) ; 5 miles east of Clovis, Hoover No. 996. 6 LEAFLETS OF WESTERN BOTANY [VOL. I, NO. I Downingta mirabilis does not extend north of the Tuolumne River, while the related D. ornatissima is not known south of Merced County. In the region where the two overlap they appear quite distinct from each other. Although the two spe- cies are often found at approximately the same locality, they never actually grow together. This has been observed southeast of Le Grand in Merced County, where they grow exclusively less than a half mile apart, separated by a broad creek bottom. Near Ryer station a large colony of D. mirabilis was seen in one “hog-wallow,” while only a few yards away D. ornatissima and D. bicornuta grew to the exclusion of the former species. When growing under favorable conditions, D. mirabilis has larger broader leaves of a lighter green than D. ornatissima. The flowers are also considerably larger and paler than in the form of D. ornatissima found in the San Joaquin Valley, in color resembling the pale-flowered D. ornatissima of the Sacra- mento Valley. Downingia mirabilis J. T. Howell var. eximia Hoover, var. nov. Lobis labii superioris corollz ascendentibus, appressis inter se vel paulum divergentibus, nec recurvis nec patentibus horizontaliter quidem in senectute. Lobes of upper corolla-lip ascending, appressed to each other by the flat surfaces or slightly divergent, not horizontally spreading or recurved even in age. Type collection: 3 miles west of Orange Cove, Fresno Co., April 28, 1936, Hoover No. 1000 (J, type, CA, UC). Also col- lected 1 mile east of Orange Cove, Tulare Co., Hoover No. 1011, and at Woodlake, Tulare Co., Hoover No. 1285. This form is confined to the vicinity of the town of Orange Cove and northern Tulare County on the edge of the foothills. Growing with it at Orange Cove were a few plants of typical D. mirabilis, which reaches its southern limit there. The latter, distinguished by having the lobes of the upper corolla-lip hori- zontally spreading with recurved tips, is the only form of the species found north of Kings River. DowNINGIA HUMILIs Greene. Near La Grange, Stanislaus , Co., on the road to Snelling, April 13, 1936, Hoover No. 970. The plants are identical with those of Sonoma County, where the species was first known. JANUARY, 1937] | NEW SPECIES OF WESTERN PLANTS 7 NEW SPECIES OF WESTERN PLANTS By ALICE EASTWOOD Aquilegia emarginata Eastwood, spec. nov. Similis ad A. truncato F. & M., sed petalis emarginatis et ubique preter stamina glanduloso- pubescens; calcaribus incurvis. Type: No. 232179, Herb. Calif. Acad. Sci., collected April 20, 1934, three miles south of Klamath Junction, Jackson Co., Oregon, by Alice Eastwood and John Thomas Howell, No. 1729. In general appearance this resembles the common A. trun- cata, but the petals instead of being truncate at the apex are emarginate. Aquilegia truncata also is smooth while this is glandular-pubescent throughout, even to the flowers and fruit. Only the stamens are smooth. The spurs in bud curve inwards, later spreading. Thelypodium stamineum Eastwood, spec. nov. Annuum, elatum, paniculato-ramosum, pilosum, pilis simplicibus vel interdum furcatis ; foliis irregulariter lobatis, lobis obtusis, alternantibus, decurrentibus rhachadi, infimis foliis 1 dm. longis; floribus parvis, viridibus; sepalis oblongis, ob- tusis, albo-marginatis, ca. 5 mm. longis, 1 mm. latis, parce pilosis; petalis linearibus, margine crispis, purpurascentibus, ca. 9 mm. longis, 0.5 mm. latis; antheris linearibus, basi sagittatis, superantibus calycem et corol- lam; siliquis filiformibus, divaricatis, longissimis 1 dm. longis. Type: No. 232117, Herb. Calif. Acad. Sci., collected on the Victory Highway, 10 miles east of Battle Mountain, Nevada, June 10, 1933, by Alice Eastwood and John Thomas Howell, No. 165b. This species is chiefly distinguished by the small greenish flowers with the exserted anthers. The numerous pods are long and thread-like, with the single row of seeds evident so as to be almost moniliform. The pubescence is more abundant at the base of the plant and is of white spreading hairs. The leaves vary, diminishing upwards, but apparently somewhat lobed. The racemes are bractless and the pedicels in fruit are about 5 mm. long. The sepals seem connivent in the obtuse buds, but later spread. The very narrow petals have strongly crisped margins and are purplish along the center. The stigma is very slightly 2-lobed. Lotus Leonis Eastwood, spec. nov. Perennis, basi ramosus, canes- cens, appresso-pubescens ; caulibus ascendentibus, dense foliosis, 1—1.5 dm. longis; foliis petiolatis, foliolis 3—5, subobovatis, 5—8 mm. longis, 3—5 8 LEAFLETS OF WESTERN BOTANY [ VOL. II, NO. I mm. latis; umbellis axillaribus pedunculatis, 3-bracteatis vel sine bracteis, 5—10 floribus, pedicellatis; calyce campanulato, 5 mm. longo, segmentis deltoideis, attenuatis, circa 2 mm. longis; corolla straminea, 1 cm. longa, vexillo orbiculato-obovato, 5 mm. longo, ungue 2 mm., alis angustis, super- antibus carinam; legumine immaturo, falcato, attenuato, appresso-pubes- cente. Type: No. 227881, Herb. Calif. Acad. Sci., collected June 3, 1934, in the dry stream bed of Hackamore River, Hackamore, Modoc Co., California, elevation about 1600 m., by Leo Whitney, No. 1920, for whom it is named. It belongs to the aggregate under Lotus Douglasw Greene, but differs from all described in the densely leafy stems with internodes very short, the grey-green color of the plant, and the pale straw-color of the flowers. Arctostaphylos obispoensis Eastwood, spec. nov. Frutex pallidus, 1—2 m. altus, erectus, ramosus ; caulibus senioribus glabris, atro-purpureis, junioribus albo-tomentosis ; foliis ovatis vel oblongis raro lanceolatis, apice mucronatis, basi cordatis vel truncatis, pallidis, leviter tomentosis, in senectute prope glabratis, maximis foliis circa 4 cm. longis et 2 cm. latis, petiolis 5 mm. longis; paniculis subsessilibus tomentosis, bracteis foliaceis lanceolatis acuminatis pedicellos superantibus; pedicellis prope glabris; floribus pallido-roseis, 7 mm. longis; fructu glabro plano-globoso rosaceo, circa 1 cm. diametro. Type in Herb. Cal. Acad. Sci., No. 165817, collected by Alice Eastwood in flower, March 7, 1928, No. 16843, and in fruit, Eastwood No. 15125, May 17, 1928, Herb. Cal. Acad. Sci., No. 158925. The bushes grew in a serpentine area up Chorro Creek, near San Luis Obispo. Collections were made also by Eastwood and Howell, No. 2278, east of the Morro dis- trict on the new road to Atascadero, San Luis Obispo County, May 7, 1936. This is closely related to A. canescens Eastwood and A. auriculata Eastwood and also approaches A. pechoensis Dud- ley. It is distinguished from A. canescens by the generally cordate bases of the leaves and the smooth fruits and from A. auriculata and A. pechoensis by the petiolate leaves. It is the dominant shrub growing in the serpentine area on the upper elevations of Chorro Creek and when in flower with the pale foliage and pink flowers (sometimes white), the effect is beautiful. The pubescence is white-tomentose, entirely without glandular or spreading hairs. The heart-shaped leaves and JANUARY, 1937] NEW SPECIES OF WESTERN PLANTS 9 general aspect of the shrubs are distinctive. However, this and the related shrubs might all be considered as subspecies under A. canescens. Oreocarya capitata Eastwood, spec. nov. Perennis, ramosa ex cau- dice lignoso, ca. 2 dm. alta, strigosa, viridis; foliis aggregatis basi, lineari- lanceolatis, subacuminatis, 10—15 cm. longis, 1—5 mm. latis, viridibus, pilis appressis, pustulatis in senectute, pustulis albis, nitidis, tessellatis ; foliis caulinis paucis, parvis, strigosis et hispido-ciliatis; inflorescentia capitata, superante folios; sepalis lineari-lanceolatis, 1—4 mm. longis, 1 mm. latis, divaricate hispidis; corolla alba, 8—10 mm. longa, lamina 8 mm. diametro, segmentis orbiculatis, fornicibus flavis, exsertis 2 mm.; staminibus supra medio tube, antheris linearibus, 3 mm. longis; nuculis dorso ovatis, obtusis, 4 mm. longis, 2.5 mm. latis, acute marginatis glabris et nitidis, ventro acute costatis, hilo basilare, parvo. Type: No. 232041, Herb. Calif. Acad. Sci., collected by the author, No. 5969, on the Hermit Trail, Grand Canyon of the Colorado, Arizona, April 9, 1917. This specimen is in flower. A fruiting specimen, Herb. Calif. Acad. Sci. No. 232040, was collected by the author on the same trail, June 18, 1916. It was collected June 23, 1933, on the north side of the Grand Canyon on the Kaibab Trail to Roaring Springs, Eastwood & Howell No. 1005, Herb. Calif. Acad. Sci. No. 232039. This species comes under the aggregate typified by O. con- fertiflora Greene. It differs from the type of which there is a duplicate in Herb. Calif. Acad. Sci. in different pubescence, the green color of herbage, the very narrow leaves, the white flowers, and a more hispid inflorescence. It is unlike any of the other species or varieties that are related in the color of the flowers. “It has been fashionable in some quarters in modern times to decry both the importance and the value of systematic botany. Because of its validity, its human interest, its practical bearing on other phases of plant science, and on our everyday life, one suspects that some of its critics have lacked the breadth of view of leaders in science, and have been misguided in criti- cising that which they did not fully understand.”—Dr. E. D. Merrill, Memoirs Brooklyn Bot. Gard. 4:57 (1936). ice) LEAFLETS OF WESTERN BOTANY [ VOL. II, NO, I THREE SPECIES OF GNAPHALIUM ADVENTIVE IN CALIFORNIA BY JOHN THOMAS HOWELL How long a weedy immigrant can flourish in our midst and escape detection because it simulates another well known species is a question which, because of its very nature, is scarcely to be answered. Yet such a questioning thought is unavoidable when one considers the occurrence of Gnaphalium luteo-album L. in California where it is already an aggressive weed in several districts and promises to become one of our commonest plant inhabitants. Until it was indicated to me by Dr. S. F. Blake of Washington, D. C., as an undetermined weed new to our flora, the plant would have been passed as a slender phase of G. chilense Spreng. which is so abundant in our lower hills and valleys. It was in June, 1934, that Dr. Blake pointed out G. luteo- album to me at Stanford University and since then observations have shown it to be one of the common weeds of the San Fran- cisco Peninsula in Santa Clara, San Mateo, and San Francisco counties. As the writer once remarked about another rampant immigrant, Lactuca saligna, our new Gnaphalium is making itself right at home and has all the aspects of a native in the territory of its adoption. Thinking that earlier Californian collections of G. luteo- album might be concealed under the label of G. chilense, I sought in the herbaria of the San Francisco Bay region for further distributional and temporal data. Only a single speci- men was uncovered, one collected by myself in 1929 on the Merced River near Merced Falls in Merced County (Howell No. 4154). No other collection from the San Joaquin Valley was seen until this year when two specimens came to my at- tention: the first a specimen collected by Gordon H. True (No. 452) near Los Banos in the western part of Merced County, the second a specimen communicated by Prof. E. E. Stanford from an alfalfa field at Lodi, San Joaquin County; and during the past summer when on an excursion to the Merced River district, I observed G. luteo-album in dense vigor- ous clumps along irrigation ditches. From Southern California only a single specimen has been seen, one collected by Miss JANUARY, 1937] GNAPHALIUM 11 Eastwood in the Rancho Santa Ana Botanic Garden in Orange County, April 5, 1936. It is not always easy to distinguish G. luteo-album from G. chilense even after the two have been seen apart. For the most part G. chilense is a larger plant with leaf-bases more decurrent and with heads larger and more decidedly yellowish or straw-color. In G. luteo-album the leaves half clasp the stem and the smaller heads are more pearly-shining with a distinctly greenish to light brownish tinge. But in the tiny flowers and achenes there are definite differences. In G. chilense the corollas are yellowish, the achenes are glabrous, and the pappus-bristles generally separate when they fall from the achene. In G. luteo- album the corollas are a deep maroon or purplish-red at the outer end, the achenes are minutely scabrous-pubescent, and the pappus-bristles tend to cohere at their somewhat thickened, slightly hairy bases. The advent and spread of G. luteo-albwm in California is not surprising since it is a weedy plant in many parts of the world. Although it is reported by Hegi as occurring in America outside of the arctic regions, it has never been listed in those floras of the United States that have come to my attention. 7 7 7 The case of Gnaphalium japonicum Thunb. is quite differ- ent. Time and again this very distinctive plant has been collected by Mr. Joseph P. Tracy, of Eureka, California, but it has remained undetermined for years although a fine series of speci- mens from Mr. Tracy has been on file in the Herbarium of the University of California. When one of Mr. Tracy’s collections came to my attention about two years ago, it attracted me at once because it was so different from any Gnaphalium I had ever seen, obviously a plant which could not “hide out” behind any known Californian species as G. luteo-album had done. Here was an annual remarkable for its inflorescence, the capitulz being densely congested into regular globular heads which were subtended by an involucre of elongate foliage leaves. This striking plant has been collected by Mr. Tracy at a number of stations in Humboldt County, California, and one collection comes from as far east as Sharber Slough on the Trinity River, Trinity County (Tracy No. 7783). Mr. Tracy’s earliest collection is his No. 4677 obtained September 12, 1915. 12 LEAFLETS OF WESTERN BOTANY [ VOL. II, NO. I During the past summer the writer found the plant on the west side of the Eel River south of Miranda, Humboldt County (Howell No. 12888). Here in a dry grassy open in the red- wood forest it was associated with no fewer than four other species of Gnaphalium and it appeared to be a thriving immi- grant. Mr. Tracy reports finding it abundant in certain recently logged areas where it behaved as a fire-weed. A widely dis- persed native of the Australasian district from Tasmania and New Zealand to Japan, G. japonicum is likely to become as common through our coastal hills as the two Australian species of Erectites which have made themselves so, much at home hereabouts. oA tf r To Mr. Tracy the writer is also indebted for material and field observations on another Australian species of Gnaphalium which has become established in northwestern California, G. col- linum Labill. This species is closely related to G. japonicum and it was referred to that species by the writer until its dis- tinctive characters were indicated by Mr. Tracy, whose remarks, interestingly enough, have been found to coincide almost exactly with Bentham’s comments on the species in Flora Australiensis. Gnaphalium collinum is a rhizotomous perennial with less com- pact heads subtended by fewer and shorter involucral leaves, and, according to Mr. Tracy as he knows it, “it acts like a thoroughly fixed species and has no particular relation to the annual G. japonicum in distribution.” It is also of interest to note that whereas G. japonicum has been collected only from August 29 (Howell No. 12888) to October 6 (Tracy No. 5118), G. collinum has not been collected later than August 13 (Tracy No. 10349) and has been collected as early as March 22 (Tracy No. 14784). This interesting introduction was found by Mr. Tracy as long ago as 1904 at Eureka, Humboldt County (Tracy No. 2134), and since then he has made a number of collections, northward along the coast to southern Del Norte County and eastward to the vicinity of Korbel, Humboldt County. His col- lections present rather diverse aspects in habit and inflores- cence, and although these are believed to be merely responses to local environmental conditions, variations only little more pronounced than these have received nomenclatorial recognition in Australia and Tasmania where the species is native. JANUARY, 1937] NOTES ON CAREX 13 NOTES ON CAREX—VIII BY J. W. STACEY Carex curatorium Stacey, spec. nov. Dioica; rhizomatibus congesto- implicatis, breviter repentibus, atro-purpureis; culmis 2.5—4.5 dm. altis, basi purpureo-rubris, valde aphyllopodis; foliis 4—6 ex singulis culmis fertilibus, laminis amplis, 5—20 cm. longis, 2—3 mm. latis, vaginis demum fractis et paulum filamentosis; spicis rectis, solitariis; spicis masculis linearibus, 2.5—3 cm. longis, 3 mm. latis; spicis femineis linearibus, 2.5—3.5 cm. longis, 3—4 mm. latis; squamis oblongis vel lanceolato-ovatis, pubescentibus, rubescenti-purpureis, pallidioribus in medio, angustioribus et multum brevioribus quam perigynia; perigyniis in spicis manifestissi- mis, late oblongo-ovatis, planis, 3.5—4.5 mm. longis, 2—2.25 mm. latis, dense pubescentibus, helvolis, abrupte brevi-rostratis; acheniis trigonis, 2 mm. longis, 1.5 mm. latis, castaneo-tinctis. Dicecious; rootstocks densely matted, short-creeping, purplish-black, lignescent, stout; culms arising several together, somewhat decumbent at base, 2.5—4.5 dm. high, exceeding the leaves, purplish-red at base, strongly aphyllopodic; leaves with well developed blades, 4—6 to a fertile culm on the lower third but not clustered, the blades erect, channeled, 5—20 cm. long, 2—3 mm. wide, attenuate; sheaths puberulent, hyaline, yellowish- tinged ventrally, breaking and becoming somewhat filamentose; spike erect, solitary; staminate spike linear, 2.5—3 cm. long, 3 mm. wide, scales acutish to obtuse, reddish-brown with lighter center; pistillate spike linear, 2.5—3.5 cm. long, 3—4 mm. wide, closely flowered above, more loosely at base, bract, if present, about 5 mm. long, inserted about 5 mm. below spike, sometimes subtending a perigynium, scales oblong or lance- ovate, pubescent, reddish-purple, with lighter center and narrow hyaline margins, narrower and much shorter than the perigynia, so that the peri- gynia are very conspicuous in the spikes; perigynia broadly oblong-ovate, flattish, densely pubescent, few-nerved, yellowish-gray, tinged with green, especially on the margins, abruptly contracted into a short beak; achenes triangular with concave sides, loosely enveloped in the middle of the perigynia, 2 mm. long, 1.5 mm. wide, chestnut-brown, short-stipitate, apiculate, jointed with the slender style; stigmas 3, short. Types: staminate plant, Herb. Calif. Acad. Sci., No. 204973, Eastwood and Howell No. 1101; pistillate plant, Herb. Calif. Acad. Sci., No. 204974, Eastwood and Howell No. 1100, col- lected June 23, 1933, on Kaibab Trail to Roaring Springs, Grand Canyon National Park, Arizona. Two other collections made with the type of the pistillate plant bear the numbers 1045 and 1089. Duplicates of some of these collections were distributed as C. pseudoscirpoidea Rydb., an early determination reported in the literature (Leafl. West. Bot. 1: 142, 1934). Carex cura- torium is named in honor of the collectors, Miss Alice Eastwood, 14 LEAFLETS OF WESTERN BOTANY [ VOL. II, NO. I curator, and Mr. J. T. Howell, assistant curator, Department of Botany, California Academy of Sciences, from both of whom the author has received the utmost assistance and courtesy in the assembling of a large collection of Carices and in the con- tinuous use of the herbarium for study. This species belongs to the section Scirpine Tuckerm., and stands out from all other species in this section in that the peri- gynia are very much exposed, giving a grayish appearance to the pistillate spikes. eee) tee The rare Carex gigas (Holm) Mackenzie was collected by J. T. Howell, No. 12725, on the summit of Scott Mts., north of Carrville, Trinity Co., California, August 24, 1936. As the plant was strongly aphyllopodic, we were puzzled by Mac- kenzie’s key in North American Flora, where it is placed as phyllopodic. Through the courtesy of the United States Na- tional Herbarium and the Gray Herbarium, sheets of the type and other material, collected at Mt. Eddy and Grizzly Hill, both in Siskiyou County, were sent. It was found that the borrowed specimens were also aphyllopodic, and that the only difference in the plants was that the one collected by J. T. Howell was taller and ranker, as might be expected from the fact that it was collected at a lower altitude. Although Mackenzie keys C. gigas as phyllopodic, in the text he has “phyllopodic (but with basal scales), which is contradictory, for if it has basal scales, this very fact makes the plant aphyllopodic. With this new col- lection, the description of Mackenzie should be revised to read “culms 3—6 dm. high” instead of “3—4.5 dm. high,” and “pistil- late spikes 1.5—4.5 cm. long” instead of “1.5—2.5 cm. long.” As Mackenzie has not very well differentiated C. gigas in his key, and as a new species has been added to the section, a key to all known species of the section Scirpine is herewith given. Culms aphyllopodic, strongly purple-tinged at base. Perigynia compressed-triangular, 2.5—3.5 mm. long. Perigynia short-whitish-pubescent; scales very narrowly Viki hele ovr) aloe i Leb flesk les Rt RT RE AOR Ep alas C. scirpoidea Perigynia hirsute, yellowish-brown or greenish-brown; scales broadly hyaline-margined ...........-...-22-.-2se-eeeee C. scirpiformis JANUARY, 1937] NOTES ON CAREX 15 Perigynia flattish, 3—4 mm. long. Perigynia oblanceolate or lanceolate.......................... C. stenochlena Perigynia broader, ovate to obovoid. Perigynia broadly obovoid; scales as long as or much longer than the perigynia —...........-.-.-.-.----ce--see---0---- C. gigas Perigynia broadly oblong-ovate; scales much shorter than the werteyiiid 27s a ee Se aa C. curatorium Culms phyllopodic, brownish or reddish-brown at base. Perigynia obscurely triangular, obovoid, 2.5 mm. long, strongly PT i ae le dies HE ot eet ae alae ae han ee C. pseudoscirpoidea Perigynia flattish, oblong-oblanceolate, 4—4.5 mm. long, minutely puberulent, or sparingly pubescent on the angles....C. scabriuscula cell: blades ( Carex nova L. H. Bailey, a Rocky Mountain species, reported heretofore only in the states of Montana, Wyoming, Colorado, New Mexico, and Utah, has been found in two more western states, Idaho and Oregon. It was collected by Prof. Morton E. Peck (No. 18493) on July 13, 1934, at Ice Lake, Wallowa Mts., Oregon, and by Prof. J. W. Thompson (No. 13555), on July 28, 1936, at Devil’s Bedstead, Sawtooth Range, Idaho. It is to be expected in the Blue Mountains of south- eastern Washington. of 7 7 Carex tumulicola Mackenzie has been collected by J. T. Howell (No. 6380) at China Harbor, Santa Cruz Island, Cali- fornia. This is a considerable extension of its range as it has not been reported before from farther south than the vicinity of Monterey, some 200 miles away. This species grows along the coast northward more or less abundantly to just beyond the Oregon-Washington boundary, and then has not been found for another 200 miles, until it reappears in the San Juan Islands. q rf 7 A new record for California is that of Carex neurophora Mackenzie, collected by J. P. Tracy at Trinity Summit, Hum- boldt County, No. 14143. This species occurs in southern Oregon so its appearance in northern California was to be expected. Mr. Tracy also collected on Trinity Summit, Hum- boldt County, C. Jonesti L. H. Bailey and C. paucicostata Mackenzie, both of which represent considerable extensions of range, as these species have been found heretofore only in the Sierra Nevada. 16 LEAFLETS OF WESTERN BOTANY [ VOL. II, NO. I Carex atrosquama Mackenzie was collected at Lone Frank Creek, Okanogan County, Washington, east side of Rock Moun- tain, August 4, 1933, by Charles B. Fiker, No. 1412. In the editorial appendix of North American Flora, vol. 18, part 7, p. 470, the editor states that in notes left by Mackenzie, C. viri- dior Mackenzie (in Abrams, Ill. Fl. Pacif. St. 1:331) is regarded as a name applied to a mixture of C. atrata L. and C. atrosquama. We have not seen the type of C. wiridior Mackenzie, Eggleston. No. 3329, collected at Sheep Mountain, Okanogan County, Washington, but since the plant collected by Fiker is in the same general locality as that collected by Eggleston, and since C. atrata L. has not yet been found in Washington, it may be inferred that both sheets are C. atro- squama Mackenzie. DIPSACUS SYLVESTRIS Huds.’ This, the so-called Common Teasel of Europe, was reported by Katherine Brandegee in Zoe 2:383 (1892) as a naturalized plant of San Francisco, growing “behind and above the Presidio proper.” It was in- — cluded by Jepson in the two editions of the Flora of Western — Middle California, but it is omitted from the Manual of the Flowering Plants of California although the plant still flourishes at the station where it was first detected and whence it has not spread in the past forty years. A specimen of this species, col- lected by Alice Eastwood at Sisson, Shasta County, in 1912, provokes the thought that the species is perhaps more widely distributed in California than is realized and that some colonies passed as Fullers’ Teasel (D. fullonum L.*) may be the Common Teasel, since in gross aspect the two plants are not dissimilar. They can be distinguished by the bracts in the heads, those of the Fullers’ Teasel being rigid and recurved, those of the Common Teasel (which with us is the rarer of the two) being pliant and straight—John Thomas Howell. 1 This name and author are assumed for the Common Teasel although it appears in the Index Kewensis as D. sylvestris Mill. and is given by Hegi in Fl. Mitt. Eur. VI, 1:281 as D. silvester Huds. 2 Dipsacus fullonum( excluding var. sativus) of Linnaeus’ Species Plan- _tarum is not the Fullers’ Teasel as we know it, but is D. sylvestris of present use. But the combination Dipsacus fullonum, which was first made by Linnaeus, can be interpreted as the name-bringing synonym for the Fullers’ Teasel, which was called by Brunsfels and other pre-Linnean writers Carduus fullonum. Thus, it would appear that the name D. ful- lonum [Brunsfels] L. can be preserved for proper use for the Fullers’ Teasel, instead of D. sativus (L.) Honck.; and certainly its ‘preservation is eminently desirable.’’ (See Greene, Pitt. 3:1-9, for further discussion and for source of matter for this note.) Lh it ee ean a Se Vo . II No. 2 LEAFLETS of WESTERN BOTANY % CONTENTS PAGE A Russian Collection of Californian Plants. . . . . I7 JoHN Tuomas Howe. The Type Locality of Gilia congesta . 21 LINCOLN CONSTANCE Notes on the Distribution of Great Basin Plants—II . . 23 Bassett MAGUIRE OS Ae OR eR es a nn AL OR BIE [a oC Ba? fs A.ice Eastwoop SG OES eet CY See A OLN Ae Rtmin ty Ae ae AR J. W. STACEY This number published with funds from the California Botanical Club SAN FRANCISCO, CALIFORNIA Aprit 19, 1937 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. AU UU pli INCHES pvaguoypeenganganaytenguccapn agg aape ages eee Owned and published by Axice Eastwoop and JoHN THomas HowkELL i i ie te APR 66 1937 YORK :OTANICAL APRIL, 1937] A RUSSIAN COLLECTION 17 A RUSSIAN COLLECTION OF CALIFORNIAN PLANTS BY JOHN THOMAS HOWELL For a collection of herbarium specimens to encircle the earth in order to be named is a rare occurrence, and yet such a col- lection recently came to me for study. The much-travelled specimens would be noteworthy if only because of their ex- tended journeying; but this particular collection holds a special interest for us who live in California because the plants were collected by the Russians in 1840 and 1841 immediately before they abandoned their Californian colony not far north of San Francisco. Until recently the existence of this unnamed col- lection in the herbarium of the great Russian Academy of Sciences in Leningrad was not generally known, and indeed it would seem likely that the memory and knowledge of these early Californian specimens had passed even from the minds of Russian botanists. More recently, the bundles were re- discovered, and, together with other American collections, were sent for determination to Dr. Ivan M. Johnston at the Arnold Arboretum. Recognizing the particular interest that would be attached to this set of plants by Californians, Dr. Johnston inquired if I would like to see and name the collection. The opportunity to examine these early Californian specimens was eagerly and gratefully accepted; and so it happened that the specimens after encircling the earth came to be named less than 100 miles from where they were collected nearly 100 years ago and where their descendants still flourish on the maritime bluffs and mesas or in the interior hills and valleys.* An interesting chapter in Californian history is recalled by this collection.2 Early in the winter of 1841, after twenty-nine years of occupation, the Russian colonists left Ross on the Sonoma coast of California.* The attempt to establish a perma- 1 The specimens were originally sent to St. Petersburg via Alaska and Siberia. On their return to California, they crossed western Europe, the Atlantic Ocean, and North America. After the specimens were determined, all were returned to Leningrad except a few duplicates which are now in the Herbarium of the California Academy of Sciences. 2 For the historical data in this account, I am chiefly indebted to the very interesting number of the California Historical Society Quarterly which dealt with the Russians in California, vol. 12, no. 3, pp. 189-276, 1934. 8 Ross was known to the Spanish-Californians as Fort Ross, the name b whith it is generally known today. It should not be confused with Ross, arin County, a modern residential district that is situated between Mt. Tamalpais and San Rafael. Leafil. West. Bot., Vol. II, pp. 17-32, April 19, 1937. 18 LEAFLETS OF WESTERN BOTANY [ VOL. II, NO. 2 nent colony in California had failed because the enterprise had not proved profitable to the Russian American Company which was primarily interested in revenue from sea-otter and seal hunting on the Californian coast and in the development of agricultural activities. After the decision of the company in the spring of 1839 to dispose of its Californian holdings, a purchaser had been sought in the Mexican pueblos of Sonoma and Yerba Buena, but finally the property was sold to Capt. John A. Sutter of New Helvetia (Sacramento). In June, 1841, just a few months before Ross was finally evacuated and while negotiations for the disposal of the proper- ties were assuming form, one of the chief accomplishments of the Russian sojourn was realized, the ascent and naming of Mt. St. Helena, the most conspicuous landmark in the Coast Ranges immediately north of San Francisco Bay. It seems probable that the one who initiated this undertaking and was responsible for its success was E. Voznesenski,* scientist- naturalist from the zoological museum of the Russian Academy in St. Petersburg. Voznesenski, 24 years of age when he came to California in 1840, was a trained entomologist and his col- lections of insects from various parts of central California have become classical; but only now, after all these years, is it known to us that he also collected plants, the same plants that came back to us in California to be identified. The collection contained 346 specimens representing 214 species and varieties. The specimens were collected in 1840 and’ 1841, in general a fairly representative collection of specimens from the flora of present-day Sonoma County: a number obvi- ously from the coastal hills and mesas in the immediate vicinity of Ross, and others from the interior, some certainly from Mt. St. Helena. But whether the specimens originated on the coast or in the interior, most of the labels accompanying the specimens carry only the printed data, “California boreal. Ross.—leg. Wossnesensky,” and on only a few appear more definite designations of locality in script. Thus on a rare and unusual form of Eriogonum vimineum is the notation “FI. 4 The orthography of this name varies in the literature on the Russian colony in California. The variant adopted here is the one accompanying the scientist’s portrait, facing page 120 in the publication of the Russian Academy of Sciences, “The Pacific, Russian Scientific Investigations,” Leningrad, 1926. APRIL, 1937] A RUSSIAN COLLECTION 19 Slavjana,” that is, the Slav or Russian River; a specimen of Anaphalis margaritacea is labelled “Bodega”; and on yet an- other, Erodium cicutarium, is the note “Rio San Ignacio.” One of the most interesting of these special designations is ‘“‘m. St. Helenz et desertum St. Rose,” the mountain of St. Helen and the desert of St. Rose. In June, 1841, in the heat of early summer, as Voznesenski and his party traversed the hot and desiccated country beyond Santa Rosa in his ascent of Mt. St. Helena, well might he have called it a desert. Those inhabi- tants of Sonoma County who now live between Santa Rosa and Mt. St. Helena would probably not be highly edified to hear their home district called the Santa Rosa Desert, but one can well imagine the feelings of Voznesenski and his companions as they traversed the brushy hills and mountains. The hard- ships endured and the difficulties surmounted can be vividly imagined by those who are acquainted with the region, and the label on these plants tells in one word, desertum, that the first ascent of Mt. St. Helena was not easily accomplished. It is to be regretted that no account by any member of the party who made the historic ascent is known. Among the specimens that carry this thought-provoking label are plants from the chaparral of the interior hills and mountains such as Adenostoma fascicu- latum, Pickeringia montana, and Eriodictyon californicum, as well as such herbs as Calochortus amabilis, Silene californica, Hypericum concinnum, Antirrhinum vagans, Antirrhinum virga, and others. Regarding the data on the labels, it only remains to be noted that while most of the specimens are credited to “Wossne- sensky,” eight were collected by Kuprianov, governor of Russian America at the time Ross was abandoned. Some of the speci- mens are also accompanied by notations of specific date and locality in Russian script on small bits of paper, evidently clipped from the original papers in which the specimens were prepared. Among the most interesting plants represented in the col- lection are the weeds. Although the weeds are not numerous, the few that there are furnish definite evidence of the occur- rence of certain species even at so early a date. Two species of filaree which are now among the most abundant plants in the lowlands of California are in the collection, Erodium cicutarium and E. moschatum. Poa annua, Silene gallica, Raphanus sativus, Malva sp. (seedling), Anagallis arvensis, and Physalis ixocarpa 20 LEAFLETS OF WESTERN BOTANY [vou. II, NO. 2 are other widely naturalized weeds of today which are included in the collection. Other plants in the collection definitely of garden origin which have never become naturalized are Pisum sativum, Ruta graveolens, Althea rosea, Borago officinalis, and Lycopersicum esculentum. And it is of special interest to note those plants which are sometimes regarded as doubtfully indige- nous to California: Arabis glabra, Fragaria chiloensis, Geranium carolinianum, Oxalis corniculata, Apocynum cannabinum, Pru- nella vulgaris, and Solanum nigrum. While there is some botanical interest in the weeds of the collection, no extraordinary scientific interest is to be attached to the native species represented, most of which are not uncommon and can still be found in the hills and valleys of Sonoma County. But in passing it is interesting to note that, whereas the Russian collectors neglected to obtain a specimen of the glory of the northern California Coast Ranges, the Redwood, they did find several plants relatively rare, such as Trifolium ameenum Greene and Antirrhinum virga Gray, as well as the most robust speci- men of Orthocarpus pusillus known to me, a plant more than 6 inches tall. But by and large, this set of plants offers a striking and impressive instance of what value accrues to a collection which is thoroughly studied and reported in botanical litera- ture, and what happens if such studies are not made. Many plants in the collection were unknown to science when they were collected and if by a bit of diligent research they had been sought out and named, the collection would now rank with those obtained by Nuttall, Douglas, and Hartweg, and by the earlier Russian collectors, Langsdorff, Chamisso, and Eschscholtz. As it is today, Voznesenski’s Californian col- lection is just another set of Sonoma County plants as far as scientific value is concerned; and what might have proved a classical collection of Californian plants is noted here as some- thing of merely botanico-historical interest. If things botanical had been done differently in old St. Petersburg, I would not have had the interesting experience of working over these plants in San Francisco so near to their original home; and instead of going to the Gray Herbarium or to Kew to study rare and precious types of certain species, Californian botanists would be journeying to Leningrad to learn what certain types collected by Voznesenski are really like! APRIL, 1937] GILIA CONGESTA 21 THE TYPE LOCALITY OF GILIA CONGESTA BY LINCOLN CONSTANCE Washington State College, Pullman, Washington The question of the type locality of Gilia congesta Hooker was left unsettled in a recent revision of this species and its allies by Mr. Reed C. Rollins and myself (Amer. Journ. Bot. 23 :443-440,—1936). We pointed out that the original col- lection was made by Douglas, on the “Sandy plains of the Columbia,” presumably in 1829 (although the collector was not even in North America at that date) and that the only com- parable subsequent collection was that by the Wilkes Expedition, from the “Upper Columbia.” Gilia congesta has been included in the flora of Washington solely upon the basis of these two early records. Rydberg judged as typical the local variant of the Black Hills of South Dakota and neighboring Wyoming (our var. pseudotypica), thus, at least by implication, moving the type locality across the Rocky Mountains. We concluded, rather lamely, that: “Not all of the Columbia drainage has been thoroughly botanized . . . and it is possible that this plant may have escaped detection in recent years.” A set of Gilia congesta recently sent me for annotation by Mr. J. William Thompson of Seattle, premier contemporary collector of the Pacific Northwest, contained a sheet of the species collected by Sister Mary Milburge, in Spokane County. Mr. Rollins compared this with an isotype of G. congesta at the Gray Herbarium, and reported an excellent match. A request to the collector for further information brought a generous response: a duplicate specimen, and a map showing the site of collection. The specimen is her No. 874, collected May 5, 1934, on a dry gravelly hillside, 15 to 25 miles northwest of Spokane, on the Spokane River, between the mouth of the Little Spokane River and Nine Mile. Inasmuch as Douglas, according to his Journal, crossed and recrossed the Spokane River in the years 1826 and 1827, and even ascended some nine miles above its mouth, we feel there need no longer be serious question as to the approximate type locality. 22 LEAFLETS OF WESTERN BOTANY [VoL. II, NO. 2 The same paper set the northern limit of Gilia congesta var. montana (Nelson & Kennedy) Constance & Rollins at Crater Lake and Mount Thielson, Oregon. An examination of speci- mens in the University of Oregon Herbarium, at Eugene, revealed a collection made in “scoria, top of South Rim of New- berry Crater, Paulina Mountains, July 7, 1928, L. E. Detling No. 6.” This extends the range of the variety some thirty-five miles northeast, into Deschutes County, and further corrobo- rates the suggestion that several species, known only from Crater Lake or thought to range only thus far north, are likely to occur, also, in the Paulina Mountains. cf if 7. Typical Gilia congesta has also been collected, “above 8000 feet, Broken Top, east of the South Sister (Cascade Moun- tains), Oregon, August 4, 1931, Ed. Easton” (L. F. Henderson No. 14169) (Univ. Oreg.). This, again, represents a consider- able extension in range along the summit of the Cascades into Lane-Deschutes counties. Although a specimen of Gulia congesta from the Wallowa Mountains of northeastern Oregon was in the herbarium of the State College of Washington at the time, we excluded it from our paper, on the suspicion that the data might be erroneous. The discovery of two other specimens from the same area verifies the authenticity of this occurrence, and the following citations should be added: OreEcon. Wallowa Co.: on the Imnaha near the source, 7000 ft. alt., Wallowa Mountains, August, 1906, W. C. Cusick No. 3083 (Univ. Oreg.); Wallowa Mountains, 8000 ft. alt., August, 1909, Cusick (Univ. Oreg.); near the sources of the Imnaha River, high Wallowa Mountains, August 12, 1911, Cusick (Wash. State Coll.). The Broken Top and Wallowa plants occupy an inter- mediate point on the line of morphological development between typical Gilia congesta and var. montana, and would probably fall into that unnatural population segregated as subsp. palmifrons (G. palmifrons Rydb.) by Brand. APRIL, 1937] GREAT BASIN PLANTS 23 NOTES ON THE DISTRIBUTION? OF GREAT BASIN PLANTS—II? BY BASSETT MAGUIRE® This series of notes‘ is designed to make better known the occurrence and distribution of hitherto unrecorded and rare plants of the Great Basin. *EQUISETUM VARIEGATUM Schleich.5 13167°®: Iron Co., Utah. Common in wet meadow parks in spruce at 10,000 ft. Dixie National Forest in the vicinity of Cedar Breaks. * SORGHUM HALEPENSE (L.) Pers. 13264: Washington Co., Utah, Aug. 3, 1934. Becoming a common weed in Washington Co.; also collected by Valgene Lehman in the Bear River marshes, Box Elder Co., Utah. *SALIX MONTICOLA Bebb.”? 1719: June 28, 1932; 1720: June 26, 1932; San Juan Co., Abajo Mts.; 1722: San Juan Co., La Sal Mts., July 5, 1935. Somewhat common at and above 8000 ft. These collections, according to Dr. C. R. Ball, extend the range of this species considerably westward from the Rocky Mts. *CLAYTONIA VIRGINICA L. 4793: Washington Co., Utah, April 1, 1934. Under thickets, moist canyon slopes, Zion Na- tional Park. A robust form frequently producing three stem- leaves. RANUNCULUS TRICHOPHYLLUS Chaix var. HISPIDULUS (E. R. Drew) W. B. Drew. (R. Grayanus Freyn.) 1812: San Juan Co., Utah, July 5, 1932. Vernal pool in Quercus Gambellu association. South trail to La Sal Pass, 8000 ft. 15369: Cache Co., Utah, May 14, 1934. Growing abundantly in Dry Lake, a large vernal pond at 6000 ft. These two stations apparently double those previously known from Utah. Tidestrom® reports 1 The use of the adjective ‘Distributional’ in the title of the first iron te ceueuintle emer tr hs ced ata 2 Contribution from Department of Botany, Utah Agricultural Experi- ment Station. Publication authorized by Director. 8 In charge of the herbarium. 4 Maguire, B. Distributional Notes on Plants of the Great Basin Region—I. Leafl. West. Bot. 1:185-188 (1935). Ne jel Lag asterisk indicates records which seem to be new for Utah or TB numbers following the plant name are the field numbers of the 7 Identified by Dr. C. R. Ball. 8 Tidestrom, Ivar, Flora of Utah and Nevada, 214 (1925). 24 LEAFLETS OF WESTERN BOTANY [voL, II, NO. 2 Batrachium Grayanum from Fish Lake, Sevier Co., and Drew® cites E. B. & L. B. Payson No. 4848 from Summit Co., near the west fork of Bear River. *BERBERIS FENDLERI Gray. 5904: San Juan Co., Utah, June 27, 1933. Lateral canyons to San Juan River, vicinity of Bluff. To be expected elsewhere in southeastern Utah since this plant is very common in adjacent Colorado along the Dolores and San Miguel rivers which join the Colorado River in Grand Co., Utah. *CRATZGUS CHRYSOCARPA Ashe. 5184: Cache Co., Utah, August 27, 1934. Growing sparingly along the stream bank, Blacksmith Forks Canyon, 5500 ft. * PRUNUS DEMISSA (Nutt.) Walp. 5185: Cache Co., Utah, August 23, 1934. Uncommon on slopes, Logan Canyon, 6000 ft. A southern extension of its known range into Utah. *Hipiscus Trionum L. 212: Davis Co., Utah, 1931. A recent weed introduction occurring in cultivated fields. This plant has been sent in from several localities in the state, and may be expected to become a generally distributed weed. * FRAXINUS MACROPETALA Eastw. 4902, 4903, 4904: Clark Co., Nevada, April 6, 1934. About small swamps eight miles north of Glendale. Known hitherto only from northwestern Arizona. *SOLANUM ROSTRATUM Dunal. 5688: San Juan Co., Utah. Occurring as a weed in fields and along canal banks in Bluff, Utah. A specimen of this weed has been sent to the writer from Salt Lake Co. OENOTHERA BREVIPES Gray var. TYPICA Munz. 4893: Mohave Co., Arizona, April 3, 1934. Common in sand of Beaver Dam Wash, Arizona Strip. DopocATHEON TETRANDRUM Suksdorf. 4219, 4220, 4222: Summit Co., Utah, August 18-19, 1933; 4221: Duchesne Co., Utah, August 19, 1933; all collected from the high Uinta Mts. in the vicinity of Mirror Lake and the Still Water Basin. These collections were distributed as D. pauciflorum (Durand) Greene. Subsequent to Goodman’s?* paper calling attention to the occur- rence of the tetramerous species in the Uinta Mts., the writer 9 Drew, W. B. Rhodora, 38:29 (1936). 10 Goodman, Geo. J. Notes on the Distribution of Some Rocky Moun- tain Plants. Ann. Mo. Bot. 18:284 1931. APRIL, 1937] GREAT BASIN PLANTS 25 again examined his material and discovered his error in identifi- cation. In addition to the above four collections of this species, the writer has before him two collections of the same plant from this general region collected by Dr. B, L. Richards and a recent collection of his own from Henry’s Fork Basin. Since the writer has seen no specimens of D. pauciflorum from the higher Uinta Mts., it would seem that here D. tetrandrum occurs, probably to the exclusion of the generally common D. pauctflorum. *MIMULUS PRIMULOIDES Benth. 13124: Washington Co., Utah. Moist place along wash in meadow. Pine Valley Mts., 10,000 ft. 14323: Summit Co., Utah, 10,800 ft. Abundant in grass-sedge and vaccinium bogs about Grass Lake. Frequent in similar situations in Henry’s Fork Basin, Uinta Mts. Outside of the general Pacific States range of this attractive little plant, the records given above are the only collections known to the writer that have been made since 1916, and con- nect by great jumps the Idaho stations of Macbride*? and Macbride and Payson*™ with that of Goodding™ in southern Arizona. This erratic distribution would lead to the belief that this species of Mimulus will be found, through more thorough field work, to have a more general distribution. Rusia TINcToRUM L. 4482: Washington Co., Utah. Locally well established. Developing vigorously in slopes of Ash Canyon, 2 miles northwest of La Verkin. APLOPAPPUS SUFFRUTICOSUS Nutt. *subsp. Typicus Hall. 13389: Iron Co., Utah, August 5, 1934. Occurring frequently under Pinus aristata in stony sandy soil in the vicinity of Mid- way Summit, 6 miles west of Cedar Breaks at 10,000 ft., Dixie National Forest. 13386: Washington Co., Utah, August 1, 1934. With Arctostaphylos pungens under Pinus ponderosa; on rocks. North slopes of Pine Valley Mts., 10,000 ft. * ARTEMISIA NORVEGICA Fries var. SAXATILIS (Besser) Jep- son. 14394, 14523, 14594: Summit Co., Utah. All collections made in upper basin of the Henry’s Fork of the Green River in the vicinity of Mt. Gilbert and King’s Peaks, at an altitude of 10,800 to 11,200 ft. This plant replaces A. scopulorum in deeper, richer, and better drained soil in more protected sites ; 11 Grant, A. L. A Monograph of the Genus Mimulus. Ann. Mo. Bot. 11:244 1924, 26 LEAFLETS OF WESTERN BOTANY [VoL. II, NO. 2 the latter species occurring commonly on stony exposed ridges and at higher elevations, up to 12,000 ft. CENTAUREA Picris Pall. 5624: Grand Co.; 5625: San Juan Co., Utah. The plant is listed by the common name, Russian Napweed, as one of the noxious weeds of Utah. It is cited by Tidestrom™ from the vicinity of Salt Lake. At the present time this troublesome weed is to be found in waste places along fence rows and in pastures over most of the state. *FLAVERIA CAMPESTRIS J. R. Johnston. This weed, ap- parently new to Utah, was collected in low pastures in the vicinity of Moab, Grand Co., September 23, 1935. * XIMENESIA EXAURICULATA (Robins. & Greenm.) Rydb. 14965: Grand Co., Utah, August 13, 1934. Occurring sparingly as a roadside weed in vicinity of Moab, 6000 ft. Common in western Colorado. To be expected to spread rapidly as a road- side and waste place weed in Utah. POLYGONUM ARGYROCOLEON IN Arizona. Among many plants collected in southern Arizona, in the autumn of 1936, by Sister Noel and Sister Teresita (O. S. F.), one Polygonum appeared to be a “stranger” to our Flora. Mr. Howell of the California Academy of Sciences had given an account of Poly- gonum argyrocoleon Steud. two years before (Leafl. West. Bot. 1: 142-144,—1934), and upon examining our specimen and comparing it with material in the United States National Her- barium, it proved to belong to that species. It compared well with a specimen collected by Kotschy “in arena insularum Tigridis pr. Mossul” and named by Steudel. This material in the United States National Herbarium evidently belongs to the original collection. Mr. Howell gave us the range of this “emigrant” in California, and to this we are now adding the Arizona locality—Ivar Tidestrom, Langlois Herbarium, Catholic University of America, Washington, D. C. POLYGONUM ARGYROCOLEON IN CALIFORNIA. In the summer of 1935, Mr. Chester Dudley collected the Silver-sheathed Knot- weed on San Marcos Creek, an affluent of the Salinas River near San Miguel, San Luis Obispo County. This is a Califor- nian locality not heretofore noted.—J. T. Howell. 12 Tidestrom, Ivar. Flora of Utah and Nevada, 621 (1925). . —————e APRIL, 1937] QUEST FOR LILIES 27 QUEST FOR LILIES BY ALICE EASTWOOD The first lily encountered in our quest in 1936 was the coast lily (Lilium maritimum Kell.). It was abundant and in fine bloom the latter part of May in that fascinating area near Mendocino City known as the White Plains of Mendocino. Here Pinus contorta, Pinus muricata and Cupressus pygmea begin to fruit when only a few inches from the ground, never becoming large trees. The lovely rhodendrons were in bloom and the squaw-grass (Xerophyllum tenax) everywhere lifted its creamy white plumes from amid its grass-like leaves. Arcto- staphylos setosissima, the tall erect hairy-stemmed manzanita, was fruiting abundantly, and the low, glossy-leaved manzanita (Schizococcus nummularius) still retained a few flowers. Plants of the latter with erect stems grow side by side with those that are prostrate. Bunches of white flowers of Labrador tea, Ledum columbianum, were conspicuous amid the other shrubs. Lilium maritimum was described from plants found in wet places in the vicinity of San Francisco from which it has long since disappeared. About forty years ago I saw it growing in the Crystal Springs Lakes region and have never seen it since nearer than Mendocino County. While each plant usually has but one flower hanging from an erect stem, sometimes a plant with six or seven flowers may be found. The stamens and style are included, the reddish bells are purple-spotted within, and the segments of the perianth recurve but little at the tips. While in Oregon and Washington during June, we saw the typical Columbian lily (L. columbianum Hanson) in many places but nowhere abundant. Along the road to Mt. Hood from Portland it must have been protected since on that much- travelled highway it grew on both sides of the road. What a pleasure it was and what an asset to the road! All that we saw had yellow flowers, red-dotted, with stamens and pistil con- spicuously exserted. The segments of the perianth fold back from about the middle to the base of the flower. The number of flowers on a stem varies, all are pendent from ascending stems. We were especially desirous of finding L. Bolanderi Wats. and L. Kelloggii Purdy, so on our way home early in July we 28 LEAFLETS OF WESTERN BOTANY [VOL. 11, NO. 2 left the main coast highway near Smith River, Del Norte County, California. Our goal was the summit between Patrick and Shelley creeks, Del Norte County, California, not far from the Oregon boundary. Many years ago I had found both lilies there and in bloom. Luckily we arrived at the blooming season and as I had expected, found them in the vicinity of my former collection. Our specimens of L. Bolanderi agree so exactly with the original description by Sereno Watson, Proc. Am. Acad. 20: 377* that they might be taken as the type. This red bell-shaped lily is allied to L. maritimum and is very unlike L. Kelloggi, which is related to L. rubescens and was described in Purdy’s 1904 catalog, page 19, of which we have a copy. The Index Kewensis gives the date as 1905. Purdy’s description is as follows. “Lilium Kelloggii. A new lily discovered by myself in northern California, 3—4 ft. high, slender, 3—15 flowers. The flowers have closely revolute petals of a pinkish color, finely dotted purple, fragrance peculiar and very delicate.” The lilies we found did not have as many as 15 flowers, but in all other characteristics agree with the description. It seems incompre- hensible that anyone could consider L. Bolanderi and L. Kel- loggii the same, though Ivan M. Johnston states in Contr. Gray Herb. new series, no. 68, page 81, that three collections were on the type sheet of L. Bolanderi: one collected by Bolander in the Red Hills of Humboldt County, one by Rattan near Arcata, Humboldt County, and one by Thos. Howell near the Oregon boundary in Del Norte County. This last was the lily described by Watson. Though Bolander’s was named first and considered the type by Johnston, it must have been an almost unidentifiable specimen for Watson to include it with the plant which he really described and which should bear the name L. Bolanderi or descriptions are of no value. Both of these lilies may eventually be found again in the region of Bolander’s collecting in what is known as Red Moun- tain, Mendocino County, and Red Hills, Humboldt County. _ ® Bulb ovate, of numerous lanceolate scales 1 to 1% inches long: stems % to 3 feet high, 1—2-flowered: leaves mostly verticillate and approxi- mate, oblanceolate, acute, glaucous beneath, 1 to 2% inches long: flowers horizontal or somewhat nodding, “dingy purple” (Rattan) or “dark brownish red” (Howell) becoming somewhat paler, spotted, the segments 1% to 1% inches long, but slightly spreading, rarely at all recurved: enthers 2 or 3 lines long, the ovary and style 9 or 10 linés jong. =e APRIL, 1937] QUEST FOR LILIES 29 The region has been very little visited by botanists. In one day in 1905 I was able to collect three species, of which these hills are the type locality. Garrya buxifolia and Narthecium cali- fornicum are both quite common near the Del Norte-Oregon boundary. Eriogonum Kelloggii was the other plant, and a new Gormania, named G. Eastwoode Britt., was also collected. In Mendocino County, this mountainous area extends from Bell Springs on the old road north into Humboldt County. There are no roads and the trails are confusing. Along Smith River between Patrick Creek and Gasquet we encountered L. occidentale Purdy, or what seems to be that species. The bulb is shortly rhizomatous. The tall stems bore many flowers on long horizontally spreading pedicels, the longest 2 dm. long. In this habit, they resembled L. Humboldtii. On the lower part of the stout stem the leaves are scattered, about the middle they become densely whorled, some whorls with 15 leaves. The leaves are long, erect, linear, glossy, the longest 2 dm. or more long and all only 5 mm. wide. Some younger plants were seen with one or two flowers. They grew in wet places. One other lily seen was what we thought to be the red- flowered variety of the Columbian Lily. This was found at two localities near the coast: the first on the Oregon Coast Highway near Ocean Lake, Lincoln County; the second on the California Redwood Highway at Patrick’s Point, Humboldt County. And so during the early summer of 1936, in the course of rambles extending from northern California to Washington, five of our lovely western lilies were seen and studied as they occur in the wild. NoLinA Parry! IN VENTURA CouNTy, CALIFORNIA. An extension of the range of Nolina Parryi Wats. to the northward is represented in the Herbarium of the California Academy of Sciences by a specimen in bud, collected April 12, 1916, by the writer. It grew along the road leading up Santa Ana Creek, a tributary of the Ventura River.—Alice Eastwood. 30 LEAFLETS OF WESTERN BOTANY [voL. II, NO. 2 NOTES ON CAREX—IX BY J. W. STACEY A rather strange confusion has occurred in the Carices of the section Macrocephale. Mackenzie (N. Amer. FI. 18:83,— 1931) gives only one species, C. macrocephala Willd., stating that the North American plant was the same as the Asiatic one. Fernald (Rhodora 32:9-11,—1930) treated the North Ameri- can plant as C. anthericoides Presl and the Asiatic one as C. macrocephala Willd., pointing out certain very obvious differ- ences. The writer (Leafl. West. Bot. 1:210,—1936) agreed with Fernald, chiefly on account of the fact that the so-called C. macrocephala Willd., introduced into New Jersey, was greatly different from the plant growing in northwestern America along the coast. It remained for a Japanese botanist, Jisaburo Ohwi, of the Kyoto Imperial University, to solve the problem (Mem. Coll. Sci. Kyoto Imp. Univ. ser. B, 5: 281,—1930; 11:249,—1936). He found that there were two species along the coast of Japan, C. macrocephala Willd., and another, which was named C. Kobo- mugi Ohwi. Immediately on receipt of his descriptions, we wrote to Prof. Ohwi, and he very kindly sent us sheets of both species, and wrote that C. macrocephala is distributed from Hokkaido (Yezo) northward, while C. Kobomugi grows in the southwestern coast of Hokkaido (Yezo), Korea, Hondo and southward. The sheets from the southern part of his territory exactly matched the introduced plants in New Jersey, while the © sheets from the northern part of his territory matched the plants growing in northwestern America. Mackenzie evidently had never seen the introduced plants from New Jersey, but probably had seen plants collected in the northern part of the Asiatic territory, but none from the southern part. Fernald had seen the plants from New Jersey, and was entirely right in differentiating them from our Pacific Coast plants, clearly pointing out all of the essential differences. The result is that our western plants retain the name of C. macrocephala Willd., the plants introduced into New Jersey become C. Kobomugi Ohwi, and C. anthericoides Pres] becomes a synonym of C. macrocephala Willd. ll = te i te Sa APRIL, 1937] NOTES ON CAREX 31 Mackenzie in December, 1931, described a new species from Mexico, Carex autumnalis. Jisaburo Ohwi in July, 1930, used this name, C. autumnalis, for a Japanese plant. A new name is here proposed for the Mexican species based on the name of the town at which the type was collected, Flor de Maria, State of Mexico, Mexico. Carex marianensis Stacey, nom. nov. Carex autumnalis Mackenzie, N. Amer. Fl. 18:66 (1931); not C. autumnalis Ohwi, Mem. Coll. Sci. Kyoto Imper. Univ. ser. B, 5:251 (1930). “C. densa L. H. Bailey” Kikenth. in Englar, Pflanzen- reich IV. 20: 167 (1909) as to Mexican specimens. 7 ri 7 Three new records for the State of Washington follow: Carex multicostata Mackenzie, collected at Indian Corral Spring, Columbia County, by H. T. Darlington, No. 65, in the herbarium of the State College of Washington, Pullman, Wash- ington. Heretofore this species has not been detected from farther north than the Wallowa Mts., in Oregon. Carex tenera Dewey, collected at Tucanon River, Columbia County, by H. T. Darlington, No. 103, also in the herbarium of the State College of Washington. This sheet was labeled C. tenereformis Mackenzie, and then annotated C. microptera Mackenzie. Prior to this collection the nearest known station was that of Big Fork, Montana. Carex Vahlii Schkuhr,* collected at Mutton Creek, northwest of Salmon Meadows, Okanogan County, by Charles B. Fiker, No. 352, in the herbarium of J. William Thompson of Seattle. THE Dyer’s Woap IN CALIFoRNIA. Leaving the main high- way at Yreka, Siskiyou County, an inviting road winds in a southwesterly direction over a high ridge to the quaint towns of Fort Jones and Etna. As one approaches Fort Jones, which occupies a central point in beautiful Scott Valley, large open fields are traversed and there on May 20, 1936, it was our pleasure to become acquainted with /satis tinctoria L. in glorious association with Lupinus albifrons Benth. While not on so vast a scale as the blue and gold Arvin fields of Kern County, the yellow and blue mass effect was equally lovely. * According to V. Kreczetowicz (Fl. U. R. S. S. 3:183.—1935), this name is antedated by Carex norvegica Retz. (Fl. Scand. Prodr. 179,—1779). As Carex norvegica Willd. is consequently invalidated, it is renamed Carex all V. Krecz. If this is correct, O. Vahliti Schkuhr becomes C. nor- vegica Retz. 32 LEAFLETS OF WESTERN BOTANY [ VOL. II, NO, 2 Isatis tinctoria is a European member of the Mustard Family and it would be interesting to know how it became naturalized in this remote Siskiyou valley, where from the farmers’ view- point it is a troublesome weed and when cut down persistently springs up again from the root. It is commonly called Dyer’s Woad and according to L. H. Bailey, “Cesar relates that the ancient Britons used the woad for staining their bodies and the word Britain itself comes from an old Celtic word meaning painted. Before indigo became common in Europe, the Dyer’s Woad produced the chief blue coloring matter for woolen cloth. The introduction of indigo in the seventeenth century destroyed this important industry, not without opposition.” We should be thankful, however, that Jsatis tinctoria has not been de- stroyed, despite opposition, for even if it is a weed, it can be a thing of beauty —E. Dales Cantelow. After sending a packet of buhach to Dr. Gustavus A. Eisen, eminent Swedish-American scientist now living in New York City, the following note was received regarding this insectifugal powder. : “This morning came the can with buhach for which I thank you most sincerely... . I have always been interested in the buhach. I knew intimately Mr. Mico, a Dalmatian, who brought over the seed. I saw the plantation at Merced from the very beginning and I grew some myself at Fresno.* Mr. Mico invented the name which he patented ; it was from the Slavonian word buhah, meaning flea. A teaspoon of the powder, if touched with a match, will glow and in five minutes every mosquito and fly in the room is dead. During my trips to Mexico and Central America, I always carried with me a large supply of buhach. In those countries you can never go to a theater unless you sprinkle your stockings inside and out with buhach. The fleas would kill you. In Guatemala every person, I was told and I found it to be true, had never less than seven fleas on him... .” * The plantation of which Dr. Hisen writes was probably not at Merced but was undoubtedly on what was known as the Buhach Ranch, about five miles northwest of Merced on the line of the Southern Pacific Railroad. Although the ranch has now been subdivided, the district is known as “the Buhach,’”’ and on the U. S. G. S. quadrangle ‘‘Atwater,’’ the place name Buhach is to be found on the railroad about 1% miles southeast of Atwater. The plant from which buhach is obtained is Chrysanthemum cinerariefolium Vis.—J. T. Howell. Vot. Il No. 3 LEAFLETS of WESTERN BOTANY ¥ CONTENTS PAGE A Provisional Key to the Species of Downingia Known in California BP SH aah la} Sa, SM) RY A a nad Rosert F. Hoover : Meropecies Of Dodecatheon ....° 0. 6.) ee 86 ALice Eastwoop ¥ 7 NI TEED. CER ae eRe aS PORN SPERTES RB LK J. W. Stacey ‘ Zygadenus fontanus, a New Species from Mt. Tamalpais . 41 i Auice Eastwoop . nnmNatatinan Piirits | 0 i gr a oe a ene Joun THomas Howe. Further Studies in Eriogonum—I.. . .-. . . . 45 SusAN G. STOKES SAN FRANCISCO, CALIFORNIA Aucust 26, 1937 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. WEstT. Bort. PAUL) BSUS RU Ue a IU A eee ee ee ee ee UE MAT MU HC EGR SUA! 011° Owned and published by Axice Eastwoop and JOHN THomMaAs HOWELL SOTANICAL AUGUST, 1937] KEY TO SPECIES OF DOWNINGIA “~~? | 33 A PROVISIONAL KEY TO THE SPECIES OF DOWNINGIA KNOWN IN CALIFORNIA BY ROBERT F, HOOVER Most of the Californian species of Downingia were first recognized and named as natural entities by Greene, but it was not until Jepson published his revision of the Californian species in 1923 (Madrofio 1:98—102) that they were distinguished by the use of adequate key-characters. The key to the species in that revision is quite satisfactory for the species included, but since that time a number of additional species have been made known from California, including two described by the present writer. The following key is intended to place these species according to their distinctive characters and apparent relation- ships. The writer wishes to express his gratitude to Professor Jepson for helpful criticisms and suggestions in connection with this key. It was the writer’s intention to include all species of the genus, but in many instances no specimens or even adequate descriptions of the extra-Californian species are available, so that the larger project is for the present impractical. It is much to be desired that all collectors of these plants study the fresh flowers, making notes in regard to the sort of characters used in this key. Certain specimens from various localities in Cali- fornia appear to represent undescribed species, but the distinctive characters are obscured in drying. It is notable that species which are closely related to one another occupy distinct geographical areas. Thus D. pulchella occurs in the Coast Ranges and Sacramento Valley northward to Oregon, D. immaculata in Southern California, and D. pallida in the Sierra foothills. Dowmningia concolor is mainly a Coast Range plant, occurring locally in Southern California and on the west side of the lower Sacramento Valley, while the appar- ently related D. bella is restricted to the San Joaquin Valley. Downingia ornatissima, D. montana, and D. mirabilis form a closely related group, but each species has developed in a different area: D. montana mainly at middle altitudes in the Sierra Nevada, reaching the upper Sacramento Valley in Shasta County; D. ornatissima in the Sacramento Valley from Butte Leafl. West. Bot., Vol. II, pp. 33-48, August 26, 1937. 34 LEAFLETS OF WESTERN BOTANY [ VOL. I, NO. 3 County southward to the lower San Joaquin Valley; and D. mirabilis mainly to the south of D. ornatissima. Downingia humilis is confined to central California from near the coast to the border of the Sierra foothills, while D. leta is known only east of the Sierra Nevada. Such geographical distinctions are often useful in combination with morphologic characters. CoROLLA MUCH EXCEEDING CALYX Lower lip not forming a sharp angle with corolla-tube; anther-tube PECHEV EU 2A et est uke aed eerie aco D. elegans (Lindl.) Torr. Lower lip forming a sharp angle with corolla-tube; anther-tube not recurved (except rarely in D. ornatissima). Sinuses between upper and lower corolla-lips not extending below plane of lower lip. Calyx-lobes shorter than corolla-tube; leaves less than 5 mm. Depress eae re eee a 2 aie a D. cuspidata Greene Calyx-lobes equalling corolla-tube; leaves mostly more than 5 mm. long. Calyx-lobes rotate in flower and fruit; lower corolla-lip with 3 purple spots at base; anther-tube long- EXSEREEC aes conan eet D. pulchella (Lindl.) Torr. Calyx-lobes ascending, at least in flower; lower corolla-lip without purple spots at base; anther-tube not or scarcely exserted. Corolla about 1.5—2 cm. broad, bright blue; upper lip with large divergent lobes, equalling lower iF coh taeda D. immaculata Munz & Johnston Corolla less than 1.5 cm. broad, very pale blue; upper lip with subparallel lobes, only half as long as jo) gal bp eae maa tT Oe D. pallida Hoover Sinuses between upper and lower corolla-lips extending more or less below plane of lower lip. Sinuses straight and parallel to axis of tube, extending a short distance below plane of lower lip. Stems spreading, stout and fistulous; corolla deep blue, lower lip with central yellow spot......D. bella Hoover Stems not fistulous, stiffly erect or, if spreading, very slender ; corolla “light blue,” lower lip with central purple spot but no yellow.............. D. concolor Greene Sinuses curved, extending well below plane of lower lip. Calyx-lobes usually rotate; sinuses between corolla-lips oblique to axis of tube, curved under base of lower lip; projections at base of lower lip dark purple, nipple-like; pair of bristles at apex of anther-tube BU ISLER LOR EUNEE ihc cscsamtepceeesct ts D., bicornuta Gray AUGUST, 1937] KEY TO SPECIES OF DOWNINGIA 35 Calyx-lobes ascending; sinuses between corolla-lips curved first toward lower lip, then at end toward upper lip; projections at base of lower lip not dark purple; lobes of upper lip linear; pair of bristles at apex of anther-tube when present not twisted together. Leaves narrowly linear, strictly entire; calyx-lobes unequal; lobes of upper corolla-lip parallel, extending in same plane as lower lip................ gh Yc Dh, thle te nhac ERs atild D. montana Greene Leaves mostly oblong, the larger dentate to crenate; calyx-lobes nearly equal; lobes of upper corolla-lip oppositely divergent, or if parallel appressed to each other by the flat surfaces. Lower corolla-lip concave, the projections at base nipple-like; corolla-tube without a fold at base of sinus between lobes of upper lip; tips of lobes of upper lip not rolled.............. NEE Sh Ne Ee D. mirabilis J. T. Howell Lower corolla-lip plane, the projections at base not nipple-like; corolla-tube with a project- ing fold at base of sinus between lobes of upper lip; tips of lobes of upper lip each rolled into a ring or sometimes simply PeCUIVEE ob gn giet D. ornatissima Greene CoroLLA NoT ExcEEDING CALYX Plants less than 5 cm. tall, the larger with numerous branches; leaves ibaa nea a File aA MOE ta ses MELE D. humilis Greene Plants over 5 cm. tall, the stems solitary and simple; leaves mostly SS EET SOL Ee eee SU DY aR ean) ee D. leta Greene* PUBLISHED SPECIES ExcLUDED From Key D. insignis Greene. No character was found by which this might be distinguished from D. elegans. D. tricolor Greene is apparently not distinguishable from D. concolor. D. sikota Applegate apparently differs from D. bicornuta only in having the lobes of the upper corolla-lip longer and more divergent. From my observations D. bicornuta seems to be the most variable species of the genus. Perhaps D. stkota should be made a variety of that species. * Downingia leta Greene, apparently unreported in the California flora before this, is included in this key to the Californian species on the basis of a collection which was determined by me and then filed away so carefully as to elude later detection. This collection, which was not seen by Dr. Hoover, was given to the Herb. Calif. Acad. Sci. by Mr. Gordon H. True, his No. 518, collected June 25, 1936, near Standish, Honey Lake Valley, Lassen County. The occurrence of this species in northeastern California is rather to be expected since it is known from eastern Oregon and northern Nevada,— John Thomas Howell. 36 LEAFLETS OF WESTERN BOTANY [ VOL. II, NO. 3 NEW SPECIES OF DODECATHEON BY ALICE EASTWOOD Dodecatheon glandulosum Eastwood, spec. nov. Glanduloso-puberu- lens, czspitosum, basi vestitum foliis brunneis marcescentibus; radicibus densis carnosis ex caudice breve; foliis oblanceolatis, 3—6 cm. longis, 5—15 mm. latis, acutis obtusisve, integris, in petiolum coarctatis; scapo folios superante, 6—12 cm. alto; umbella pauciflora, floribus 4-meris, brac- teis lanceolatis, acuminatis, 1 cm. longis; tubo staminum ca. 1 mm. longo, antheris obtusis, 6 mm. longis, purpureo-sulcatis ; capsula sepalis breviore, apice dehiscente, stylo antheras superante, infra glanduloso. Type: Herb. Calif. Acad. Sci. No. 40897, collected by the author, July, 1904, in Desolation Valley, Lake Tahoe region, California. Another specimen in the herbarium was collected by Mr. S. L. Berry, July 10, 1902. Other specimens from the same region are as follows: above Grass Lake, Dr. E. C. Van Dyke, July 4, 1915; Gilmour Lake, Eastwood, July, 1904. It belongs to the section Etubulosa Pax & Knuth? and seems most closely related to D. Jeffreyi Van Houtte. It is much smaller in every way and could never be confused with that robust species. The stamen-tube is below the corolla in the D. Jeffreyi group while in this the short tube is visible above the corolla. Dodecatheon Jeffreyi Van Houtte var. odoratum East- wood, var. nov.? Dodecatheon Jeffreyi was beautifully illustrated in Van Houtte, Flore des Serres et des Jardins de l'Europe, v. 16, t. 1662. It is a robust plant with 4-merous flowers and the staminal tube below the corolla. Dr. H. M. Hall in Bot. Gaz. 31: 392 described a strongly scented variety growing in wet meadows or along streams in the Sierra Nevada which he named var. redolens. In this the flowers are 5-merous, but in general appearance it resembles the type. Enid Michaels has collected specimens in Peregoy Meadow, Yosemite region, which seem intermediate between these two. The flowers are 4-merous like the type but they give off the most delicious fragrance. They grow in drier places than the common 1 Das Pflanzenreich IV, 237. 2 Dodecatheon Jeffreyi Van Houtte var. odoratum Eastwood, var. nov. Flores 4-meri, odorati, annulo flavo, Superiore margine porphyro-marginato. Type: Herb. Calif. Acad. Sci. No. 243409, collected by Enid Michaels in Peregoy Meadow, Yosemite region, California, July 4, 1937. AUGUST, 1937] | NEW SPECIES OF DODECATHEON 37 variety. The leaves are oblanceolate, acute, ca. 1 mm. wide, much surpassed by the long scapes which are over 3 dm. high. The flowers are violet-rose with stellately spreading petals, the ring at base bright yellow somewhat raised and edged along the top with a narrow red-brown rim. Dodecatheon subalpinum Eastwood, spec. nov. Glabrum, cespitosum ex radicibus densis, fibrosis et tuberiferis; foliis oblanceolatis, obtusis, integris, in petiolum coarctatis, ca. 4—7 cm. longis, 3—12 mm. latis; scapo 4—10 cm. alto, folios superante; umbella 1—3-flora, floribus nutantibus, 5-meris, bracteis membranaceis, lanceolatis acuminatis; calyce 3—5 mm. longo, lobis acuminatis; corolla phcenicea, marcescente, pallida, lobis 5—7 mm. longis, 2—3 mm. latis, acuminatis ; tubo staminum 4 mm. longo, papil- loso, purpureo; antheris zquilongis obtusis flavescentibus, connectiva obscura ; stigmate subclavato antheras superante ; capsula 6—10 mm. longa, erecta, circumcissa, calycem superante 4—5 mm. Type: Herb. Calif. Acad. Sci. No. 159999 (flower), collected July 7, 1928, by Mrs. Charles Derby on Silliman Crest, 10,000 ft., Sequoia National Park, Tulare County, California. Specimens seemingly the same species have been taken as the type of the fruit, Herb. Calif. Acad. Sci., No. 40781. They were collected by the author on the trail to the Monarch Mine, near Mineral King, Tulare County, July 21, 1903. It was a subalpine region, where, among other alpine species, were Primula suffrutescens, Penstemon Davidsoni, and Epilobium obcordatum. This may be the same as D. Hendersoni Gray var. yosenuti- anum Mason (Madrofio 1: 187) which it resembles in the small tubers among the roots and the rather long staminal tube. It belongs to the section Purpureo-tubulosa Knuth. Dodecatheon zionense Eastwood, spec. nov. Rhizoma breve, crassum, foliis ellipticis vel spatulatis, obtusis, integris, in petiolum latum coarctatis, ca. 2 dm. longis, 5 cm. latis, petiolis laminis brevioribus; scapo 3—4 dm. alto, folios superante; umbella ca. 10-flora, pedicellis gracilibus, strictis, 1—4 cm. longis, minute pulverulentibus, bracteis lanceolatis, attenuatis ; calyce campanulato, 5 mm. longo, venoso, segmentis deltoideis, acuminatis, 4 mm. longis; corolla purpurascente, segmentis oblongo-linearibus, obtusis, ca. 1 cm. longis, 3—4 mm. latis ; tubo staminum breve, aurantiaco; antheris obtusis, 1 cm. longis, connectiva ovata, rugulosa; capsula calyce longiore, apice dehiscente et purpurascente. Type: Herb. Calif. Acad. Sci. No. 232003, collected July 25, 1933, in Zion Canyon, Utah, by Alice Eastwood and John Thomas Howell (No. 1144). This large-leaved species, with the 38 LEAFLETS OF WESTERN BOTANY [ VOL. II, NO. 3 scape and umbel surpassing the leaves, grew on the wet cliffs up the canyon where the beautiful variable columbines, ferns, and other plants formed an arch of loveliness on the side of the cliff out of which water oozed from the clefts in the rocks. It belongs in the section Etubulosa Pax & Knuth. The description of the flowers is drawn from withered flowers that persisted on the ripe fruit. New Locatitiges In North AMERICA FOR SPH#ROPHYSA SALsuLA (Patu.) DC. My attention was called to this intro- duced plant by the article in Torreya, March and April of this year, in which E. J. Alexander gave two localities, one in Colo- rado and one in Utah, and a general description of the species. In 1936, June 28th, it was collected by Eastwood and Howell (No. 3510) two miles north of Hermiston, Umatilla County, Oregon. Tentatively it was recorded in the note book as an Astragalus but it had an alien look about it that caused it to be set aside for future investigation. It was a conspicuous plant with salmon-colored flowers and papery inflated pods. Under unnamed Astragalus another collection was brought to light. This was collected by Susan Delano McKelvey, May 21, 1934 (No. 4565), between Winslow and Holbrook, Navajo County, Arizona. Mrs. McKelvey reported the flowers as coral- color turning bluish as they fade and the plant making a showy mass. Mr. Alexander gives the color as orange and DeCandolle reports the flowers as “ruberrimi” and a related species with the flowers “‘sordide diluteque purpurei.” Ideas of color vary with different individuals and perhaps the flowers do vary in color. The only real difference lies in the style: Spherophysa has a bearded style and Astragalus, a style without a beard. Taubert in Engler and Prantl Pflanzenfamilien includes Spherophysa in Swainsona. Three specimens are in the her- barium of the California Academy of Sciences from other lands, two from Asia Media and one from “Songaria chin. ad lacum Saisang-Nor’’ (ex Herb. Acad. Petrop.). Still further North American localities for this recently identified species may be hidden in herbaria under unknown Astragalus.—Alice Eastwood. AUGUST, 1937] NOTES ON CAREX 39 NOTES ON CAREX—X BY J. W. STACEY Carex Sartwelliana Olney (C. yosemitana L. H. Bailey), endemic in California, has for some time presented some prob- lems in differentiation, and with a study of material from three herbaria, it has been found that two distinct species have been distributed under this name as the species had been interpreted by Mackenzie. Carex Sartwelliana Olney was described from two collections from the Yosemite Valley, collections of Brewer and of Bolander, from an altitude of about 6000 feet. The proposed segregate grows at a higher altitude, at about 10,000 feet, always, as far as known, at or above timberline. As this plant has already been named C. Congdonu by L. H. Bailey (Bot. Gaz. 21:6,—1896), this name will have to be adopted, although the description is very meager, and misses the principal characteristic difference. Carex Congdoni is a taller and stouter plant than C. Sartwelliana, with a thicker and more purplish base, and has a much different aspect, so much so that Mr. Frank W. Peirson of Pasadena, who is a very discriminating collector, called our attention to it two years ago, but at that time enough material was not available for comparison. The following key points out the obvious differences: Culms phyllopodic, or very slightly aphyllopodic; leaves softly pubes- cent; perigynia broadly obovoid, abruptly beaked, 2.5—3.5 mm. long; staminate scales prominently white-ciliate; achenes broadly SO rh OE UP es oS LC Say ees a CEs TOM FRE C. Sartwelliana Culms very strongly aphyllopodic; leaves glabrate or puberulent; peri- gynia of the lance-ovoid type, tapering into the beak, 3.5—4 mm. long; staminate scales not prominently white-ciliate; achenes Les CoM” fe ea Fes ed See ENE oy ERA Re AACR EM ee IRD Ae C. Congdonit The great difference in the two species is in the degree of phyllopody. This, together with the differences in the leaves, the scales, the perigynia, and the achenes, would seem to pre- clude a mere varietal distinction. Besides, if one was a variety of the other, it would be natural for the coarser variety to be at the lower altitude and the more slender one at the higher, but the converse is true. As an adequate description of C. Cong- donii has not been written, we are giving one that will perhaps serve the purpose of differentiating it from C. Sartwelliana. 40 LEAFLETS OF WESTERN BOTANY [ VOL. II, NO. 3 Carex Concponi L. H. Bailey. Cespitose from stout rootstocks; the culms stout, 4—9 dm. high, very strongly aphyllopodic, from little to much-exceeding the leaves, strongly purplish-red at tht base; leaves glabrate or puberulent, the blades light green, 1—4 dm. long, 3—8 mm. wide, roughened toward the attenuate apex; terminal spike staminate, with a few perigynia, androgynous, or sometimes even gynzcandrous, short-peduncled, rarely with a smaller spike below, linear to clavate or ovoid, 1.5—3.5 cm. long, 3—8 mm. wide, the scales narrowly oblong or lanceolate, not conspicuously ciliate, acute or obtusish, purplish-brown with lighter center; pistillate spikes 3 or 4, approximate or somewhat separate, erect, sessile, or somewhat peduncled, cylindric, 1.5—5 cm. long, 4.5—7 mm. wide, lowest bract leaf-like, lightly sheathing, from shorter to longer than the inflorescence; scales ovate, appressed-hairy, awned, mucronate or acute, purplish-brown with white hyaline margins, and lighter 3-nerved center, narrower than the perigynia; perigynia 3.5—4 mm. long, 1.25—1.75 mm. wide, shortly white-pilose, the body lance-ovoid or oblanceolate, short-stipitate, tapering into the beak; achenes elliptic, 2 mm. long, 1—1.25 mm. wide, yellowish, substipitate, short-apiculate; stigmas reddish-brown, slender. A list of exsiccate of the two species follows. The abbrevi- ations used are: (CA) California Academy of Sciences, San Francisco; (DS) Dudley Herbarium of Stanford University ; and (UC) University of California, Berkeley. CarEx Coneponir L. H. Bailey. Tulare Co.: Sawtooth Peak, Hall & Babcock No. 5687 (UC), Dudley No. 1616, 1617 (DS); Alta Peak, Dudley No. 1533 (DS); Mt. Silliman, Dudley No. 1504 (DS) ; head of Middle Kaweah River, Dudley No. 1261 (DS). Inyo Co.: Rock Creek Lake Basin, Peirson No. 9106 (UC), 10835 (CA), 11396 (CA). Mariposa Co.: Mt. Buena Vista, Congdon in 1895 (DS). Tuolumne Co.: trail from Lundy to Tioga, Congdon in 1894 (UC); Mt. Dana, C. W. Sharsmith No. 2364 (CA); Cathedral Creek, Hall No. 11906 (UC). CAREX SARTWELLIANA Olney. Riverside Co.: Mt. San Jacinto, Parish in 1892 (DS) ; Idyllwild, Wilder No. 918 (UC) ; twenty miles north of Idyllwild, M. E. Jones in 1926 (CA). Tulare Co.: Oriole Lake, Dudley in 1900 (DS). Mariposa Co.: Big Oak Flat Road, Congdon in 1896 (UC), Congdon in 1897 (UC); Yosemite, Brewer No. 1636 (DS, UC); Yosemite Big Trees, Bolander No. 6221 (DS, UC), Congdon in 1895 (DS, UC). Tuolumne Co.: Tuolumne Meadows, Dudley in 1901 (DS). AUGUST, 1937] ZYGADENUS FONTANUS 41 ZYGADENUS FONTANUS,? A NEW SPECIES FROM MT. TAMALPAIS BY ALICE EASTWOOD This has been known to the author for many years. It grows in springy places on Mt. Tamalpais especially on the south side amid the azaleas and blooms some time after the common early spring species, Z. Fremontt, is in fruit. Near Rock Spring it flourishes along the rivulet fed by the spring and is also to be found in the marshy area in the Potrero. It has a large bulb 4 cm. in diameter with the outer coat chestnut-brown. The basal leaves are long, often surpassing the flowering stems, the largest about 2 cm. wide. The plants grow to a height of 8 dm. (almost 3 ft.) and the stems of the larger plants have a diameter of 15 mm. The horizontally spreading flowering branches begin about the middle of the scape, the terminal much longer than the laterals. The white star-shaped flowers are on spreading pedicels that become hori- zontal in fruit. The perianth is about 1 cm. broad with the seg- ments cordate at base and marked within by a yellow gland serrate along the truncate apex above the short claw. The cap- sules are about 12 mm. long with the perianth persisting. In the second volume of the Botany of California Geol. Surv. p. 183, at the end of the description of Z. venenosus Watson, reference is made to a plant collected in Sonoma County by Bolander, who stated that the bulbs were eagerly eaten by hogs and to the farmers of the region it was known as “Hogs’ Potato.” Zygadenus venenosus is especially common in the upper meadows of the Sierra Nevada. On account of the poisonous qualities of its bulb from which its specific name arises, it is generally called “Death Camass.” The quality of the bulb of Z. fontanus is not known but it seems probable that Bolander’s “Hogs’ Potato” may be the same. Specimens agree- 1 Zygadenus fontanus Eastwood, spec. nov. Bulbus rotundus, 3—4 cm. diametro, exteriore castaneus. Scapus 5—10 dm. altus, robustus, glaber, medio divaricate paniculatus; foliis radicalibus scapos superantibus, 1—2 em, latis, ramis panicule horizontali-divaricatis, bracteis longis lineari- attenuatis; floribus albis, 1 cm. diametro, pedicellis horizontali-divaricatis, bracteolis parvis; perianthii segmentis ovatis, obtusis, basi cordatis et unguiculatis, glandibus flavis, apice dentatis et truncatis; capsulis circa 2 cm. longis, perianthiis marcescentibus. Type: No. 241860, Herb. Calif. Acad. Sci., collected by John Thomas Howell, No. 12656, in springy places on serpentine near Bootjack on the south side of Mt. Tamalpais, Marin County, California, June 7, 1936. 42 LEAFLETS OF WESTERN BOTANY [ VOL. II, NO. 3 ing with the original description of Z. venenosus are in the herbarium of the California Academy of Sciences from the Santa Lucia Mountains in Monterey County. These mountains may be the region where Douglas made his collection, which is the type of Z. venenosus. Zygadenus fontanus comes nearest to Z. micranthus East- wood, of which it might be regarded as a gigantic subspecies. The horizontally spreading peduncles and pedicels are similar, but in flowers, leaves, height, etc., it is very much larger. Zygadenus micranthus is more frequently simply racemose than paniculate. Zygadenus fontanus likes springy places in the ser- pentine areas. It grows in a similar environment at Tiburon and on higher ground in the Big Carson mountainous region of Marin County. NEW CALIFORNIAN PLANTS BY JOHN THOMAS HOWELL Eriogonum vestitum J. T. Howell, spec. nov. Annuum, erectum, 1—4 dm. altum, albo tomento omnino vestitum; caulibus teretibus fere simplicibus ex basi, di- vel tri-chotome ramosis superne, ramis ascendenti- bus; foliis alternantibus basi, verticillatis ad nodos inferiores, superne reductis ad bracteas subulato-triangulares, foliis inferioribus ellipticis, 1—3 cm. longis, 0.5—1.5 cm. latis, obtusis acutisve, cuneatis basi, petiolis usque ad 3.5 cm. longis, margine integris vel undulato-crispis; pedicellis paullo crassis, erectis, usque ad 6 cm. longis; involucris campanulatis, 2 mm. longis, 5-lobatis, sinibus inter lobos parvos prope completis membrana scariosa; bracteolis tenuissimis, papillas divaricatas pinnate ferentibus; segmentis perianthii albis vel carneis, subsimilibus, 1.5 mm. longis, paullo accrescentibus post anthesin, oblongo-ovatis, obtusis, externe dense et tenuissime papillosis, margine integris et paullo incurvis; filamentis pubes- centibus basi; achenio opaco, 2.5 mm. longo, rostro tenuissime papilloso et perianthio cincto zquilongo vel rostro paullo longiore. Type: Herb. Calif. Acad. Sci. No. 243299, taken from a talus of disintegrating shale, 4 miles from Idria on road to Panoche, San Benito County, May 3, 1937, Eastwood & Howell No. 4295. The species was also collected the same day in a less densely tomentose form on Griswold Creek southeast of Panoche, Eastwood & Howell No. 4276. Eriogonum vestitum would seem to be a distinct addition to the subgenus Ganysma and appears to be most nearly related to AUGUST, 1937] NEW CALIFORNIAN PLANTS 43 E. argillosum J. T. Howell which also inhabits the dry inner South Coast Ranges. Not only in general aspect are the two readily separable (EZ. vestitum being clothed throughout with a persistent generally thick white tomentum and E. argillosum being glabrous throughout or early glabrescent except for the thin persistent tomentum on the lower side of the leaves), but they are distinct in important details of the involucre and flower. In E. argillosum the inner wall of the involucre bears long tomen- tose hairs and in E. vestitum the inner involucral walls are glabrous. In &. argillosum the bractlets within the involucre bear long tangled hairs (chiefly near the base) as well as short stout papillz of the kind which alone are found along the bract- lets of E. vestitum. The perianth-segments of £. argillosum tend to be broadest above the middle while those of E. vestitwm are broadest near or below the middle, and the perianth of the former, although cellular in texture and sometimes papillate near the base, lacks the conspicuous papillate protuberances which every- where cover the outer surface of the perianth of the latter. Ceanothus gloriosus J. T. Howell, spec. nov. Frutex prostratus etiam ad apices ramulorum, caulibus radicantibus subter, ramulis et pedunculis magis minusve sericeo-tomentulis; foliis viridulo-olivaceis supra, griseo- viridulis et areolato-tomentulis infra, planis vel paullum plicatis in longi- tudinem, late oblongis usque ad suborbicularibus, late cuneatis basi, subtruncatis vel emarginatis apice, usque ad 4.5 cm. longis et 3.5 cm. latis, conspicue pinnato-nervosis infra, margine zqualiter dentatis, dentibus 15—30, rigidis et spinescentibus, petiolo brevi, stipulis maximis usque ad 3.5 mm. longis, suberosis, primo ferrugineis; floribus minimis cyaneis in corymbo magno et denso secundum ramos compositis vel corymbis paucis et apicibus ramorum subcapitatis, floribus plerumque ultra 20 in corymbum, 6 mm. latis; capsulis subglobosis, 4—5 mm. latis, 3—4 mm. altis, cornibus subapicalibus, erectis vel divergentibus, paullum rugulosis, 1—2 mm. longis, cristis mediis obsoletis vel obscuris et rugosis in suturis inter lobos humilis capsule; seminibus oblongis vel ellipticis, 2—3 mm. longis, nigris, nitentibus. Types are from Anchor Bay, Mendocino County: that for flower, Moffitt & Orr, Apr. 4, 1937, Herb. Calif. Acad. Sci. No. 242192; that for fruit, Eastwood & Howell No. 4493, May 31, 1937, Herb. Calif. Acad. Sci. No. 246064. Specimens have also been seen from Sonoma County (8 miles south of Stewart’s Point, J. T. Howell No. 11742) and from Marin County (Point Reyes Peninsula, Eastwood & Howell No. 2123, the type 44 LEAFLETS OF WESTERN BOTANY [ VOL. i, NOs locality of C. rigidus Nutt. var. grandtfolius Torr.,* the varietal epithet which has been variously combined to designate our plant). Two forms of C. gloriosus are to be distinguished: the one here regarded as the type of the species, prostrate to the very tips of the stems, which is found on coastal slopes and mesas in the immediate vicinity of the sea from Marin County to Mendo- cino County; the other an erect variant, with widely spreading virgate fastigiate-divaricate branches, which is found back from the coast in brush and chaparral of hills and valleys. The striking habital difference between these two plants is here given nomen- clatorial recognition. Ceanothus gloriosus J. T. Howell var. exaltatus J. T. Howell, var. noy. Frutex erectus circa 2 m. altus, subrigide multiramosus, ramis ultimis elongatis virgatis fastigiato-divaricatis, cortice fusco vel cinerascenti, inflo- rescentiis virgato-racemosis, 2—3 dm. longis. The types were collected along an arroyo in the Vine Hill district north of Sebastopol and about 10 miles west of Santa Rosa, Sonoma County, where the shrubs grow in sandy clay soil: J. T. Howell No. 5781 (flowering type, Mar. 14, 1931, Herb. Calif. Acad. Sci. No. 194493) and J. T. Howell No. 6694 (fruit- ing type, June 15, 1931, Herb. Calif. Acad. Sci. No. 194481). Other collections have been seen from Marin County (Bolinas Ridge, Eastwood No. 3476), Sonoma County (near Sebastopol, J.T. Howell No. 6692 ; Vine Hill district, M. S. Baker No. 5236, J. T. Howell No. 6693), and Mendocino County (5 miles east of Fish Rock, James Moffitt; Mendocino City, Eastwood No. 11411; Mendocino barrens, Charlotte Hoak). It seems likely that C. gloriosus var. exaltatus occurs in the Santa Rosa and Mt. Hood ranges (cf. M. S. Baker’s transplant from Los Guilicos Valley) ; but all specimens which have been seen from south and east of Santa Rosa which approach C. gloriosus appear to be variants between C. gloriosus, C. divergens Parry, and C. cuneatus (Hook.) Nutt. that have perhaps originated through hybridization. * Ceanothus rigidus Nutt. in T. & G. var. grandifolius Torr., Pac. RR. Rep. 4:75 (1857) ; Trelease, Syn. Fl. 1, pt. 1:417 (1897) ; McMinn, Contrib. Dudley Herb. 1:145 (1930). C. crassifolius Wats., not Torr., in part, Bibliog. Ind. 164 (1878). OC. verrucosus Nutt. in T. & G. var. grandifolius (Torr.) K. Brandg., Proc. Calif. Acad. Sci., ser. 2, 4:207 (1894). C. pros- Sper Benth. var. ‘grandifolius (Torr.) Jepson, Man, 624 (1925) ; FL Calif. > 479 (1936). AUGUST, 1937] FURTHER STUDIES IN ERIOGONUM 45 A study of C. gloriosus over a period of years would seem to indicate that it does not find its nearest relative in any of those species to which the oft-disposed var. grandifolius has been referred. Rather it appears to be most closely related to C. pur- pureus Jepson, a local and apparently distinct endemic in the southern part of the Napa Range, Napa County. However, C. purpureus and C. gloriosus are distinct in one or more charac- ters of habit, leaves, flowers, and seeds. Together, C. purpureus and C. gloriosus can be separated from other related species of Ceanothus on characters of foliage and fruit. A collection from Guerneville, Sonoma County, which has not been seen but which is perhaps referable to C. gloriosus, is cited by Prof. Jepson (Fl. Calif. 2: 478) as C. Jepsonit Greene var. purpureus (Jepson) Jepson and would seem to indicate that he too recognizes the close affinity between C. purpureus and what is here considered a distinct species. It only remains to note the great beauty of C. gloriosus in cultivation. I feel certain that for the rock garden the typical prostrate maritime variant will prove a desirable and delightful shrubby creeper; and as for var. exaltatus, no Ceanothus is to me more effective in habit, flower, and foliage than a fair-sized specimen in full bloom. FURTHER STUDIES IN ERIOGONUM—I BY SUSAN G. STOKES Eriogonum vimineum Dougl. var. caninum Greene (FI. Fran. 150; type locality, Tiburon, Marin County, California) was described from a specimen which has the low spreading habit of E. Nortoni Greene, but the terminal branches are shortly extended and the involucres are narrower. The “flexuously divaricate” forms with slender branchlets and narrow involucres which are common in northern California should be given the name E. vimineum Dougl. var. californicum Gdgr. (Bull. Soc. Bot. Belg. 42: 199; type locality, Petaluma, Sonoma County, California). 7 5 A 7 The following names were not accounted for in the “Genus Eriogonum”’ (Stokes, 1936) : 46 LEAFLETS OF WESTERN BOTANY [ VOL. II, NO. 3 ERIOGONUM ROSENSE Nelson & Kennedy, Proc. Biol. Soc. Wash. 19: 36 (1906). Very near to E. ochrocephalum Wats. subsp. anemophilum (Greene) Stokes, pubescence of the same character, size smaller, flowers yellow as in E. ochrocephalum subsp. typicum Stokes. Type locality: “summit of Mount Rose, Washoe County, Nevada, elevation 10,800 feet,’ P. B. Kennedy No. 1180. ERIOGONUM RHODANTHUM Nelson & Kennedy, Proc. Biol. Soc. Wash. 19:35 (1906). An alpine form of E. ovalifolium Nutt. subsp. vineum (Small) Stokes, not so reduced as typical var. nivale (Canby) Jones, forming dense mats on hard rocky ground, caudex branched and decumbent rather than stubby ; slightly resembling subsp. extmium (Tidestr.) Stokes. Differs from E. ochrocephalum Wats. subsp. anemophilum (Greene) Stokes in the character of the pubescence which is close and white, also in the flowers which have broad sepals as in E£. ovali- folium Nutt. which is rather polymorphous. Type locality: “summit of Mount Rose, Washoe County, Nevada, elevation 10,800 feet,’ P. B. Kennedy No. 1184. ERIOGONUM STRICTUM Benth. var. LACHNOSTEGIUM Benth. in DC., Prodr. 14: 16 (1856). E. lachnostegium ( Benth.) Rydb., Fl. Rocky Mts. 221 (1917). Appears heavier than E. strictum subsp. typicum Stokes because of the rather dense pubescence; the subsp. typicwm has a glabrous, coriaceous surface. Type locality : “in collibus ad Snake river (Fremont!).” Eriogonum chrysocephalum Gray subsp. Cusickii (Jones) Stokes, comb. nov. E. Cusickti Jones, Contrib. West. Bot. 11: 10 (1903), not E. Cusick Gdgr., Bull. Soc. Bot. Belg. 42: 193 (1906). Dwarf perennial closely related to E. chrysocephalum Gray subsp. typicum Stokes, but differing in that the inflores- cence is subumbellate, some of the involucres are sessile while others are borne on short rays 4 to 8 mm. long, and the flowers are pale. Interpreted as a survival of a primitive ancestor with branched inflorescence. Type locality: “on a stony desert, Harney Co., Oregon,” Cusick No. 2603. The type collection of E. microthecum Nutt. var. idahoense (Rydb.) Stokes (E. tdahoense Rydb.) was made at Weiser, Washington County, Idaho, July 7, 1899, by M. E. Jones, No. 6511. AUGUST, 1937] FURTHER STUDIES IN ERIOGONUM 47 In their revision of the genus Eriogonum, Torrey and Gray adopted the name FE. Wrightu Torr. instead of E. trachygonum Torr. (Proc. Amer. Acad. 8:176) and E. brevicaule Nutt. instead of E. campanulatum Nutt. (1. c., 172). According to Art. 56, International Rules of Botanical Nomenclature (1935), the following names should be used: Eriogonum Wrightii Torr. subsp. typicum Stokes, nom. nov. E. Wrightii Torr. in Benth. in DC., Prodr. 14:15. E. trachygonum Torr. subsp. Wrightii (Torr.) Stokes, Gen. Eriog. 58. Eriogonum Wrightii Torr. subsp. dentatum (Stokes) Stokes, comb. nov. E. trachygonum Torr. subsp. dentatum Stokes, Gen. Eriog. 60. Eriogonum Wrightii Torr. subsp. glomerulum (Stokes) Stokes, comb. nov. E. trachygonum Torr. subsp. glomerulum Stokes, Gen. Eriog. 59. Eriogonum Wrightii Torr. subsp. membranaceum (Stokes) Stokes, comb. nov. E. Wrighttt Torr. var. membranaceum Stokes in Jepson, FI. Calif. 1:416. E. trachygonum Torr. subsp. membranaceum (Stokes) Stokes, Gen. Eriog. 59. Eriogonum Wrightii Torr. subsp. Pringlei (Coult. & Fish.) Stokes, comb. nov. E. Pringlei Coult. & Fish., Bot. Gaz. 17:351. E. trachygonum Torr. subsp. Pringlei (Coult. & Fish.) Stokes, Gen. Eriog. 59. Eriogonum Wrightii Torr. subsp. subscaposum (Wats.) Stokes, comb. nov. E. Wrightii Torr. var. subscaposum Wats., Bot. Calif. 2:29. E. trachygonum Torr. subsp. subscaposum (Wats.) Stokes, Gen. Eriog. 59. ErroconumM WricuHtTit Torr. subsp. TAXIFOLIUM (Greene) Parish, Erythea 6:87. E. taxifolium Greene, Pitt. 1:267. E. trachygonum Torr. subsp. taxifolium (Greene) Stokes, Gen. Eriog. 59, a combination errone- ously credited to Parish. Eriogonum Wrightii Torr. subsp. trachygonum (Torr.) Stokes, comb. nov. &. trachygonum Torr. in Benth. in DC., Prodr. 14:15. E. trachygonum Torr. subsp. typicum Stokes, Gen. Eriog. 58. Eriogonum brevicaule Nutt. subsp. typicum Stokes, nom. nov. E, brevicaule Nutt., Jour. Phil. Acad. Sci., ser. 2, 1:163. E.campanulatum Nutt. subsp. brevicaule (Nutt.) Stokes, Gen. Eriog. 77. Eriogonum brevicaule Nutt. subsp. campanulatum (Nutt.) Stokes, comb. nov. E. campanulatum Nutt., Jour. Phil. Acad. Sci., ser. 2, 1:163. E. campanulatum Nutt. subsp. typicum Stokes, Gen. Eriog. 77. Eriogonum brevicaule Nutt. subsp. grangerense (Jones) Stokes, comb. nov. E. grangerense Jones, Contrib. West. Bot. 11:12. E. campanu- latum Nutt. subsp. grangerense (Jones) Stokes, Gen. Eriog. 77. Eriogonum brevicaule Nutt. subsp. leptothecum (Stokes) Stokes, comb. nov. E. campanulatum Nutt. subsp. leptothecum Stokes, Gen. Eriog. 78. Eriogonum brevicaule Nutt. subsp. orendense (A. Nels.) Stokes, comb. nov. E. orendense A. Nels., Bot. Gaz. 34:21. E. campanulatum Nutt. subsp. orendense (A. Nels.) Stokes, Gen. Eriog. 78. 48 LEAFLETS OF WESTERN BOTANY [ VOL. II, NO. 3 Eriogonum pedunculatum Stokes, spec. nov. Annuum gracile aspectu E, viminei, sed involucris pedunculatis, pedunculis gracilimis 1—10 mm. longis, involucris et pedunculis rectis strictis. Slender annual with narrow erect habit, 3—4 dm. tall; leaves nearly round, clustered in a rosette at the base, also, in smaller size, within the axils of the lower circles of bracts, densely tomentose beneath, less so above; stems di- or tri-chotomous, lower part tomentose, upper glabrous and coriaceous, final branches virgate, bearing from several to about 10 involucres, nodes bearing minute, acute bracts; involucres pedunculate, narrowly turbinate, 1.7—2.3 mm. high and 1 mm. wide, grooved, teeth 5, acute; peduncles delicate, 1—10 mm. long, bearing the involucres close to and parallel with the stem; flowers few, usually but one appearing at a time, narrow at base, glabrous, pale, about 1.5 mm. long. Type: No. 131669, Herb. Calif. Acad. Sci., collected by F. E. Blaisdell at Mokelumne Hill, Calaveras County, California. Eriogonum cernuum Nutt. subsp. tenue (T. & G.) Stokes var. multipedunculatum Stokes, var. nov. A subspecie tenwi differt: habitu diffuso, ramulis numerosis recurvis capita globosa formantibus, internodiis et pedunculis brevibus, pedunculis gracilibus refractis, inferioribus circa 1 cm. longis, superioribus sepe 1—2 mm. longis; involucris brevioribus, turbinatis; floribus minoribus. Type: No. 191116, Herb. Calif. Acad. Sci., collected by J. T. Howell, No. 7988, 40 miles west of Austin, Lander County, Nevada, Aug. 25, 1931. This variety approaches E. vimineum Dougl. subsp. Batley: (Wats.) Stokes in habit, coriaceous stems, and small involucres ; the outer floral segments, like those of typical E. cernuum, are slightly widened at the base. This portion of the Great Basin produces several forms of the E. deflexum and E. nutans types which vary in the length of the peduncles and the greater stout- ness of the parts. Even subsp. tenue is seldom so delicate as is var. multipedunculatum. Eriogonum apiculatum Wats. var. subvirgatum Stokes, var. nov. A specie differt: ramulis ultimis virgatis, involucris sessilibus vel sub- sessilibus, involucris in axillis inferioribus pedunculatis ; floribus aut acuto- turbinatis aut obtusis basi. Type: No. 168499, Herb. Calif. Acad. Sci., collected by Ralph Hoffmann, Sept. 8, 1929, on Mt. San Jacinto, Riverside County, California, alt. 7000 ft. The typical form of EF. apicu- latwm has long internodes and well developed peduncles. Speci- mens of the variety, of which there are eight, are low, 1—2 dm. tall, the involucres nearly sessile. Vor, il LEAFLETS of No. 4 WESTERN BOTANY Y CONTENTS Notes on Schizococcus Auice Eastwoop A Remarkable New Phacelia JoHN THomas Howe Lt Further Studies in Eriogonum—II Susan G. STOKEs New Species of Western Plants ALice Eastwoop Notes on Western Plants HucuH O’NEILL New Varieties of Western Plants—I JoHN Tuomas Howe. A Collection of Douglas’ Western American Plants—I JoHNn THomas Howe Li Notes on Carex—XI J. W. STacey SAN FRANCISCO, CALIFORNIA NoveMBER 20, 1937 PAGE 49 51 52 54 56 57 59 63 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. Pe INCHES jliiint METRIC QHUnnpanngganedgUUn A) O000 O00 U0 Lg Owned and published by Atice Eastwoop and JoHN THomas Howeg.Li NOV 2 6 1937 NOVEMBER, 1937] NOTES ON SCHIZOCOCCUS 49 - NOTES ON SCHIZOCOCCUS, WITH A KEY TO THE SPECIES BY ALICE EASTWOOD Schizococcus Eastwood was differentiated from Arctostaphy- los Adans. in the first volume of this magazine, page 98, because of the dehiscent fruit, a character unknown in any other genus of the group. The fruits never can be called manzanitas (little apples ) since, unlike the species that are rightfully so named, the fruits fall to the ground, and the nutlets, encased in the dry outer coat, separate as they fall and can be found only by diligent search under the bushes. KEY TO THE SPECIES OF SCHIZOCOCCUS Parts of the flower in fives; Sierran species. Foliage pallid, old stems shreddy.............-2..----:::-t0sessee+- S. nissenanus Foliage green, old stems smooth........0.......-.-----:ecseeeeeeeeeeees S. myrtifolius Parts of the flower in fours; coast species. MECN: HIP) iS 01 3 Cy RR ne S. nummularius Fruit with generally 2, rarely 3, nutlets -......................-.----- S. sensitivus SCHIZOCOCCUS NISSENANUS (C. H. Merriam) Eastwood." This was described by Dr. C. Hart Merriam as Arctostaphylos nissenana (Proc. Biol. Soc. Wash. 31:102). The shrubs were without flowers or fruits, but Dr. Merriam, who knows Cali- fornian manzanitas, at once recognized it as distinct. I sought for years to locate it in the vicinity of the type locality, two or three miles north of Louisville, Eldorado County. An account of its rediscovery with a full description was published by Mr. Howell in vol. 1, p. 253, of this magazine. The character of the fruit places it in the genus Schizococcus. It has been reported from other localities and may be more widely spread than is known at present, perhaps buried in the dense chaparral areas so common in the foothill region of the Sierra Nevada. It is related to S. myrtifolius (Parry) Eastwood from Ione, both having the parts of the flowers in 5’s, but is quite distinct because of the pallid foliage and shreddy bark of the old stems. 1 Schizococcus nissenanus (C. H. Merriam) Eastwood, comb. no leita nissenana C. H. Merriam, Proc. Biol. Soc. Wash. 31: 102 Leafi. West. Bot., Vol. II, pp. 49-64, November 20, 1937. 50 LEAFLETS OF WESTERN BOTANY [voL. II, NO. 4 ScHIzOCoccUS NUMMULARIUs (Gray) Eastwood. Schizo- coccus nummularius (Gray) Eastwood and S. sensitivus (Jepson) Eastwood ” differ from the preceding in the parts of the flowers in 4’s, On the Mendocino Plains adjacent to Mendocino City, the type locality of S. nummularius, erect and prostrate plants grow side by side, indistinguishable except by habit of growth. Herbarium specimens would scarcely show this difference, but since the species was described as erect, this form has to be con- sidered the type. Some three miles east of Point Arena, Mendo- cino County, on the road above the Garcia River, a variety of this species occurs. It is less delicate in all its parts, with larger and broader leaves, and is always erect. I am naming this plant S. nummularius var. latifolius.* Both the species and the variety have globular fruits containing four or five seeds. A hybrid apparently between S. nummularius and the neigh- boring Arctostaphylos setosissima Eastwood was collected by the author near Mendocino City, June 28, 1922, No. 11458. It has the small leaves and habit of S. nummularius but the hairy stems and pubescence of A. setosissima. The few fruits on the specimens are those of Schizococcus and contain five seeds. SCHIZOCOCCUS SENSITIVUS (Jepson) Eastwood. This is a common species on Mt. Tamalpais, the type locality. Compared with S. nummularius it is more robust, compact, always erect, and with larger leaves. On the south side of Mt. Tamalpais amid other chapparral it sometimes becomes over six feet tall, but is generally about three feet. The most distinctive difference, however, between this and S. nummularius is in the fruit. It is flat, sulcate longitudinally down the middle, and contains two seeds, but it is never globular. The species is also common on the hills above the Big Basin, Santa Cruz County, as well as in other parts of the Santa Cruz Mountains. 2 Schizococcus sensitivus (Jepson) Eastwood, comb. nov. Arctostaphylos sensitiva Jepson, Madronfio 1:85, 94 (1922). 8 Schizococcus nummularius (Gray) Eastwood var. latifolius Eastwood, var. nov. A specie differt: robustior, in omnibus partibus minus delicatis, foliis majoribus latioribusque. Type: Herb. Calif. Acad. Sci. No. 247005, collected on hills east of Point Arena, Mendocino County, California, May 31, 1937, by Eastwood and Howell, No. 4459. NOVEMBER, 1937] A REMARKABLE NEW PHACELIA 5I A REMARKABLE NEW PHACELIA BY JOHN THOMAS HOWELL Phacelia Dalesiana J. T. Howell, spec. nov. Humilis, perennis, foliis et inflorescentiis ex apice rhizomatis lignei directi simplicis vel pauci- partiti; foliis rosulatis oblongis vel ellipticis ad ovatis, 1.5—4.5 cm. longis, 0.5—2.5 cm. latis, acutis vel subobtusis, cuneatis usque ad subcordatis basi, laminis breviter et anguste decurrentibus, tenuiter pubescentibus supra et infra, pinnate venosis, petiolis tenuibus, 1.5—7 cm. longis, villosulis, viscidulis, basi dilatatis; inflorescentiis ex axillis foliorum basalium emanantibus, pedunculis viscidulo-pubescentibus decumbentibus, 3—6 cm. longis, sub cyma 2 (vel 3) folia reducta rhombico-ovata 1—2 cm. longa petiolata ferentibus, cyma laxe pauciflora vix scorpioidali, usque ad 6 cm. longa, pedicellis usque ad 2 cm. longis; calyci usque ad basin partito, lobis paullum inzqualibus, oblongo-lanceolatis, obtusis, 3 mm. longis, in fructu accrescentibus, inzequalibus, usque ad 6 mm. longis, pubescentibus, sub- coriaceis; disco nullo; corolla fere rotata, albida, demum decidua, in fauce purpureo-maculata, intus glabra, extus minute pubescenti, 1—1.5 cm. lata, lobis subrotundis vel late elliptico-ovatis, 1 cm. longis, apice rotundis; squamis 10, fere semiorbicularibus, cum corolla neque filamento coalitis, 2 mm. longis; filamentis glabris, 6 mm. longis, basin corollz insertis, antheris purpureis, 2 mm. longis; stylis sparse barbatis basi, 7 mm. longis, ramis 6 mm. longis; ovario setaceo, uniloculari, placentis apice vix incras- satis, ovulis 2 ad quamque placentam, pendulis; capsula suborbiculari, coriacea, 4 mm. diametro, valvis prominentes parietales placentiferas carinas ferentibus, sepe 2-sperma; seminibus 2.5—3 mm. longis, hemi- sphericis, facie plana tenuiter excavata et humili-carinata, plus minusve spongioso-reticulatis. This Phacelia is named in honor of E. Dales (Mrs. H. C.) Cantelow, who first brought it to me from the summit of the Scott Mts. in Trinity County north of Carrville in May, 1936. In the following August, I was taken to Trinity County on a botanical excursion by Mr. and Mrs. Cantelow, at which time I was able to collect fruiting material (J. T. Howell No. 12736). The type of the species, which is in flower and young fruit, was collected at the same station, June 25, 1937, Eastwood & Howell No. 5014 (Calif. Acad. Sci. Herb. No. 243434). Additional fruiting material was obtained on July 30, 1937, J. T. Howell No. 13691. Several remarkable characters distinguish this Phacelia, The habit, which is like that of some otherwise entirely unrelated 52 LEAFLETS OF WESTERN BOTANY [ VOL. Il, NO. 4 Mexican species, is, to my knowledge, unique among western American species of the genus and is unlike the habit of any members of the section Euphacelia to which our plant is refer- able on characters of flowers and ovules. In fact the origin of the several inflorescences of each plant from the axils of the rosulate leaves brings to mind the distinctive habit of Hespero- chiron and the resemblance is heightened when the few-flowered, scarcely scorpioid cyme of P. Dalesiana is compared with those forms of Hesperochiron in which the peduncles are not always 1-flowered as they have usually been described. In the important characters of the corolla-scales and of the well developed placental ridges on the walls of the capsule, the plant is truly Phacelia-like, but in the position and attachment of the ovules there is diver- gence. The position of the ovules at the top of the placental ridge is not found in species of Phacelia which I have examined, but it is characteristic of the genus Draperia. Our plant lacks the bulbous placental projections to which the ovules of Draperia are attached, but the pendent position of the ovules is like that of the ovules in Draperia and in certain few-ovulate types of Hesperochiron. From a consideration of these critical aspects, it would appear that P. Dalesiana might represent a reliquial expression of an ancient type or complex from which not only the genera Draperia and Hesperochiron developed but also from which species-groups of modern Phacelia with less specialized habit and placentation have diverged. FURTHER STUDIES IN ERIOGONUM—II BY SUSAN G. STOKES Eriogonum reliquum Stokes, spec. nov. Ad E. niveum et E. vimineum affine, perenne, album; inflorescentiis virgatis ; sepalis obovatis. Low perennial, base woody, decumbent, rather loosely branched, 2—3 dm. tall, white-tomentose throughout except flowers ; leaves broadly ovate, 2—3 cm. long, closely tomentose, petiole 1—2.5 cm. long; lower bracts foliaceous ; branching rather strict, the involucres appressed to the stems, about 2.5—3 mm. long, slightly campanulate; flowers narrow at base, pale yellow to white with dark vein, 3 mm. long, outer segments obovate, glabrous. NOVEMBER, 1937] FURTHER STUDIES IN ERIOGONUM—II 53 Type: No. 248453, Herb. Calif. Acad. Sci., collected by I. W. Clokey, No. 7491, August. 9, 1937, on a gravelly brushy wash in the yellow pine belt, 2270 m., Charleston Park, Charles- ton Mts., Clark County, Nevada. No. 5444 was collected at the same locality on August 7, 1935. The name was suggested by its close resemblance to both annual and perennial members of the virgate group. It appears to be a survival of the primitive form from which the annuals could have been derived, left when its relatives drifted to the north. Eriogonum umbellatum Torr. subsp. stellatum (Benth.) Stokes var. subaridum Stokes, var. nov. A subspecie stellato differt: foliis tenuibus et mollibus petiolatis et ovatis ; ad subsp. stellatum et subsp. cognatum affine in inflorescentia pauciradiata, radiis lateralibus bracteatis ; dentibus involu- cralibus brevibus, deflexis; a subsp. cognato differt: foliis basi cuneatis et tenuiter tomentosis. Type: No. 248450, Herb. Calif. Acad. Sci., collected in the juniper belt at 2200 m. alt., Kyle Canyon, Charleston Mts., Clark County, Nevada, by I. W. and C. B. Clokey, No. 7492, July 17, 1937. Other collections made by Mr. Clokey in the Charleston Mts. are: No. 5441 and 5442, Charleston Park; No. 5440, Deer Creek, 2700 m.; No. 5452, Lee Canyon, 2450 m. ; No. 5450, Deer Creek, 2750 m. In the latter the inflorescence is very compound like that of E. umbellatum var. bahieforme (T. & G.) Jepson. In No. 5441 and No. 5440 the flowers are yellow, in No. 5452 and No. 5450 the flowers are pale.* This variety appears to be derived from the same stock as subsp. stellatum which returned to the north through western California, leaving isolated survivors in southern California. The present variety and E. umbellatum subsp. cognatum (Greene) Stokes appear to have ascended from the Colorado River depres- sion and to have become established on the mountains adjacent, var. subaridum to the north and subsp. cognatwm to the south. Both are nearer to the stellatum type than to the polyanthum. * Experiments in the garden at Stanford University indicate that color is not very constant except the fundamental color which is pale. Beautiful purple, crocus-yellow, and pink flowers, where grown side by side from seed collected in isolated spots, have reverted to the pale-flowered in the first generation. 54 LEAFLETS OF WESTERN BOTANY __[VOL. II, NO. 4 NEW SPECIES OF WESTERN PLANTS BY ALICE EASTWOOD Boisduvalia pallida Eastwood, spec. nov. Ramosa basi et supra, pallida, minute tomentosa, in senectute glabrata; ramis gracilibus, fastigiatis, sepe rubescentibus ; foliis lanceolatis, acuminatis, integris, 2—5 cm. longis, 3—10 mm. latis; floribus axillaribus, rubris, 2 cm. longis, ovario tereti, attenuato, rostrato, ca. 13 mm. longo; calyce 7 mm. longo, tubo cuneato, segmentis lanceolatis, acutis, 4 mm. longis; petalis ca. 8 mm. longis, bilobatis; cap- sulis 2-, 3-, 4-angulatis, 2—3 cm. longis, basi 1—2 mm. diametro, apice rostratis; seminibus brunneis, oblongis, 2 mm. longis. Type: No. 243301, Herb: Calif. Acad. Sci., collected by the author, No. 1021, July 11, 1912, at Goose Valley, Shasta County, California, on the George Dillman Ranch, where I had the privi- lege of collecting as their guest. The large flowers ally this species with B. macrantha Heller, but that has larger flowers, and the leaves are serrate rather than entire. The ribbed pod suggests an affinity with B. cleistogama Curran. The branches are slender, reddish, erect, and spring from the base as well asabove. The fine white tomentum which clothes the pallid foliage disappears with age, but is present on the flower- ing parts. Flowers appear in the earliest axils and are larger than those of the common species. The petals are deeply lobed, the slender seed pods are ribbed with from 2 to 4 ribs, sometimes somewhat torulose and with a long slender beak that points out- ward. The seeds are few and large and in some pods are so evident that the pod becomes almost moniliform. The type of B. macrantha Heller, with which this has been compared, is in the Herbarium of the California Academy of Sciences, No. 125. In B. macrantha the leaves are close and somewhat imbricated ; in this the leaves are distant, much smaller, and the entire plant is more slender and wiry. Phlox czsia Eastwood, spec. nov. Humilis cespitosa, 5—9 cm. alta, omnino glandulosa, dense foliosa ; foliis congestis, rigidis, bisulcatis, lineari- lanceolatis, aristatis, basi amplexicaulibus, 1—2 cm. longis, basi 2 mm. latis; floribus 1 vel paucibus, sessilibus, terminantibus caules; calyce ovoideo, ca. 9 mm. longo, segmentis 6 mm. longis, aristatis, bisulcatis, tubo membranaceo intra nervos; corolla cesia, hypocrateriforme, tubo 15 mm. longo, glanduloso, lamina 15 mm. diametro, lobis subrotundatis, 5—7 mm. NOVEMBER, 1937] NEW SPECIES OF WESTERN PLANTS 55 latis; antheris sessilibus, 3 infra medium tubz, 2 supra; ovario oblongo, stylo 4 mm. longo, ramis 1 mm. Type: Herb. Calif. Acad. Sci. No. 211123, collected in Red Canyon near the entrance to Bryce Canyon, Utah, June 20, 1935, by Alice Eastwood and John Thomas Howell, No. 792. Another collection was made in the same locality, June 18, No. 668. This lovely species grew amid the rocks in the red soil from which the canyon receives its name. The dense mats with con- gested spiny-tipped leaves gave off a mephitic odor from the abundant glands covering all parts, even the tube of the large lavender-blue corolla. The many stems arise from a woody taproot. Rudbeckia glaucescens Eastwood, spec. nov. Glabra et glauca ca. 1 m. alta; caule simplice, scaposo, monocephalo; foliis radicalibus late lanceolatis, acuminatis, contractis ad longum petiolum, integris vel sub- dentatis, lamina prope 3 dm. longa, ca. 6 cm. lata; petiolis prope 3 dm. longis ; foliis caulinis ovatis vel ovato-lanceolatis, basi cuneatis et sessilibus, uninervatis; involucri squamis lineari-lanceolatis, uniseriatis, radiis brevi- oribus; ligulis 1.5—3.5 cm. longis, prope 10 mm. latis, lanceolatis, apice bidentatis; paleis receptaculi linearibus, carinatis, achenia amplectentibus, apice deltoideis, vestitis pilis brevibus et crispis; disco fructifero cylin- draceo, brunneo, 2—4 cm. longo; floribus disci tubulosis, acheniis brevi- oribus ; acheniis angulatis, coronatis breve dentata cupula. Type: Herb. Calif. Acad Sci. No. 12054, collected by the author, Sept. 7, 1923, on the old Gasquet road in a Darlingtonia swamp, No. 12150. Other specimens from other localities in Darlingtonia swamps are in the herbarium from Del Norte County, California. This differs from R. occidentalis Nutt. in the radiate heads, the peculiar pale 1-nerved leaves, the cauline sessile with cuneate base, the radical attenuate to long petioles, not winged. In general appearance it resembles R. californica Gray from which it differs in the smooth pale foliage, the shape of the leaves, and the angled akenes crowned by a short dentate cup instead of four lobes. Stephanomeria Haleyi Eastwood, spec. nov. Perennis ?, intricate ramosa, glauca; foliis minutis, ovatis, reflexis, supra lanatis, lana decidua, tenace in axillis; capitulis multis, lateralibus, pedunculis 5—10 mm. longis, dense bracteatis, bracteis minutis, imbricatis, supra lanatis, involucri squamis linearibus, apice lanatis, margine crispis et parce lanatis, squamis exterioribus brevioribus et lanatioribus; radiis linearibus, apice truncatis 56 LEAFLETS OF WESTERN BOTANY _ [VOL. Il, NO. 4 et 5-dentatis; styli lobis 2 mm. longis; pappo fusco, plumoso ad infra medium, acheniis gracilibus, linearibus, 4-costatis. Type: Herb. Calif. Acad. Sci. No. 232019. It was collected on the west side of Walker Lake, Mineral County, Nevada, Aug. 16, 1927, by Dr. George Haley in whose honor it is named. Dr. Haley has made most valuable contributions to the herbarium from the Pribilof and Aleutian islands, from Iceland, Norway and other Arctic districts, as well as a few from Nevada. It is a bushy plant about 2 dm. high forming a round mass of intricately branching stems. The leaves are soon deciduous, leaving tufts of white wool along the stems. The heads of flowers terminate short, densely woolly-bracted stems. The outer bracts of the involucre are shorter and more woolly than the inner which are linear with crisped and somewhat woolly margins. The pink rays are truncate at the 5-toothed apex and about 2 mm. long; the pappus is brownish and plumose to a little below the middle and the plumes seem to fall separately ; the immature akenes are linear, slender and 4-ribbed. NOTES ON WESTERN PLANTS BY HUGH O'NEILL Langlois Herbarium, Catholic University of America A Cyperus New To Ca.irornia. Dr. Leroy Abrams on September 25, 1920, collected Cyperus albomarginatus Mart. & Schrad. three miles below Three Rivers, Tulare County, Cali- fornia. The collection is Abrams No. 7715 and is in the Dudley Herbarium, No. 108045. This seems to be the first record of this species from California. It was originally described from Brazil. CyPERUS FUSCUS IN CALIFORNIA. A specimen collected by Dr. Leroy Abrams in sand 3 miles northwest of Stockton on the Calaveras River, September 14, 1917, is probably the first record of Cyperus fuscus L. in California. New To Arizona. Pistia Stratiotes L., the so-called Tropical Duckweed, has been collected by Sister Mary Noel; O.S.F., in an irrigation canal in the vicinity of Yuma. Her specimen, No. 44, apparently the first record of this plant from Arizona, is in the Langlois Herbarium. NOVEMBER, 1937] NEW VARIETIES OF WESTERN PLANTS—I 57 NEW VARIETIES OF WESTERN PLANTS—I BY JOHN THOMAS HOWELL Streptanthella longirostris (Wats.) Rydb. is widely dis- tributed from the deserts of northwestern Mexico and California northward and eastward as far as central Wyoming and western Colorado. Although characterized by an intriguing and some- what perplexing morphological constitution when considered together with its near relatives in the Crucifere, it is, within itself, a nearly uniform entity mostly devoid of variation, save such as might in any season be imposed by fluctuation of tempera- ture and moisture in a desert environment. To find, then, a variant worthy of name in this species has been unusually interesting. It has been of interest, too, to note that in known distribution our new plant is nearly or quite confined to desert sands at or near sea level about the ancient head of the Gulf of California. Undoubtedly our variety was once the maritime aspect of the species before the Colorado River had unloaded its plains of silt and when the waters of the gulf extended far up the Coachella Valley toward Palm Springs in California and spread over the lowlands around Yuma in southwestern Arizona. Many gener- ations of our plant have come and gone since last it grew along the seashore in California and Arizona, but in Mexico it can still be found flourishing along the present shores of the gulf. Streptanthella longirostris (Wats.) Rydb. var. derelicta J. T. Howell, var. nov. Caulibus foliisque viridibus, haud glaucis; foliis inferioribus usque ad 10 cm. longis, pinnato-partitis, rachide angusta segmenta 3 vel 4 divaricata angusto-oblonga acuta utrimque ferenti; foliis superioribus re- ductis, minus prominenter lobatis vel integris ; fructu maturo ignoto. Type: Herb. Calif. Acad. Sci. No. 248447, collected on sand hills near La Quinta, Riverside County, California, by Lewis S. Rose, No. 36835, Dec. 29, 1936. Other collections which have been seen from Riverside County, California, are: Point Happy near Indio, Jaeger in 1925 (Herb. Univ. Calif.) ; Coachella Valley, Winblad in 1937 (Herb. Calif. Acad. Sci.) ; Painted Canyon, McGregor No. 752 (Herb. Dudley.). One collection has been seen from Arizona, Monnet No. 1103, from dunes between Wellton and Dome, Yuma County (Herb. Calif. Acad. Sci.) ; and one has been seen from Sonora, Mexico, Pringle in 58 LEAFLETS OF WESTERN BOTANY _ [VOL. II, NO. 4 1884, on sandy plains near the Gulf of California (Herb. Univ. Calif.). From typical S. longirostris, var. derelicta is very distinct with its non-glaucous shoot and its pinnately parted leaves. So different is it in appearance that it was believed to be specifi- cally distinct but no characters of flower or fruit were noted that separated it. When mature fruit of the variety is found, the character of the silique or seed may supply the character needed for specific differentiation. Abs ? 7 From the east slope of the Sierra Nevada below the summit of Sonora Pass, Mr. Lewis S. Rose recently collected a variant of Chrysothamnus nauseosus ( Pall.) Britt. with leaves as wide as or wider than in any of the numerous forms of that complex that have been made known. By reference to the treatment of this species by Hall and Clements (Carn. Inst. Wash. Publ. 326 : 209—229), Mr. Rose’s collection was found to be most nearly related to C. nauseosus subsp. bernardinus (Hall) Hall & Clements. From that subspecies, whose entire distribution is far removed from the locality where our plant is found, the Sonora Pass plant differs in its wider 3-nerved leaves and acute involucral bracts. In general aspect, it bears a striking resemblance to the broad-leaved form of C. nauseosus subsp. speciosus ( Nutt.) Hall & Clements which is found in the same general region on the east side of the central Sierra Nevada and which is treated by Hall and Clements as minor variation no. 58 (J. c., p. 221). From this, our plant can be separated not only by the broader 3-nerved leaves but also by the involucres which are a quarter to half again as long. Chrysothamnus nauseosus (Pall.) Britt. var. macrophyllus J. T. Howell, var. nov. Frutex circa 7 dm. altus; caulibus ex basi humili lignoso, erectis simplicibus foliatis ; pubescentia caulis subcompacta griseo-viridula ; foliis oblanceolatis, 4—6.5 cm. longis, 46.5 mm. latis, acutis, attenu- atis basi, tenuiter tomentosis vel subvillosis trinervatis, costa conspicua, nervis lateralibus submarginalibus; involucris 12—14 mm. longis, bracteis 5-seriatis carinatis subglabris acutis; corolla 1 cm. longa, lobis 1—1.3 mm. longis, tuba tenuiter pubescenti; ramis styli 3 mm. longis, appendice 2 mm. longa; acheniis pubescentibus. Type: Herb. Calif. Acad. Sci. No. 248449, collected by Lewis S. Rose, No. 37684, 6 miles east of summit of Sonora Pass, elevation 7400 ft., Mono County, California, Sept. 28, 1937. ’ NOVEMBER, 1937] A COLLECTION OF DOUGLAS’ PLANTS—I 59 A COLLECTION OF DOUGLAS’ WESTERN AMERICAN PLANTS—I BY JOHN THOMAS HOWELL In this journal I recently gave a short account of a collection of plants which I received on loan from Leningrad for determi- nation, plants collected by the Russians in California nearly a hundred years ago.1 It was pointed out that that collection, though of local historical interest, did not possess the scientific interest it would have if it had been critically studied and reported on when it first arrived at the Academy of Sciences in St. Peters- burg in 1842. Incontrast, what deep feelings of scientific interest and regard were aroused by the small set of duplicates of western American specimens collected by David Douglas* which were included for determination in the same loan from Leningrad! Here was a collection classical in the botany of California and the West, a collection containing many species of plants which had been named by Douglas himself or by eminent botanists who had studied his specimens. So fundamental is a consideration of any of Douglas’ collections for an adequate understanding of many of our western American species, it seems proper to pub- lish here the complete list of this botanically rich collection which is to be found in the Russian Academy of Sciences in Leningrad. Of Douglas’ plants there were 90 specimens, 17 from north- western America and 73 from California. The former carried the printed label: “Herb. Hort. Soc. Lond. America Boreali- occidentalis. D. Douglas”; and the latter were labelled: “Herb. 1 Leafl. West. Bot. 2:17—20 (1937). 2 A short biographical notice with such dates as indicate Douglas’ move- ments in the West seems appropriate at this point. David Douglas was born at Scone, Scotland, in 1798. He made two trips to western America as collector for the Royal Horticultural Society of London and not ont did he obtain botanical specimens but he collected seeds of attractive plants for cultivation by members of the society. Many of our most beautiful western wild flowers found their way to England through Douglas’ activity and there they have been cherished as garden plants to the present time, receiving that regard which they deserve but which they rarely receive in their native land. On his first trip, Douglas arrived at Fort Vancouver on the Columbia River in 1825 and remained in the Columbia River country until 1827. On his second trip to America, Douglas turned his attention to California, arriving at Monterey on De- cember 22, 1830 (acc. Jepson, Madrofno 2:97), and remaining until August, 1832, the first botanical explorer in California to remain so long. On leaving California, Douglas went to Hawaii and from there he returned to the Pacific Northwest, landing on the Columbia River in October, 1832, where he was until October, 1833. In November, 1833, he was again in California (cf. Madrofio 2:98) on his way to Hawaii where he was killed by an infuri- ated bull on July 12, 1834. ‘hb. 60 - LEAFLETS OF WESTERN BOTANY [VoL. II, NO. 4 Soc. Hort. Lond. Nova California. Douglas, 1833.”% In spite of the fact that many of the specimens are duplicates of the type collections of certain species, almost none carried a determination and only a few had the generic name written on the label. This would seem to indicate that the specimens were distributed by the Royal Horticultural Society very early and that the set sent to St. Petersburg was sent in the same condition as those sent to the English herbaria, to Asa Gray, to DeCandolle, and to others. In the Russian set nearly all the specimens are numbered but the numbers were undoubtedly assigned in a taxonomic sequence for the distribution and not according to any order the specimens might have had when they were originally collected. So that no specially interesting or valuable specimen would be overlooked, considerable care was taken in the identification of the Douglas collection. In some cases it was possible to deter- mine that a certain specimen was part of the type collection of a species, definitely cited in the literature. In other cases it was not clear how much the specimen from the wild had figured in an original description and how much a plant grown from seed of Douglas’ collecting had been used. Whenever one of the specimens seemed to have anything to do with a plant name based on Douglas’ material, special care was taken to verify the determi- nation and to check the specimen with the original description of the plant. Then when such a determination was established, the specimen was compared with the specimens of the species _ in the herbarium of the California Academy of Sciences, and if one could be found which corresponded to the Douglas collection under consideration, the Academy specimen was annotated to the effect that it agreed with the type collection of the species in the Russian Academy. The purpose of making and noting these comparisons was twofold. First, it seemed desirable to have for reference in the Academy herbarium specimens which have been compared with authentic specimens of the species in question. These annotations have already proved of great assistance inthe critical study of several puzzling species of western American plants and notes 3 By reference to the sketch given in the preceding footnote, it will be seen Douglas spent scarcely any time in California in 1833. This date repre- sents not the year in which the specimens were collected but the year in which Douglas’ Californian collections were received in England. Similar discrepancies are known to occur with some of Douglas’ plants from the Pacific Northwest. NOVEMBER, 1937] A COLLECTION OF DOUGLAS’ PLANTS—I 61 on these are given in the following enumeration of the Douglas plants. Second, in searching out what appeared to be a close match for a Douglas plant, an attempt was made to fix the geo- graphic region in California from which the particular Douglas plant might have come. No detailed field data accompany Douglas’ Californian collections and very little is definitely known about his journeys during his Californian sojourn.* By this study it was possible to place rather definitely several of the more important specimens in the collection. Most of the studies and determinations were made by the writer. The specimens of Potentilla were referred for deter- mination to Miss Ethel Crum, the specimen of Delphinium variegatum was examined by Mr. Joseph Ewan, and that of Downingia pulchella by Dr. Robert F. Hoover. For the oppor- tunity to study this collection, the writer is indebted to Dr. Ivan M. Johnston. SPECIMENS FROM NorTHWEST AMERICA No. 142. Crematis Douctasi1 Hook. This Douglas col- lection, which is in fruit, is probably not a part of the collection from which Hooker drew the original description. That collection was in flower and is well represented by tab. I in Fl. Bor. Amer. vol. 1. Compared to specimens in Herb. Calif. Acad. Sci., the specimen from Leningrad agreed most nearly with a collection from near Austin, Grant County, Oregon, Henderson No. 5451. However, in the Douglas collection the leaf-segments were a trifle more acute and not so noticeably impress-veined on the upper side. Clematis Douglasu is treated by Piper (Fl. Wash. 266) as a synonym of C. hirsutissima Pursh. No. 144. Berrperts AQUIFOLIUM Pursh. In flower. No. 134. Horketia Hrrsuta Lindl. ? Miss Ethel Crum, who critically examined the Douglas collection, was not sure of its identity. She writes about it as follows: “The Douglas specimen from Leningrad may be Horkelia hirsuta Lindl. of which I have seen no authentic material. The inflorescence, however, is more open than originally described for that species; also, the radical stipules disagree with the 4 As recounted by Douglas in his letter to Hartnell (Madrofio 2:98—99), his field journal giving an account of his Californian collections was lost in the Fraser River in 1833. From a table of bearings for certain Californian localities sent by Douglas to Figueroa and published in Bancroft’s History of California (3:403, 404), we know at least that his journeys in California took him from Santa Barbara on the south to Sonoma on the north. 62 LEAFLETS OF WESTERN BOTANY [ VoL. II, NO. 4 description in that they are rarely, if ever, forked. The specimen resembles also H. congesta Dougl. ex Hook., but differs from typical material in the more open inflorescence. There may be a difference also in the quality of the pubescence. The pubes- cence of the Douglas specimen is hirsute; that of H. congesta, not mentioned in the original description, was understood by Rydberg to be silky. However, Rydberg does not cite, and apparently had not seen the type collection. “The floral characters of the Douglas specimen agree wat those of the illustration of H. congesta (Bot. Mag. tab. 2880,— 1829).° In the original description of H. hirsuta the floral struc- ture was not described in detail. The type collection was seen by Rydberg, however, and his illustration (Mem. Dept. Bot. Columbia Univ. 2, tab. 77,—1898) is not essentially different from that of H. congesta. “Tt is quite possible that H. congesta and H. hirsuta are not specifically distinct, but this can be decided only by comparing the types.” Photographs of this Douglas collection which is in flower are in Herb. Calif. Acad. Sci. and Herb. Univ. Calif. The type locality for H. hirsuta is given as “California, Douglas” (Bot. Reg. sub tab. 1997) and for H. congesta as “Cape Mendocena and on the low hills of the Umptqua River...” (Bot. Mag. tab. 2880). In gross aspect the Douglas collection corresponds to a collection made in 1936 in the Willamette River Valley near Monmouth, Eastwood & Howell No. 2863, and determined by Miss Crum as “Horkelia hirsuta Lindl. ex char.” No. 152. AsTRAGALUS REVENTUS Gray. Douglas’ specimens furnished the type of this species which over a period of many years was a puzzle and a problem to Gray. Originally noted by Torrey and Gray under Phaca leucophylla in 1840 (Fl. N. Amer. 1: 694), it was not until 1879 (Proc. Amer. Acad. 15:46) that Gray “completely identified” and named the inadequate fruiting material of Douglas. The specimen which we examined had only a single fruit attached but the plant was unmistakable. The leaflets were oblong, obtuse, up to 17 mm. long and 3 mm. wide, subglabrous above, cinereous-strigellous below; the single pod was 2 cm. long and 1 cm. wide at the truncate base. 5 The drawing was prepared from plants ‘‘which flowered in August, 1828, from seeds brought home by Mr. Douglas’’ to the Roy. Hort. Soc. Lond. The seeds may or may not have come from ‘“‘Cape Mendocena’’ or the “Umptqua River.’’—J. T. H NOVEMBER, 1937] NOTES ON CAREX—XI 63 NOTES ON CAREX—XI BY J. W. STACEY Carex sonomensis Stacey, spec. nov. Laxe czspitosa stolonifera, culmis 4—6 dm. altis, erectis, foliis longioribus, basi fibrillosis et brunneis ; foliis 6—10 ex singulis culmis fertilibus, laminis amplis, planis, 5—40 cm. longis, 3—6 mm. latis ; spica terminali gynecandra androgyna vel pistillata, sessili, ovoidea obovoidea vel oblanceolata, 8—10 mm. longa, 4—8 mm. lata, paullum excedenti spicam proximam; spicis pistillatis 3—6, aggregatis supra, oblongis, 7—15 mm. longis, 5—9 mm. latis; squamis ovatis obov- atisve, obtusis, hyalinis omnino, flavescentibus, in medio viridibus et tri- nervatis, costa squama breviore; perigyniis triangulari-lanceolatis, 4—4.5 mm. longis, 1.5—1.75 mm. latis, prominenter costatis utrimque, in rostrum bidentulatum subserrulatum apice hyalinum attenuatis ; acheniis late ellips- oidalibus 1.75 mm. longis, 1 mm. latis, breviter stipitatis, tenuiter longi- rostellatis. Loosely cespitose, stoloniferous, the rootstocks short, blackish, the stolons slender; culms 4—6 dm. high, erect, exceeding the leaves, phyl- lopodic, smooth, very fibrillose and dark brown at base; leaves with well developed blades, 6—10 to a fertile culm, clustered near the base, the blades flat, thickish, light green, 5—40 cm. long, 3—6 mm. wide, slightly involute on edges, long-attenuate to acute, roughened toward the apex; terminal spike gynecandrous, androgynous, or nearly pistillate, obovoid, ovoid, or oblanceolate, sessile or nearly so, 8—10 mm. long, 4—8 mm. wide, little exceeding the next uppermost spike; pistillate spikes 34, the upper closely approximate, the one or two lower distant, little to strongly exsert- peduncled, the spikes oblong in shape, 7—15 mm. long, 5—9 mm. wide; bracts long-sheathing, the blades leaf-like, shorter than the culms, the sheaths tubular, scarcely enlarged upward; scales nearly as wide as but shorter than the perigynia, ovate, obtuse, nearly hyaline throughout with a yellowish tinge toward the center, with a darker greenish three-veined center, the middle vein not extending to the apex; perigynia triangular- lanceolate, 4—4.5 mm. long, 1.5—1.75 mm. wide, not inflated, glabrous, light green or yellowish-green, strongly nerved on both faces, round- tapering at base and substipitate, tapering at apex into a bidentulate, slightly serrulate, hyaline-tipped beak; achenes broadly ellipsoid, sharply angular with concave sides, obscurely nerved as seen under a lens, brownish, short- stipitate, slenderly long-apiculate, jointed with the straight style; stigmas 3, slender, reddish-brown. Type: Herb. Calif. Acad. Sci. No. 246086, Howell & Stacey No. 13042, collected June 6, 1937, at Pitkin Marsh, five miles north of Sebastopol, Sonoma County, California. Another col- lection from the same place in the Herb. Calif. Acad. Sci. is Howell & Stacey No. 12681. The species is named after Sonoma County. 64 LEAFLETS OF WESTERN BOTANY __ [VOL. II, NO. 4 This species belongs to the section Ferruginee Tuckerm. and is related to C. luzulina Olney and C. Lemmonii W. Boott. It is probable that there are one or two more new western species in this difficult section, and a complete key will be drawn up later. In the meantime the following artificial key will differentiate these three related species : Perigynia hyaline-tipped ; midvein of scales not extending to apex. Uppermost spike staminate, strongly overtopping the pistillate spikes; uppermost pistillate spikes not strongly aggregated; scales reddish-brown; achenes obovoid....................------ C. Lemmonii Uppermost spike gynzcandrous, androgynous, or pistillate, little exceeding the strongly aggregated uppermost pistillate spikes; scales almost completely hyaline; achenes broadly ellipsoid AES cial Ee aM 0 NR: ota FN eat ae Ce ee C. sonomensis Perigynia dark purplish-tipped; midvein of scales extending to the apex; the staminate spike not much exceeding the strongly aggregated uppermost pistillate spikes... C. lusulina tA 7 ? Since the large-headed species of Carex introduced along the coast of New Jersey in the pine barrens was determined as Carex Kobomugi Ohwi,' this species has also been detected on the Pacific Coast, introduced on ballast at Portland, Oregon. Two sheets were found in the Suksdorf collection at the State College of Washington at Pullman, Suksdorf’s No. 1270 and 1811. His number 1811 is also in the herbarium of the California Academy of Sciences, San Francisco. 5 id 7 Carex Douglasu Boott was collected by Eastwood and Howell (No. 4269), fourteen miles west of Panoche, San Benito County, California. This is a considerable extension of its former range, as it has been found before in California only along the Sierra Nevada and eastward, and locally in the high mountains of Southern California. In the herbarium of the New York Bo- tanical Garden are several sheets of this species, labeled as col- lected by Torrey in Santa Barbara County. Mackenzie has written on these sheets, indicating that without doubt they bear the wrong label. In the light of this recent collection, it may be likely that Torrey really did collect these specimens somewhere in Santa Barbara County. 1 Leafl. West. Bot. 2:30 (1937). Vot. II LEAFLETS of WESTERN BOTANY ¥ CONTENTS Interesting Western Plants—I Poitier A. Munz New Varieties of Western Plants—II . JoHN THomas Howe. Further Studies in Eriogonum—III SusAN G. STOKES Two New Wallflowers ALicre Eastwoop A Collection of Douglas’ Western American Plants—II . JoHN THomas Howe. New Records of Noteworthy Northwestern Plants . WALTER J. EyYERDAM Botanical Itinerary of Marcus E. Jones . Puiuie A. Munz A New Mimulus. .. . Joun Tuomas Howey Notes on Carex—XII J. W. STAcEY SAN FRANCcIscOo, CALIFORNIA January 24, 1938 PAGE 65 70 72 73 74 78 78 79 80 ¥ LEAFLETS Es re i< : ie ee STE Seg Ee ae ee ae ee numbers, 40c. Address: John Thomas Howell, California Academy of Sciences, Golden Gate Park, San Francisco, California. WESTERN BOTANY ye heh A publication on the exotic flora of California and on the Tf native flora of western North America, appearing about four ¥ a times each year. Subscription price, $1.00 annually; single BT Cited as LEAFL. WEstT. Bor. PUUEEEP TEETER) TPE ETP ERP PEEP) ey INCHES Se ee ee URL ELE AM AU TAAL abaal ee Owned and published by i ; AuicE Eastwoop and JoHN THomMAS HowkLi JANUARY, 1938] INTERESTING WESTERN PLANTS 65 INTERESTING WESTERN PLANTS—I BY PHILIP A. MUNZ Pomona College, Claremont, California ALLIUM CRISTATUM Wats., Proc. Am. Acad. 14: 232 (1879). In my Manual So. Calif. Bot., 86 (1935), I give A. cristatum as occurring in the Providence Mts. of the eastern Mohave Desert in California, and I include under A. Parishii Wats. a series of plants from the Little San Bernardino Mts. More abundant material and further study would now lead me to restrict A. Parishu* to the San Gabriel and San Bernardino mountains, where it grows at elevations mostly above 6000 it., although some- times as low as 4200 ft.- It is characterized by rose-purple perianth-segments mostly 14—16 mm. long. The plants from the Little San Bernardino Mts., on the other hand, have perianth- segments 8—12 mm. long and the perianth is more campanulate. They have the same ovary-crests and other characters as do plants of A. cristatum from Utah and Arizona. The description of this species must be modified somewhat from that in my Manual for perianth-length and for color, which is deeper than I indicated. It grows at lower elevations than does A. Parishii, for example: Keyes Ranch, Little San Bernardino Mts., at 3500 ft., Munz & Johnston No. 5263; Quail Springs, same range at 3500 ft., Munz & Johnston No. 5238; Contact Mine, same range, at 4000 ft., Munz No. 13760; Stubby Spring near Inspi- ration Point, same range, at 5000 ft., Sara Schenck in 1937; Twenty-nine Palms, Jones in 1927; White Tanks, Hitchcock No. 12226 ; and Bonanza King Mine, Providence Mts., at 5000 ft., Munz, Johnston & Harwood No. 4218. In the above references and for all other specimens cited in this paper, it may be understood that herbarium specimens are in the Herbarium of Pomona College. PHORADENDRON JUNIPERINUM Engelm., Mem. Amer. Acad. n. s. 4: 58 (1849). As this species was delimited by Trelease in The Genus Phoradendron (22,—1916), it has been known only from east of California. A collection of it was made on Juniperus utahensis at Keystone Spring, New York Mts., eastern Mohave * Since the above note was written, Mr. Joseph Ewan has presented evi- dence for separating the plants of the San Gabriel Mts. as A. monticola Davidson and restricting A. Parishii to the San Bernardino Mts. (cf. Bull. Torr. Bot. Club 64:509-511,—1937). I have again gone over our material, but I am unable to maintain this segregation. Leafi. West. Bot., Vol. II, pp. 65-80, January 24, 1938. 66 LEAFLETS OF WESTERN BOTANY _ [VOL. II, NO. 5 Desert, San Bernardino Co., California, October 13, 1935, Munz No. 13868. It is certainly distinct from P. ligatum Trel. on J. occidentalis in the absence of the groove at the base of the scales, although resembling it in size and habit. In these last respects it differs decidedly from the much larger P. Libocedri Howell on Libocedrus decurrens. SILENE INFLATA Smith, Fl. Brit. 467 (1800-4). A rather poor specimen of this was sent me in October, 1936, by Mr. Ethelbert Johnson, as “adventitious in orchard near Anaheim,” Orange Co., California. Aquilegia mohavensis Munz, spec. nov. Tab. I, fig. 1—3. Herba perennis; caulibus pluribus, ascendentibus, 4—6 dm. altis, glabris glau- cisque; foliis basalibus infimisque glabris, petiolis 1—2 dm. longis; laminis biternatis aut triternatis, glabris, glaucis; petiolulis 1—4 cm. longis; foliolis distinctis, 1.5—2.5 cm. longis, 1—2.5 cm. latis, cuneatis, lobatis; foliis superioribus aliquanto reductis, illis inflorescentie ad bracteas reductis, simplicibus aut paucilobatis; caulibus ramosis; floribus nutantibus ; sepalis 7—8 mm. longis, porrectis, elliptico-ovatis, cum marginibus rubidis, ciliatis ; laminis petalorum subflavis, subtruncatis, 3 mm. longis, 4 mm. latis; cal- caribus flavo-rubidis, rectis, 13—15 mm. longis; staminibus flavis, 5—13 mm. longis; antheris 1 mm. longis, flavis; stylis circa 8 mm. longis ; ovariis glanduloso-puberulentibus ; folliculis circa 15 mm. longis, 1.5 mm. crassis; seminibus brunneis. Perennial herb, the stems several, ascending, 4—6 dm. high, glabrous and glaucous through most of their length; basal and lower leaves on peti- oles 1—2 dm. long, glabrous, their blades biternate or triternate, glabrous, glaucous on both surfaces but more strongly so beneath, petiolules glabrous, up to 4 cm. long, those of the second order very fine; leaflets more or less completely separate, mostly 1.5—2.5 cm. long, 1—2.5 cm. wide, cuneate, mostly divided into two portions, each of which is again more shallowly incised with 2 or 3 short obtuse to rounded lobes, upper cauline leaves somewhat reduced, those of the inflorescence becoming simple or few-lobed leafy bracts; stems openly branched above, each ultimate branchlet 5—12 cm. long, glandular-puberulent and ending in a nodding flower; flower 2.5—3 cm. long from tips of spurs to tips of stamens; sepals 7—8 mm. long, spreading, reddish on the margin, greenish yellow in the center, elliptic- ovate, acutish, ciliate; lamina of petals lemon-yellow, rounded-truncate, slightly notched, 3 mm. long, 4 mm. wide, spurs yellowish-red, straight, 13—15 mm. long, swollen at the ends, quite glabrous, slightly divergent; stamens yellow, 5—13 mm. long, anthers 1 mm. long, yellow; styles about 8 mm. long, ovaries glandular-puberulent; follicles about 15 mm. long, 1.5 mm. thick; seeds brown. Type, Munz No. 14687, June 3, 1937, from plant grown at Claremont, California, from rosette taken in October, 1935, from Keystone Spring, New York Mts., eastern Mohave Desert, Cali- JANUARY, 1938] INTERESTING WESTERN PLANTS 67 Explanation of Plate. Aquilegia mohavensis Munz: 1, leaf; 2, follicles; 3, flower. Petalonyx Gilmanii Munz: 4, branch; 5, sepal; 6, petal 68 LEAFLETS OF WESTERN BOTANY __[VOL. II, NO. 5 fornia (Pomona College Herbarium No. 228,694; isotypes at California Academy, Gray Herbarium, University of California). The proposed species would fall in the Rhodanthe of Pay- son’s revision (Contrib. U. S. Nat. Herb. 20: 138,—1918). In the key there proposed, it would not fall easily into either group: those with “Sepals horizontally spreading or reflexed, usually equaling or exceeding spurs,” or those with “Sepals slightly spreading or erect, shorter than the spurs,’’ since it has sepals horizontally spreading and shorter than the spurs. If in the first group it would come with those with distinct lamina, but differs from them (A. wawawensis, A. formosa and subspecies, and A. Shockleyi) by its decidedly smaller flowers and glabrous narrow leaflets. Of these, it would seem to be nearest to A. wawa- wensis, but has spreading, not reflexed sepals and sepals distinctly shorter than the spurs. From the second group in Payson’s key (A. canadensis, A. elegantula, A. desertorum, A. rubicunda, A. triternata, and A. Skinner1) it differs by the more spreading sepals and smaller flowers. In flower-size it seems nearest to A. rubicunda, but that plant is described as viscid-puberulent throughout, instead of glabrous and glaucous, and has glabrous instead of pubescent ovaries. Aquilegia formosa Fisch. var. czlifax (Payson) Munz, comb. nov. A. formosa subsp. celifax Payson, I. c., 144. This plant of the Charleston Mts. and adjacent Nevada grows in the Panamint Mts. of Inyo Co., California, in Thorndyke’s Canyon at 7400 ft., near the upper end of Wildrose Canyon, Munz No. 14858. . STREPTANTHUS CORDATUS Nutt. ex T. & G., Fl. N. Am. 1:77 (1838). In Southern California known heretofore from the New York Mts. It occurs also in the Panamint Mts.: Thorndyke’s, Wildrose Canyon, at 7000 ft., Munz No. 14845; and in Wood Canyon at 6000 ft., M@. French Gilman No. 2479. CAULANTHUS GLAUCUS Wats., Proc. Am. Acad. 17: 364 (1882). Previously reported from Nevada, and in California from Bishop and the White Mts. Collected in the Grapevine Mts., Death Valley, at 6000 ft., M. French Gilman No. 2518 and 2520. PsoRALEA CALIFORNICA Wats., Proc. Am. Acad. 12:251 (1877). Known in Southern California from Mt. Pifios and the San Bernardino Mts. ; also from Baja California. Locally abun- oO JANUARY, 1938] INTERESTING WESTERN PLANTS 69 dant in the public camp ground at Nightingale’s, on the Palms to Pines Highway at about 4000 ft., Santa Rosa Mts., Riverside Co., Munz No. 15104. ASTRAGALUS BRAUNTONII Parish, Bull. So. Calif. Acad. Sci. 2:26 (1903). There has been some question as to the habitat of this species of which very few herbarium specimens exist. On May 10, 1937, Mr. Rupert Barneby wrote me as follows: “There are two very fine stands of this magnificent species on firebreaks above Santa Monica. It is a robust five-foot perennial with (in older plants) a thick woody trunk, from which the year’s flowering branches rise. It appears to prefer the very driest situations.” ASTRAGALUS MALACUs Gray, Proc. Am. Acad. 7 : 336 (1868). Known heretofore in California from Owens Valley northward and from adjacent Nevada. Collected on dry limestone hills about five miles south of Barnwell, eastern Mohave Desert, San Bernardino Co., California, at 4500 ft., May 4, 1935, Munz No. 13711. Petalonyx Gilmanii Munz, spec. nov. Tab. I, fig. 46. Frutex ramo- sus et cinereus usque ad 1 m. altus et latus; caulibus junioribus pilosis, cum capillis mollibus et 0.5—1 mm. longis; foliis sessilibus, deltoideo-ovatis, subintegris, undulatis, subcordatis, pilosis; foliis przcipuis 10—20 mm. longis, 8—13 mm. latis ; superioribus 4—7 mm. longis ; spicis densis, sessili- bus, 1—3 cm. longis; bracteis subcordatis, sessilibus, 4—6.5 mm. longis, pubescentibus; sepalis membranaceis, lanceolato-linearibus, 2 mm. longis, pubescentibus ; petalis albis, 3—4 mm. longis, unguiculis tubulatis, 1.5—2 mi. longis, partibus extensis triangularibus ; staminibus 5 mm. longis; stylis 6 mm. longis; capsulis cylindraceis, dense pubescentibus, 2 mm. longis. Diffusely branched cinereous shrub, apparently rounded and up to 1 m. tall and broad; young stems densely spreading-pilose, the hairs up to 1 mm. long and fairly soft ; leaves sessile, deltoid-ovate, subentire but some- what wavy, subcordate at the base, somewhat elongate-acute to obtuse at apex, pilose above and below, the hairs below somewhat stiffer than those above, those of the margins quite stiff, making a scabrous margin; leaves of main branches 10—20 mm. long, 8—13 mm. wide; those of finer upper branches reduced, 4—7 mm. long and about as wide; spikes dense, not peduncled, 1—3 cm. long, the bracts subcordate, sessile, 4—6.5 mm. long, greenish when young, straw-colored in age, rather thin, stiff-pubescent ; sepals membranous, lance-linear, 2 mm. long, pubescent; petals white, 3—4 mm. long, pubescent without, the tube-like claws connivent, scarcely 2 mm. long, the spreading triangular portion 1.5—2 mm. long; stamens all fertile, 5 mm. long, anthers about 0.6 mm. wide; styles 6 mm. long; cap- sules cylindrical, densely pubescent, about 2 mm. long. 70 LEAFLETS OF WESTERN BOTANY [ VOL. II, NO. 5 Type from Ryan Wash, Death Valley, Inyo Co., California, at 1500 ft. altitude, May 20, 1937, M. French Gilman No. 1568, Pomona College Herbarium No. 228,696; isotype fragment at California Academy of Sciences. Differentiated from P. nitidus and P. Parryi, both of which it resembles in its broad, sessile, subcordate leaves, as well as by the much smaller flowers and longer and finer hairs. It differs from P. Thurberi in the more upright habit with the general appearance of having more slender growth, by the finer hair, by the broader and more entire leaves, and by the smaller flowers with shorter stamens. It differs from P. linearis, with which it agrees in the small size of the flowers, by the more triangular leaves and smaller floral bracts, and from P. crenatus by the smaller flowers and wider connivent petals. It is a pleasure to dedicate this species to its discoverer, Mr. French Gilman, who has in recent years done much to add to our knowledge of the plants of the Death Valley region. NEW VARIETIES OF WESTERN PLANTS—II BY JOHN THOMAS HOWELL In that rugged and mountainous region of the South Coast Ranges in southern San Benito County and adjacent Monterey County in California, there is to be found a variety of Arcto- staphylos glauca Lindl. which seems worthy of recognition. In typical A. glauca, all parts are glabrous, but in the variety here indicated, the young stems, leaves, and bracts and branches of the inflorescence are more or less finely hairy. The distinctions may appear trivial, but, when considered in the line of criteria used to distinguish entities in Arctostaphylos, they are sufficiently important. Moreover, the variety would appear to have geo- graphic significance, correlated perhaps with certain elements of desert origin to be found in the region. Arctostaphylos glauca Lindl. var. puberula J. T. Howell, var. nov. A specie differt: ramis foliis et inflorescentiz bracteis ramulisque pubes- centia brevi velutina vestitis. Type: Herb. Calif. Acad. Sci. No. 188525, collected by the writer on Sept. 7, 1931, 4 miles northwest of Hernandez, San Benito County, California. Other collections which have been JANUARY, 1938] | NEW VARIETIES OF WESTERN PLANTS yx! seen are: Bear Valley and the Pinnacles, Dr. C. Hart Merriam; 14 miles northwest of Priest Valley on Lewis Creek, Monterey County, L. S. Rose No. 36293, Eastwood & Howell No. 2480. 4 z, 7 A near relative of the rare and very local Cirsium fontinale (Greene) Jepson has recently been found near San Luis Obispo, California, about 200 miles to the south of the only known station for C. fontinale. The situation where the southern thistle grows is quite similar to the restricted habitat of the northern: around seepages or springs in moist or wet clay soil overlaying serpen- tine. The southern plant was found on the more gentle slopes and terraces which extend up to the rugged mountains of serpen- tine about the headwaters of Chorro Creek. Certain differences between the northern and southern plants would indicate that the latter is a variety. The southern plant is more obviously arachnoid-tomentose, and the underside of the leaves is nearly white-lanate. The tangled woolly hairs almost or quite conceal the shorter stubby trichomes which impart to the pubescence of C. fontinale its mealy appearance. The bracts of the involucre are also some- what different, in the variety the middle bracts are rarely if ever drawn out into the slender lanceolate tips that in the species exceed the budding heads. Also the bracts are usually less pubescent in the variety than in the species. The achenes of the species are light brown and the rim surrounding the apical hollow is thin; in the variety the achenes are a trifle more turgid with sides more prominently angled, the color is tinged with purple, and the apical rim is somewhat thicker. In the species the achenes are smooth and in the variety they are minutely and sparsely roughened near the apex. Cirsium fontinale (Greene) Jepson var. obispoense J. T. Howell, var. nov. Caulibus foliisque plus minusve tomentosis, foliis sublanatis subter; bracteis involucri ovato-lanceolatis usque ad ovatis, subglabris; acheniis turgidis, minute scabris, purpurascentibus, utrimque angulatis, ora apice paullum crassa. Type: Herb. Calif. Acad. Sci. No. 249328, collected on a boggy flat near serpentine, Chorro Creek, San Luis Obispo County, California, by Eastwood and Howell, No. 2218, May 6, 1936. 72 LEAFLETS OF WESTERN BOTANY [ VOL. II, NO. 5 FURTHER STUDIES IN ERIOGONUM—III BY SUSAN G. STOKES Eriogonum Howellii Stokes var. subracemosum Stokes, var. nov. A specie differt: habitu pumilo, caulibus ligneis, ramulis inflorescentiz gracilimis, elongatis, racemosis, nodis ultimis ramulos minimos steriles ferentibus ; involucris et floribus. parvis. Low shrub with slender branches and divaricately branched inflores- cence, the year’s growth about 5 cm. long, tomentose; leaves scattered, lanceolate, tomentose, about 1 cm. long, margins revolute, petioles appressed to stem, about as long; inflorescence widely branched, common peduncle short, internodes about 1 cm. long or less, final divisions racemose, epi- dermis crusted and papillate; involucres minute, campanulate, yellowish; flowers pale, less than 2 mm. long, glabrous, outer segments broad, inner narrower. Type: No. 240217, Herb. Calif. Acad. Sci., Little Colorado River Gorge, 17 miles west of Cameron, Coconino County, Arizona, collected by Kearney and Peebles, No. 12818, Sept. 26, 1935. Eriogonum Howellii Stokes var. argense (Jones) Stokes, comb. nov. EF. sulcatum Wats. var. argense Jones, Contrib. West. Bot. 11:15. The specimen from the Argus Mts. is not sulcate but is strongly papillate. It is therefore moved to E. Howelli. Kerr’s No. 73, from Independence, Inyo County, California, is included in the var. argense. According to Mr. Wheeler, who examined the type for me in the Gray Herbarium, EF. sulcatum Wats. is sulcate and the epidermis not at all papillate. Eriogonum annuum Nutt. subsp. chihuahuaense Stokes, subsp. nov. A specie differt: foliis linearibus revolutis, 2—4 cm. longis. Type: No. 195924, Herb. Calif. Acad. Sci., collected by M. E. Jones, Sept. 28, 1903, at Colonia Juarez, Chihuahua, Mexico, in the Sierra Madre Mts., at an elevation of 5200 ft. Eriogonum czspitosum Nutt. subsp. Douglasii (Benth.) Stokes var. sublineare Stokes, var. nov. A subspecie Douglasii differt : caudice fruti- culoso apertiore, pedunculis altioribus, foliis longioribus angustioribus. Low shrub with slender woody stems, twigs numerous, bearing clusters of narrow leaves; leaves about 2.5 cm. long when fully grown, a little silky-tomentose ; peduncles 1—2 cm. long, with leaf-like bracts near middle; involucres solitary, lobes reflexed; flowers pubescent. Type: No. 215024, Herb. Calif. Acad. Sci., collected by J. William Thompson, No. 8256, May 21, 1932, among sage- brush of rim rocks, Rattlesnake Hills, Ellensburg, Kittitas County, Washington. JANUARY, 1938] TWO NEW WALLFLOWERS 73 TWO NEW WALLFLOWERS BY ALICE EASTWOOD Erysimum filifolium Eastwood, spec. nov. Caulis viridis, simplex 3—4 dm. altus, striatus, pubescens paucis villis furcatis; foliis filiformi- bus 3—10 cm. longis, 0.5 mm. latis, integris vel minute dentatis ; racemis floralibus capitatis, racemis fructuosis elongatis, pedicellis divaricatis fili- formibis 5—10 mm. longis; calyce 1 mm. longo, 2 sepalis exterioribus basi saccatis, 4 mm. latis, 2 sepalis interioribus 2 mm. latis, apice obtusis et callosis, margine membranaceis; corolla primo aurantiaca in senectute aurea; laminis petalorum orbiculatis, 5 mm. diametro, unguiculis margi- natis, 1 cm. longis; antheris sagittatis, filamentis planis ; capsulis immaturis 4-angulatis, scabro-pubescentibus, stigmate lato sessili. Biennial or winter annual from a slender tap-root, stem bright green, simple, 3 or more dm. high, ribbed, pubescence scanty, of flat 2-branched hairs; leaves filiform 3—10 cm. long, less than 1 mm. wide, entire or with scattered minute teeth; racemes in flower densely capitate, in fruit elongat- ing with filiform spreading pedicels 5—10 mm. long; calyx about 1 cm. long, 2 outer sepals saccate at base, 4 mm. broad, 2 inner 2 mm. broad, with a darker obtuse thickened apex, margin membranous; corolla at first orange becoming yellow, petals with orbicular blade 5 mm. across, taper- ing to a membranously margined claw 1 cm. long; stamens with sagittate anthers and flat filaments; immature pods 4-sided, rough-pubescent, with broad sessile stigmas. Type: No. 77723, Herb. Calif. Acad. Sci., collected April 1, 1914, at Glenwood, Santa Cruz Co., California, by the late Honorable Horace Davis. It differs most noticeably from other species or varieties of Erysimum in the simple stems and filiform leaves. Another specimen which seems to be similar is from the collection of A. D. Elmer, No. 5034, collected at Mocho Creek, Alameda Co., California. It also has the large flowers, simple stems, filiform leaves, and the petiole bases of dead leaves at the base of the stem. Erysimum moniliforme Eastwood, spec. nov. Caulis basi et supra ramosus, circa 5 dm. altus, pallido-viridis, striatus, pubescens villis fur- catis appressis, foliis lineari-lanceolatis, 4—5 cm. longis, 5—8 mm. latis, integris vel foliis infimis pauci-dentatis, apice acutis et callosis; racemis ramos terminantibus, in fructibus elongatis, capsulis sparsis, pedicellis gra- cilibus, 3—5 mm. longis; calyce circa 1 cm. longo, sepalis exterioribus basi saccatis, apice acuminatis, 2 mm. latis, interioribus angustioribus, apice obtusis; corolla citrina; petalorum laminis orbicularibus, 6 mm. latis, un- guiculis marginatis, 1 cm. longis, capsulis gracilibus divaricatis, 8 cm. longis, submoniliformibus, apice attenuatis ad stylum crassum, stigmate lato, vix 2-lobato ; seminibus oblongis, 4mm. longis, 1 mm. latis, apice alatis, cotyledonibus accumbentibus. 74. LEAFLETS OF WESTERN BOTANY [ VoL. II, NO. 5 Biennial or winter annual from a tap-root, stems branching from the base and above, about 5 dm. high, pale green throughout, somewhat striate, pubescence of flat appressed 2-branched hairs; leaves linear-lanceolate, entire or with a few teeth on the basal leaves, 4—5 cm. long, 5—8 mm. wide, acute and callous-tipped; racemes terminating each branch, lengthening in fruit with scattered pods, pedicels.slender, 3—5 mm. long; calyx about 1 cm. long, outer sepals saccate at base, 2 mm. wide, apex acuminate, inner narrower with obtuse apex; corolla yellow, blade of petals orbicular, 6 mm. wide, claw margined, 1 cm. long; pods slender, submoniliform, divaricate, spreading, some 8 cm. long, tapering to a stout style, tipped by a broad stigma indistinctly 2-lobed; seeds in one row, oblong, 4 mm. long, 1 mm. wide, tipped with a wing 1 mm. long, cotyledons accumbent. Type: No, 139529, Herb. Calif. Acad. Sci., collected by the author at Alcalde, Fresno Co., California, April 1, 1926, No. 13569. Also collected in Oil Canyon, Fresno Co., March 18, 1931, by J. T. Howell, No. 5850; and by Eastwood and Howell in the Temblor Range above the Carrizo Plains, No. 4113, April 29, 1937. It differs from other species or varieties in the slender pods somewhat constricted between the oblong seeds and the prevail- ingly entire leaves. The lemon-yellow flowers and branching habit differentiate it from EF. californicum Greene, the type locality of which is Mt. Diablo where conditions are dissimilar to the almost desert localities where this is found. A COLLECTION OF DOUGLAS’ WESTERN AMERICAN PLANTS—II BY JOHN THOMAS HOWELL No. 150. AstracALus SoNNEANUS Greene. A. Hookeri- anus (T. & G.) Gray, non Dietr. The specimen from Leningrad represents part of the original collection and is in fruit. In Herb. Calif. Acad. Sci. there was no specimen which corresponded exactly with the Douglas specimen which was originally col- lected in the “interior of Oregon, probably near the Rocky Moun- tains,” but those plants from northern California which are referred to Astragalus siskiyouensis (Rydb.) Thompson seemed to be nearest. The following notes were taken from Douglas’ collection : leaflets broadly oblanceolate to linear-oblong, 1.5 x 5 mm. to 3 x 8 mm., obtuse to acute, strigellous-pubescent ; pod oblanceolate-obovate, to 5.5 cm. long and 2.5 cm. wide, obtuse, narrowed at base into a short stipe 3 mm. long. JANUARY, 1938] | COLLECTION OF DOUGLAS’ PLANTS 75 It is to be regretted that this plant, which was originally described as Phaca Hookeriana, must, in Astragalus bear a name carrying neither historic nor geographic connotation. By those who do not follow the specific segregation proposed by Rydberg for this group, our plant would be called A. Whitneyi Gray var. Sonneanus (Greene) Jepson (FI. Calif. 2: 347). No. 148. AsTRAGALUS STENOPHYLLUS T. & G. The speci- men from Leningrad carried flowers and very young fruit. No. 151. AstTRAGALUS SUCCUMBENS Dougl. ex Hook. With the original description, Hooker cites two places where Douglas collected this species, “‘on the barren grounds of the Columbia, and near the Wallawallah River” (Fl. Bor. Amer. 1:151). It is not known with which of the collections the specimen from Russia corresponds. In nearly all of the material of A. suc- cumbens in Herb. Calif. Acad. Sci., the leaflets are obovate and rounded or retuse at the apex, but in the Douglas collection the leaflets are conspicuously acute or mucronulate. This character found correspondence in part of M. E. Jones’ collection of 1902 from Umatilla, Oregon: part of the collection has leaflets broader and rotund-obtuse, part corresponding most nearly to the Douglas plant has leaflets frequently tapering to more or less pointed tips. At the appropriate season, the “barrens of the Columbia” are still beautified by the masses of attractive light pink flowers of this species (cf. Eastwood & Howell No. 3489). No. —. KENTROPHYTA IMPENSA (Sheldon) Rydb. This is undoubtedly the plant reported by Torrey and Gray as K. mon- tana Nutt. in T. & G. (Fl. N. Amer. 1: 694). It still remains too little known. (Cf. St. John, Fl. SE. Wash. 219.) The label was one of the few which carried no number, but in all other respects was like the labels of the northwestern plants. No. 149. Lupinus ornatus Dougl. ex Lindl. This lupine, abundant through the desert borders of the Columbian region, is noted in the original description from several locations where Douglas found it (Bot. Reg. tab. 1216). It is not possible to say to which of Douglas’ collections the specimen from Leningrad belongs and in the herbarium it was found in general to compare favorably with the specimens passing under this name. The Douglas specimen, which carries buds, flowers, and very young fruits, however, does not agree well with the plate which ac- 76 LEAFLETS OF WESTERN BOTANY _ [VOL. II, NO. 5 companies the original description, probably because the drawing was taken from cultivated plants. In C. P. Smith’s key to the genus in Oregon, L. ornatus is distinguished by “banner usually pubescent on the back” (Contr. Dudley Herb. 1:14), but in the Douglas collection the banner is glabrous except for a few hairs along the median line. The keel is ciliate along its entire length or almost to the tip. The inflorescence is borne on a short peduncle and is rather closely subtended by the upper leaves. No. 155. Lupwicia patustris (L.) Ell. This species is in Fl. Bor. Amer. (1: 215) as Jsnardia but Hooker does not cite Douglas’ collection. No. 165. GENTIANA CALYcoSA Griseb. in Hook. Although Grisebach was the first to study and report on certain collections of Gentiana made by Douglas, he does not record having seen Douglas’ collection of this species. No. 146. GiILiaA MINUTIFLORA Benth. in DC. The Russian specimen of the type collection of Gilia minutiflora found a good match in Suksdorf’s collection from Bingen, Washington, No. 7146. No. 169. LinaNTHUs BICOLOR (Nutt.) Greene. Neither in his treatment of Douglas’ Californian collections of Polemom- acee in 1833 (Bot. Reg. sub tab. 1622) nor in his review of the family in 1845 (DC. Prodr. 9:302—322) does Bentham treat of this plant which was not described until even later by Nuttall in 1847 (Journ. Phila. Acad. n. ser. 1: 156). No. 167. NAVARRETIA INTERTEXTA (Benth.) Hook. Doug- las’ collection appeared to match in every detail of habit, leaves, inflorescence, and pubescence the specimen in Herb. Calif. Acad. Sci. collected by E. Hall on Silver Creek,® Oregon, No. 421. Since it was from plants collected by Douglas that the species was originally described, an effort was made to determine if these Douglas specimens were a part of the type collection, because the specimens of Douglas and of Hall are more or less different from most of the Californian plants which pass as N. intertexta. Localities for the Douglas collection or collections of this species differ in the literature. In his original description of 6 Both Prof. M. E. Peck at Willamette University and Prof. A. R. Sweetser at the University of Oregon believe that Hall’s “Silver Creek” is the one in Marion County which flows down the west slope of the Cascade Mts. into the Willamette River. An Academy specimen of Allocarya which bears Hall’s number 407 was collected at Silverton, Marion County, and substantiates this opinion. Prof. Peck writes that ‘‘the Navarretia ... is common along the eastern slope of the Willamette Valley.” a a? a JANUARY, 1938] COLLECTION OF DOUGLAS’ PLANTS ra 4: gochloa intertexta, Bentham cites “California and North-West America Douglas” (Bot. Reg. sub tab. 1622). At the time Hooker reduces Bentham’s genus to Navarretia, the location for the only collection cited is “New Albion, N. W. America. Douglas” (F1. Bor. Amer. 2:75). And in Bentham’s treatment of the Polemoniacee in 1845, again only a single collection is given, this time, “in herbidis ad flum. Multaomah in America boreali-occid. (Douglas!)” (DC. Prodr. 9: 310). From this it would seem that only a single collection of this Navarretia was eventually considered typical of the species, the one from Oregon. If, however, there is a Californian collection by Douglas in Herbarium Benthamianum at Kew, that specimen must be considered by the one who at length rules on the typifi- cation of the species since “California” is first cited in the original description. Until a really detailed study is made, it cannot be stated how considerable are the differences between the plants of California and Oregon now referred to the species. It may be that the apparent differences will never deserve taxonomic recognition, in which case no great importance will be attached to the question of type locality. But in any case, from the view of studies corre- lating geographic distribution and evolutionary tendencies, the problem should be solved by a critical examination of the original specimen or specimens at Kew. No. 168. OREOCARYA LEUCOPH#A (Dougl. ex Hook.) Greene. So close was the agreement between the original col- lection made by Douglas and that made by Eastwood and Howell in 1936, the two might easily have come from the identical sandy wastes and “arid barrens” along the Columbia River. Eastwood & Howell No. 3520 was collected from wind-swept dunes near Paterson Ferry, Morrow County, Oregon. No. 170. PENstemMon Douctasi1 Hook. The fruiting specimen of the Douglas collection from the Russian Academy adequately represents the original collection of this Penstemon which is treated as a subspecies of P. fruticosus (Pursh) Greene by Pennell and Keck (in herb.). No. 171. PENstemon GarrpDNERI Hook. The original col- lection of this species, also in fruit, found close correspondence in habit and leaves with Henderson’s collection made near Prairie City, Grant County, Oregon, No. 5502. 78 LEAFLETS OF WESTERN BOTANY __[VOL. Il, NO. 5 NEW RECORDS OF NOTEWORTHY NORTH- WESTERN PLANTS BY WALTER J. EYERDAM Seattle, Washington SAXIFRAGA OPPOSITIFOLIA L. In Alaska I have collected this pretty alpine plant on Mitrofania Island (June, 1928) and at Unalaska, Unalaska Island (June, 1932). At the latter locality, about a mile from the village, on the east shore of Illuliak Bay, I found very old plants on the frost-cracked, disintegrating top of a sea cliff that were over two feet square. There are several transitional forms and this one approaches subsp. asiatica (v. Hayck) Engl. & Irmisch., but it is not the same. In the Olympic Mountains, I have taken this species on May 30, 1936, on Mt. Angeles and on August 11, 1937, on Mt. Elinor at about 4500 ft. elevation. On July 27, 1935, I collected this species at 5000 ft. elevation in rock crevices between Silver Lake and Twin Lakes, Monte Cristo district, Snohomish Co., ° Washington. CAKILE EDENTULA (Bigel.) Hook. var. cALIFoRNICA (Heller) Fern. In 1936 Eric Hultén reported this plant as new to the flora of Alaska in his account of “New or Notable Species from Alaska—Contributions to the Flora of Alaska I’ (Svensk Bot. Tids. 30: 522). The report was based on two of my collections from Kodiak Island: Three Saints Bay, August 5, 1931, No. 525; and, Old Harbor, September 11, 1931, No. 640. I have included the Alaskan report of this plant, which ranges far to the south, because the journal in which it is published is not generally accessible to most students of western botany. BoTANICAL ITINERARY OF Marcus E. Jones. There have been deposited manuscript-copies of a journal, giving in some detail the collecting itinerary of the late Marcus E. Jones, in the libraries of Gray Herbarium, United States National Herbarium, California Academy of Sciences, and Pomona College. Botanists having difficulty with the place names used by Mr. Jones on his locality labels will be able by consulting this journal to find the approximate location of such places through reference to dates involved. The journal covers the years from 1875 to 1919.— P. A. Munz. a a JANUARY, 1938] A NEW MIMULUS 79 A NEW MIMULUS BY JOHN THOMAS HOWELL Mimulus cleistogamus J. T. Howell, spec. nov. Humilis, compactus, 0.7—3 cm. altus, 1—3 cm. latus, undique paullum glanduloso-pilosus, glandi- bus capitatis, caulibus et ramis brevissimis et rigidis; foliis linearibus ad oblongo-oblanceolatis vel subovatis, usque ad 2.5 cm. longis et 7 mm. latis, obtusis vel subacutis, infra ad petiolum brevem et latum attenuatis ; floribus sub anthesi circa 2 mm. longis, corolla calyce septa, post anthesin calyce et ovario magnopere accrescenti; calyce fructifero 6—8 mm. longo, denti- bus inzqualibus, obliquis, basi multum gibboso supra, costis et dentibus viridibus, tubo calycis sub sinibus scarioso; fructu oblique ovoideo, rigide coriaceo; seminibus anguste obovatis, sparse lepidotis, apice proximo apiculato. Type: Herb. Calif. Acad. Sci. No. 243436, collected May 3, 1937, on an open hillside 14 miles west of Panoche near the head of the Tres Pinos Creek Canyon, San Benito County, where plants were locally common in fine gravelly clay soil, Eastwood & Howell No. 4267. When this extraordinary little Mimulus was collected, the whole plant seemed to be made up entirely of a few divergent fruits and their subtending leaves since the tiny cleistogamous flowers and the very short stems were quite hidden. So in- conspicuous was the whole thing it would probably have been entirely overlooked if it were not growing closely associated with Gnothera graciliflora H. & A., the condensed subglobose fruiting specimens of which were being collected when the Mimulus was detected. But when once examined, the unique character of the little plant was recognized—a monkey-flower without a face! However, by dissection of a young flower under considerable magnification and with proper illumination the corolla of a Mimulus can be seen, and the minute stamens and pistil are visible within the tiny flower if it is opened at the time of anthesis or just before. After anthesis the corolla and stamens form a tight withered knot that is carried upward as a little cap on the beak of the greatly accrescent fruit. The fruit is rigid and boney- cartilaginous and the seeds bear scattered scurfy processes on the otherwise smooth coats. The character of the fruit and seed definitely relates this little Mimulus to M. Douglasii, M. modestus, M. tricolor and others which Dr. A. L. Grant groups together in Gray’s section 80 LEAFLETS OF WESTERN BOTANY _ [VOL. II, NO. 5 CEnoe. So different from all, however, is M. cleistogamus that a new section is here proposed to receive it, section Cleisanthus.* NOTES ON CAREX—XII BY J. W. STACEY A new record for California is Carex microptera Mackenzie. This species was collected recently by J. T. Howell on the South Fork of Salmon River, 5000 ft., Siskiyou County (No. 13252) and at Big Flat, Trinity County (No. 13573). Both of these numbers are in the herbarium of the California Academy of Sciences. Hitherto its known range was from Alberta to Wash- ington, southward to Utah, Nevada and Oregon. Its presence in northern California was to be expected. Its closest relative is Carex festivella Mackenzie, from which it may readily be differ- entiated by the more echinate appearance of the heads, by the lighter colored scales, and by the narrower perigynia. iw, 7 7 Carex specifica Bailey was collected on Mt. Rose, Nevada, by Lewis S. Rose (No. 37626). This is a new record for Nevada as the species has been previously found only in California. ff v 7? Over sixty years ago Bolander collected a Carex near Mendo- cino City which was named by Olney Carex mendocinensis. Only one collection of this species has been known for all of this time (Bolander No. 4701). Miss Alice Eastwood, J. T. Howell, and I visited the region near Mendocino City May 30, 1937, to rediscover, if possible, this rare species. We were unable to find it, but the next day it was found in a swamp near Point Arena, Mendocino County (Eastwood, Howell & Stacey No. 4466). Growing with it were Carex californica Bailey, C. debiliformis Mackenzie, C. salineformis Mackenzie, C. gynodynama Olney, C. sub-bracteata Mackenzie, C. luzulina Olney, C. obnupta Bailey, and C. phyllomanica W. Boott. w, 7 vf Carex spissa L. H. Bailey was collected by Eastwood and Howell (No. 2217) at Chorro Creek, San Luis Obispo County, California. This has not been reported before, as far as the writer knows, from farther north than Los Angeles County. * Mimulus sect. Cleisanthus J. T. Howell, sect. nov. Pedicellis brevis- simis ; floribus minimis, cleistogamis, corolla calyce breviore ; capsula rigida, cartilaginea ; placentis separatis ; seminibus subobovatis, sparse lepidotis. = ores & Vo. II No. 6 oF wa LEAFLETS of WESTERN BOTANY ¥ CONTENTS PAGE The Perennial Lupines of California—I . . . .... 81 Y Axice EAstwoop ' Interesting Western Plants—II . ........ 87 , Puitie A. Munz ; NE BP TES hs a Meee at eee ee tee : J. W. Stacey 4 The Tobacco Collected by Archibald Menzies on the 3 Northwest Coast of America . . ...... . 92 Auice Eastwoop A Collection of Douglas’ Western American Plants . . . 94 JoHN THomas Howe. A Botanical Visit to the Vancouver Pinnacles . . . . . 97 JoHN THomas Howe. sowo New Scrophulariacee . . ....+ =. +... 104 ALIcE EAstwoop This number published with funds from the California Botanical Club SAN FRANCISCO, CALIFORNIA Apri 16, 1938 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. WEstT. Bort. NSU LIU DUA INCHES PCO TORT ELSA ELS YL Pe METRIC Owned and published by Axice Eastwoop and JoHN THomas HoweLi APRIL, 1938] PERENNIAL LUPINES OF CALIFORNIA 81 THE PERENNIAL LUPINES OF CALIFORNIA—I BY ALICE EASTWOOD Although the annual species of Californian lupines have been studied by C. Piper Smith and the results of his investigations published in several papers, very little has been done among the more difficult perennial lupines. In Jepson’s Flora of California vol. 2, pt. 3, pp. 246—283, synonyms are made of plants described as species, without any explanation of the characters by which their authors differentiated them. In the series of articles which the author proposes to publish from time to time, the original descriptions will be carefully studied and the types when possible. When in England in 1911, the author took photographs and made notes on the types in the Lindley Herbarium at Cambridge, the Royal Herbarium at Kew, and the British Museum Herbarium at South Kensington. The species will be in groups, some groups ecological, some geographical, and some according to natural affinities. Key TO THE LuPINES GROWING NEAR THE SEACOAST OF CALIFORNIA EISEN TE ERED Sh hn a een NESE EORE A NEE RO OR eS OT 2 I UR es Lerectnehin wdmanpcnpmansomoncntiroges 5 ERG ea ee eR L. arboreus Sims EE SS Ee ae ee 3 IE” NUON, SUNG age srinreesteceereva rnveanerescenerwencoboree 4 3. Pubescence not silvery.................-...:000..c0.00 L. rivularis Dougl. ex Lindl. ET EDLY: REL e Ge AER Bi SLES a L. Chamissonis Esch. ET EY SAS RE RE SEE aoe 1 ne L. albifrons Benth. 5. Flowers large, almost 15 mm. long and wide, varicolored and OSIRIS Eat OP Sa SE AE a SO a eee 6 5. Flowers smaller, generally sand-binders..................-.-s---s-cececeseceeeseeeseceeeeneee 7 6. Pubescence silvery appressed........................:--csececeeseeee L. variicolor Steud. I er csc sae veseninnatencivascosene L. eximius Davy 7. Flowers generally verticillate, showy; plants decumbent....................-. 8 7. Flowers verticillate or congested, not showy; plants prostrate............ 9 TE), a a TR L. littoralis Doug]. ex Lindl.* ES LE I EET TE .L. Micheneri Greene DENT RESCOTIOE AUT OSSE ooo cnc Ce Re ae ALIce Eastwoop SAN FRANCISCO, CALIFORNIA NoveMBER 10, 1939 it | 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. He UU UL OM al Ma a ei | peusunnnaggecrapuneygennaperngganengenngycnnepenaggennapennggeccypocagpentaysnngpente ety Owned and published by Atice Eastwoop and JoHN THomaAs HowELi NOVEMBER, 1939] ASTRAGALUS VERSUS OXYTROPIS 209 ASTRAGALUS VERSUS OXYTROPIS BY LOUIS CUTTER WHEELER Department of Botany, University of Missouri Columbia, Missouri The genus Oxytropis A. P. DeCandolle, Astragalogia, 24, 66 (1802) depends mainly on the apiculation of the keel to distin- guish it from Astragalus. But, as noted by M. E. Jones, Re- vision of North American Species of Astragalus, 15 (1923), A. acutirostris S. Wats. and A. nothoxys A. Gray both have acuminate keels. Jones was so impressed by this that he trans- ferred A. acutirostris to Oxytropis, Proc. Calif. Acad. ser. 2,5: 677 (1895). Later, in his Revision, he repudiated this and returned it to Astragalus. There is somewhat of a habital unity in the often scapose character of the species assigned to Oxytropis. However, when examined in detail, this supposed habital unity of the so-called genus Oxytropis fails in two ways. Oxytropis pilosa (L.) DC. (based on Astragalus pilosus L.) has long stems. On the other hand Astragalus mollissimus Torr. bears such a strong habital resemblance to Oxytropis Lamberti Pursh, a species with the habit considered typical of Oxytropis, that it may be more than a coincidence that these two species share the distinction of being the two worst “loco-weeds.” Likewise, Astragalus alpinus L. bears a strong resemblance, especially in more cespitose indi- viduals, to Oxytropis foliolosa Hook. Hegi, Syn. FI. Mittel- Europa 4(3) : 1402 (1923),* in defiance of tradition, treated Oxy- tropis DC. as a synonym of Astragalus L. Likewise Tidestrom, Proc. Biol. Soc. Wash. 50: 19 (1937), has rejected Oxytropis and transferred several species to Astragalus. Unfortunately he did so without ascertaining whether the names were pre- occupied. Consequently a few more homonyms were added to Astragalus. I shall rename only the one familiar to me, which is: Astragalus Munzii L. C. Wheeler, nom. nov.; based on Oxytropis oreophila A. Gray, Proc. Amer. Acad. Arts & Sci. 20: 3 (1885). Type: Aquarius Plateau, Utah, altitude 10,500 feet, Aug. 9, 1875, L. F. Ward No. 541 (Gray Herbarium!). 1 Date according to Becherer,‘‘Les dates de publication de flore de l'Europe Centrale,” Candollea 5: 344 (1934). Leafi. West. Bot., Vol. II, pp. 209-216, November 10, 1939. KR a 1, a Ry Ys SOTA ot} GARUB 210 LEAFLETS OF WESTERN BOTANY [voL. II, NO. 12 Astragalus oreophilus (A. Gray) Tidestrom, Proc. Biol. Soc. Wash. 50: 19 (1937); not A. oreophilus (Phil.) Reiche, Anal. Univ. Chile 97 : 561 (1897) ; nor A. oreophilus Rydb., Bull. Torr. Bot. Club 31: 561 (1904). The specimen chosen as type is selected for the reason that, of the several referred by Gray to this species, it is the best speci- men. The Californian specimens differ but little except that they are generally more compact, as would be expected from their habitat on the bleak alpine summit of San Gorgonio Peak, San Bernardino Mountains. It is highly probable that Oxytropis oreophila was published in 1884. The paper which included it was communicated May 14, 1884, and the next paper (printed with it) was communicated June 11, 1884. The articles in the next series by Gray were com- municated Oct. 8 and Dec. 10, 1884, and, according to the title page of the reprint, the series was issued Jan. 26, 1885. Unfor- tunately there is no title page for the separate, at Gray Herba- rium, of the first series, and the whole volume of the Proceedings of the American Academy of Arts and Sciences bears only the date 1885. NEW PLANT RECORDS IN UTAH AND IDAHO BY ARTHUR CRONQUIST Utah State Agricultural College, Logan Collections of Spherophysa Salsula ( Pall.) DC. have re- cently been reported by Alexander? from Colorado and Utah, and by Eastwood? from Oregon and Arizona. In July, 1937, Mr. Joseph Pechanec, of the U. S. Forest Service, called to the writer’s attention a red-flowered astragaline plant which he had observed for several years along highway U.S. 91, one mile north of Roberts, Jefferson County, Idaho. A collection was made on July 12, 1937, (Cronquist No. 697), and was determined early in 1938 by Professor Ivar Tidestrom as Swainsona galegifolia R. Br., with the brief note, “introduced from Australia.’”’ A close check with Alexander’s description of his 1936 collection of Spherophysa Salsula indicates that the plant from Roberts is 1 Alexander, E. J., Torreya 37: 35 (1937). 2 Eastwood, Alice, Leafil. West. Bot. 2:38 (1937). NOVEMBER, 1939] NEW PLANT RECORDS 211 doubtless the same as that of Alexander. It is interesting to note that another recent introduction from Asia, Hymenophysa pubes- cens C. A. Mey.® was found growing within a rod of the Sphero- physa. Specimens from the Spherophysa collection have been deposited in the herbaria of the Utah State Agricultural College and the University of Idaho, Southern Branch. Oryzopsis Webberi (Thurb.) Benth. is reported by A. S. Hitchcock* from Colorado, Nevada, and California. Mr. Pe- chanec collected this plant in 1935 on the grounds of the U. S. Sheep Experiment Station, near DuBois, Clark County, Idaho. Verification was received from Mrs. Agnes Chase at the Grass Herbarium of the U. S. Department of Agriculture in 1937. On July 28, 1937, the writer made a collection (Cronquist No. 767) consisting mostly of overmature fruiting plants from the same station as Pechanec’s original collection. Specimens have been deposited in the herbaria of the Utah State Agricultural College and the University of Idaho, Southern Branch. Agropyron albicans Scribn. & Smith is listed by A. S. Hitch- cock * as reaching its western limit in Wyoming. A single poorly developed but readily recognizable plant of this species was brought to the writer in the summer of 1937 by an A. A. A. employee who found it in Camas Meadows, Clark County, Idaho. The specimen is retained in the writer’s personal collection. J. T. Howell® has recently recorded some new localities for Tragopogon dubius L. and adequately pointed out the characters which distinguish it from T. pratensis L. Tidestrom® does not list JT. dubius from Utah or Nevada, and Rydberg’ records it only from Colorado. According to Rydberg’s treatment and Howell’s more detailed description, the common yellow-flowered Tragopogon in the vicinity of Logan, Utah, and Pocatello, Idaho, is undoubtedly T. dubius. Occasional plants of T. pratensis are also to be found in Logan. 8 Fogg, John M., Jr., Rhodora 39: 190-192 (1937). 4 Hitchcock, A. S., Manual of the Grasses of the United States (1936). 5 Howell, J. T., Leafi. West. Bot., 2: 89 (1938). 6 Tidestrom, Ivar, Flora of Utah and Nevada (1925). a. P. A., Flora of the Rocky Mountains and Adjacent Plains 212 LEAFLETS OF WESTERN BOTANY [ VoL. II, NO. 12 CARDUUS IN CALIFORNIA BY JOHN THOMAS HOWELL It was not long after I had reported the occurrence of Car- duus neglectus in California (Leafl. West. Bot. 1 :22,—1932) when I began to doubt the correctness of this name which had been applied (im herb.) by Dr. F. Petrak to a collection made by Miss Eastwood at Fort Bragg. In 1935, while studying at Kew and at the British Museum, I looked into the problem and decided that the obnoxious Californian weed was definitely not C. neg- lectus but was probably C. pycnocephalus L. Several months ago, Dr. S. F. Blake wrote to me that he had determined certain Californian collections as C. tenuiflorus Curt., and more recently the problem was again brought to my attention by Dr. Mary L. Bowerman, who, in trying to identify specimens she had collected on Mt. Diablo, made the critical observation that two entities seemed to be present. In order to answer Dr. Bowerman’s ques- tions and to correct my error in the reported determination, I undertook a critical examination of the Californian specimens of Carduus, a study which has shown that both C. pycnocephalus and C. tenuiflorus are naturalized in northern California. Although these species frequently exhibit evident characters by which they may be easily recognized, they are closely related and under varying environmental conditions individuals of one species may be found which almost exactly simulate in general appearance those of the other species. Although I regard these Old World thistles as two distinct species, all botanists do not hold that opinion and two recent workers in floristic problems of the Mediterranean area have indicated this: Borg (Descript. FI. Maltese Isls. 618,—1927) treats C. tenuwiflorus as a variety of C. pycnocephalus, while Knoche (FI. Balearica 2 :487,—1922) expresses the definite opinion that the two are distinct. Both plants are annuals with spiny-winged stems which vary from 0.5 to 2 m. in height and bear small more or less clustered heads with light purplish-rose flowers. In the Old World, C. pycnocephalus is nearly confined to lands bordering the Medi- terranean Sea and C. tenuiflorus is found widespread through central Europe. In California, C. pycnocephalus has been seen only from Sonoma, Contra Costa and San Francisco counties, but C. tenuiflorus has been seen from Humboldt, Mendocino, NOVEMBER, 1939 | CARDUUS IN CALIFORNIA 213 Marin, San Francisco, San Mateo, Santa Clara, Contra Costa, Alameda, and San Benito counties. The two species may be distinguished by the following synopses : CARDUUS PYCNOCEPHALUS L. Stems usually with narrow spiny wings which are frequently narrower and more or less interrupted below the heads; heads at ends of branches usually few (1—5) ; involucral bracts not membranous-margined, more or less persistently floccose-tomentose. the tips of the outer and middle bracts more rigid, the margins and backs bearing tiny rough upwardly appressed trichomes especially on the promi- nent mid-vein; corolla-lobes generally about 3 times as long as the corolla- throat; achenes light tan or buff, usually with about 20 nerves, pappus 1.5—2 cm. long. CARDUUS TENUIFLORUS Curt. Stems usually with broad spiny wings which extend up to the heads; heads usually numerous (5—20) ; involucral bracts more or less membranous-margined, tomentum usually scant, the tips of the outer and middle bracts glabrous and smooth except on the sub- ciliate margins ; corolla-lobes generally 1.5—2.5 times as long as the corolla- throat; achenes gray-brown, usually with 10 to 13 nerves, pappus 1—1.5 cm. long. The Californian collections of Carduus generally compare favorably with European specimens whose determinations were checked with Hegi (Fl. Mitt. Europ. 6:846, 861—863) and with Rouy (FI. de France 9:67, 71—73). However, some specimens from more inland stations match almost exactly material from France determined as “x Carduus Therioti Ry.,” a reputed hybrid between C. tenuiflorus and C. pycnocephalus. Here, how- ever, such specimens are referred to C. tenuiflorus since the heads and involucres are definitely those of that species. Hegi does not recognize hybridization between these two species (/. c., p. 864). Turrill (Kew Bull. 1938, p. 386) states that C. tenuiflorus is one of those species which in England occur “near the coast”’ in “sea-air’”’ and it may be that the difference observed in the plants from inland stations in California results from the warmer drier habitat of the interior. Carduus neglectus Ten., to which Petrak referred Miss East- wood’s collection from Fort Bragg, is figured by Tenore (FI. Napol., atlas 4, tab. 187) as a slender plant with solitary heads on elongate unwinged hoary-tomentose peduncles, a plant unlike anything I have seen from California. Tenore’s species is now generally treated as a synonym of C. acicularis Bert. 214 LEAFLETS OF WESTERN BOTANY [ VOL. II, NO. 12 EUGENIA MYRCIANTHES BY ALICE EASTWOOD Some years ago Cecil Hart sent me a fruit from a tree grow- ing on the Moore place, Rideout Heights, a garden in Whittier, California. It was a fleshy fruit, yellow on the outside, about the size of a small sickle pear, with a sort of a knob at the top, formed by the base of the calyx. The yellow flesh within clung to the round nut as in the mango. The globular nut, about the size of a small walnut, had such a hard, thick, woody outer coat that it could not be opened and a hammer and a saw had to be used. This certainly was a puzzle. Later he sent me flowering speci- mens which undoubtedly put it into the Myrtle Family near Eugenia. It seemed incredible that any member of this group could have such a fruit. Specimens with both flowers and fruit were needed to convince me, but I was convinced by later speci- mens in flower and with immature fruit. Among all the genera with fleshy fruit in this family none is described with fruit like this. : I sent specimens to the New York Botanical Garden and Mr. Moldenke kindly sent me this letter with the identification : The specimen which you sent for identification arrived in very good condition and proved ample for me to determine it. It has proved to be Eugenia Myrcianthes Niedenzu. The earliest name for the species appears to be Myrcianthes edulis Berg in Mart. Fl. Bras. 14, 1: 353 (1857), based on a Sellow specimen from Montevideo, Uruguay. The species was later transferred into the genus Eugenia by Bentham & Hooker, and the name Eugenia edulis (Berg) Benth. & Hook. f. (ex Griseb., Goetl. Abh. 24: 126,—1879) is still fre- quently used for it. However, the specific name edulis cannot be used for this species because of the Eugenia edulis of Velloso (Vell. Fl. Flum. 208; Icon. 5:t. 34,—1825). The correct name, therefore, seems to be Eugenia Myrcianthes Niedenzu (in Engler & Prantl, Nattir. Pflanzenfam. 3, 7: 81,— 1893). It is perhaps also worth noting that the Eugenia edulis Kiaersk., Enum. Myrt. Bras. 162 (1895), is still another plant, being based on the Phyllo- calyx edulis of Berg (in Mart., Fl. Bras. 14, 1: 327,—1857). If retained in the genus Eugenia, it, likewise, cannot retain the specific designation edulis and must receive the binomial Eugenia Selloi Hort. ex Jackson, Ind. Kew. 1: 911 (1893), 2: 513 (1894). All this in spite of the fact that Velloso’s Eugenia edulis has now been removed from the genus and bears the name Myrciaria edulis (Vell.) Skells, U. S. Bur. Pl. Ind. Bull. 148: 14 (1909) ! NOVEMBER, 1939] TWO NEW LUPINES 215 It seems probable that no botanist has ever before seen fresh, ripe fruit. It was originally collected by Sellow, in Uruguay, who reported the pulp as acid-sweet and fruit with one or two nuts. The genus Myrcianthes with which it was first placed is de- scribed as having a membranous outer coat to the subreniform nut. In neither of these characteristics does this apply to the nut of this fruit, which, as above stated, is globose, about the size of a small walnut, with a hard, woody, outer coat; nor is there any Eugenia described with a similar fruit. To me it seems to belong to a distinct genus, neither Eugenia nor M yrcianthes, but in this complicated aggregate of subgenera, where such diverse opinions have been expressed by more experi- enced botanists, I hesitate to take such a course. TWO NEW LUPINES BY ALICE EASTWOOD Lupinus caudiciferus Eastwood, spec. nov. Caules plures ex caudice magno et ligneo, graciles, 3—4 dm. alti, simplices, adpresso-piloselli, foliosi, terminati racemis verticillatis, pedunculis 2—8 cm. longis; foliolis 5—7, oblanceolatis, 15—20 mm. longis, 3—6 mm. latis, apice acutis apiculatis, viridibus, adpresso-pilosellis, infra densioribus supra, petiolis gracilibus foli- orum inferiorum 6—8 cm. longis, superiorum multo brevioribus, floribus 8 mm. longis, pedicellis gracilibus, 2 mm. longis; calycis labio inferiore herbaceo, ovato, apice acuto, 4 mm. longo, labio superiore breviore, ovato, membranaceo, basi saccato, apice truncato; corolla circa 1 cm. longa, vexillo reflexo, orbiculato, 8 mm. diametro, violaceo, luteo-maculato, ungue lato, saccato, brevissimo; carina curvata, 3 mm. lata, alba, apice purpura- scente, ciliata ex medio ad basin; bracteis deciduis, pilosis, equilongis alabastris. Type: Herb. Calif. Acad. Sci. No. 263363, collected on Elk Mountain, Lake County, California, May 17, 1938, by East- wood and Howell, No. 5702. This dainty lupine springs from a large woody caudex with many slender leafy simple stems 3 to 4 dm. high, terminating in whorled racemes with the peduncles sometimes surpassing the leaves and sometimes very much shorter. The flowers are violet, 8 mm. long, with a distinct space of about 4 mm. separating the banner and the wings. The spot on the banner is conspicuous and turns brownish when old. The white back of the keel is noticeable. The pubescence is on both sides of the green leaves, 216 LEAFLETS OF WESTERN BOTANY [VvoL. II, NO. 12 not dense, but more so on the lower surface and appressed. The petioles of the lower leaves are long and slender and those of the upper much shorter ; the bracts are deciduous, about equal to the flowers in bud. This belongs to the L. latifolius group and is nearest to L. viridtfolius Heller, the type locality of which is Dunsmuir. Comparing it with specimens from Dunsmuir, Shasta Springs, and other places in the valley of the upper Sacramento, it appears quite different from L. viridifolius. The plant is much smaller with rather sparse appressed hairs on both sides of the leaves, simple stems, leaflets narrowly oblanceolate, pointed at apex, peduncles generally much longer and bracts not surpassing the flowers in bud. The woody caudex is so marked a feature that it suggested the specific name. Lupinus Isabelianus Eastwood, spec. nov. Caules ramosi basi ex cau- dice ligneo, foliosi, steriles et floriferi dense congesti, albi pilis sericeis argenteis dense adpressis et laxe divaricatis; petiolis gracilibus, infimis 8—10 em. longis; foliolis 5 ad 8, anguste oblongo-lanceolatis, circa 1 cm. longis, apice apiculatis; stipulis basi breviter conjunctis, filiformibus, vil- losis, 5 mm. longis; racemis verticillatis, pedunculis folia superantibus, bracteis deciduis, circa equalibus alabastris, floribus violaceis, circa 1 cm. longis; calycis inferiore segmento oblongo, obtuso, 5 mm. longo, superiore breviore, prope disjuncto ad basin; vexillo breviore alis, reflexo, dorso piloso, aurantiaco-maculato, in senectute purpureo-maculato, orbiculato, 8 mm. diametro; alis oblongis, 6 mm. latis; apice carine exserto, relinquo tecto alis; carina prope stricta, pauce ciliata infra apicem, ungue 4 mm. longo. . Type: Herb. Calif. Acad. Sci. No.'264244, collected May 17, 1938, by Eastwood and Howell, No. 5687, above the ranger station on Elk Mountain, Lake County, California, amid the pines, hanging over a cut on the road. It is very lovely and is named in honor of Dr. Isabel McCracken, eminent entomologist from Stanford University, who was our delightful companion on the trip. It belongs in the L. albifrons group and in habit somewhat resembles L. collinus Greene. It is smaller in all its parts, pubes- cence not so closely appressed. The banner is a beautiful violet color with an orange spot about the middle, dotted with violet and turning purple when old. It is hairy on the back. VoL. II LEAFLETS of No. 13 WESTERN BOTANY ¥ CONTENTS The Sterculias Cultivated in California ALIcE Eastwoop Californian Plants, Mostly New A. A, HELLER New Information Regarding Calyptridium and Spraguea Rosert F. Hoover New Species of Lupinus A ice Eastwoop Studies in Ceanothus—III JoHN THomas Howe. SAN FRANCISCO, CALIFORNIA JANuArRY 31, 1940 PAGE 217 219 222 226 228 LEAFLETS 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. Vt INCHES TUCO LE COS OS eee oe ee ee ee ae Ag Mt Belg Gil! li) Owned and published by Atice Eastwoop and Joun THomaAs Howe. JANUARY, 1940] STERCULIAS IN CALIFORNIA 217 THE STERCULIAS CULTIVATED IN CALIFORNIA BY ALICE EASTWOOD Species of Sterculia are more commonly cultivated in Southern California than in the north and with one exception belong to the subdivision Brachychiton, They are also known under that name. Two species are well known, S. acerifolia A. Cunn., the Flame Tree, and S. diversiloba G. Don, the Bottle Tree. The former one has maple-like leaves and large panicles of brilliant red flowers, generally in bloom when the trees are leafless. The latter has usually ovate acuminate leaves and greenish-yellow bell-shaped flowers in axillary panicles. The trunk is stout, broadening to the base. It isa common street tree in Los Angeles. The pods of these two are smooth. The trees with large pods, brown on the outside from a rusty tomentum, have been known under three different names, S. Bidwillii Hook., S. discolor F. Muell. and S. lurida F. Muell. All have tomentose flowers and pods. Sterculia Bidwilli is illustrated in Curt. Bot. Mag. t. 5133. The red narrow tubular bell-shaped flowers are in large almost sessile clusters on naked stems or in the axils of upper leaves. It is said to be a shrub or tree from a large tuberous root. Probably no true S. Bidwillii is cultivated in California, as I have seen none with flowers resembling those illustrated. The other two seem to be the same. The differences between them fail when many trees are observed. Since S. discolor was published first, that will be the name if they prove to be the same. Bentham in Flora Australiensis 1 : 228 expresses a doubt as to these closely related plants being specifically distinct. Dr. J. R. Brown’s observations, which I quote, sustain this view: “After close examination of all trees: Elysian Park, Huntington Botanic Garden, University of California, Los Angeles Botanic Garden and many smaller lots, there appears to be but one species. The leaves vary from entire to seven-lobed and a great many vari- ations between. Color changes are produced on the under surface of many leaves, some more than others; often turning from green to white with apparent loss of the tomentum; this on the whole or only part of the under surface. This condition as a rule Leafl. West. Bot., Vol, II, pp. 217-240, January 31, 1940. 218 LEAFLETS OF WESTERN BOTANY _ [VOL. II, NO. 13 most marked on parts of the tree with the greatest sun exposure. A great many trees do not have leaves showing this condition ; some having green leaves on both sides and those with white leaves on the lower surface are often both on the same tree. Almost mature fruit, open flowers and others about to open are seen on the same tree, long bloomers. Many trees do not blossom each year, others set little or no fruit. The total number of mature fruits is never great. Many trees are naked but a short time, almost evergreen, while others take plenty of time, weeks and months to return to full foliage. The flowers appear as the leaves are being shed and continue to complete return of foliage.” Specimens showing leaf variations and flowers were sent to me to substantiate Dr. Brown’s observations. The flowers and fruits of both species are the same, and in the key to the species in Flora Australiensis the only differences are in the character of the leaves: S. lurida with leaves green on both sides, palmately 5- or 7- lobed, and S. discolor with leaves white underneath, angular or obscurely 5- or 7-lobed. Sterculia platanifolia L., the Japanese Varnish Tree, is known also as Firmiana platanifolia Schott & Endl. This has very different pods. Instead of follicles with the seeds at the bottom of the open pod as in Brachychiton, this has the seeds on the edges of the widely spreading leaf-like valves. It is more widely cultivated in the southeastern United States than in California. In the Herbarium of the California Academy of Sciences are specimens from Niles Nursery, the Bard Garden at Hueneme, Ventura County, and also one from Wilson’s collection in West Hupeh, China. LatHyrus ApHaca L. This odd-looking, yellow-flowered, Eurasian pea is naturalized in middle and southern Oregon west of the Cascades but heretofore I have not known it from Cali- fornia. During the summer of 1938 it was collected in Fort Bragg, Mendocino Co., where it flourished as a garden weed (Eastwood & Howell No. 6168). Its further occurrence in Cali- fornia on moist slopes of the northern coastal hills is to be expected.—John Thomas Howell. 5) JANUARY, 1940] CALIFORNIAN PLANTS 219 CALIFORNIAN PLANTS, MOSTLY NEW BY A. A, HELLER Spraguea pulcherrima Heller, spec. nov. Annua?; caulibus purpura- scentibus, ad 27 cm. altis, erectis vel suberectis, bracteatis omnino; foliis rosulatis, oblanceolatis, ad 9 cm. longis, ad 15 mm. latis, attenuatis in petiolos alatos bis longiores, apice subacutis, costa prominenti ; inflorescentia late capitata, multiflora; pedicellis filiformibus, purpurascentibus, 10—12 mm. longis; floribus saturo-roseo-purpurascentibus ; sepalis 2, orbiculari- bus, 7 mm. diametro, subcordatis basi, emarginatis apice; petalis oblongo- spatulatis, circa 6 mm. longis, 3 mm. latis supra, 1.5 mm. latis basi, 2 petalis costatis, 2 non prominenter costatis sed sub medio abruptius contractis ; staminibus luteis, maturis circa 1.5 mm. longis. The type, in the Heller Herbarium, is Heller No. 15132, collected May 30, 1938, in granite sand along the Feather River near Bucks Creek Power House, Plumas Co., California, on State Highway No. 24. It was fairly common for several miles along the highway and was not noted in formation other than granite. Delphinium armeniacum Heller, spec. nov. Perenne ad 35 cm. altum, ex rhizomatibus longis horizontalibus breviter positis, glabrum omnino; caulibus purpurascentibus, foliosis ad inflorescentiam ; foliis distantibus, non succulentis, inferioribus orbicularibus, 3—7 cm. diametro, partibus princi- palibus cuneatis vel cuneato-oblongis, 3—5-lobatis, lobis acutis vel sub- acutis, mucronatis, superioribus sepe 3-lobatis, lobis linearibus acutis; inflorescentia patenti, ad 17 cm. longa; pedicellis filiformibus, infimis 5—6 em. longis, brevioribus supra, subulato-bracteatis basi et 1 vel 2 bracteas distantes supra ferentibus ; floribus armeniacis, maturis circa 25 mm. longis, horizontalibus in pedicellis; sepalis circa 10 mm. longis, plus minusve ciliatis margine, obtusis vel subacutis, calcare directo, circa 15 mm. longo, spe gracili apice sed interdum crassiusculo, obtusiore, paululum curvyato ; petalis coloratis similibus sepalis, superioribus biserratis, apice paulum irregulariter laceratis ciliatis, inferioribus obovatis, apice paulum coartatis mucronulatis obtusis. The type, in the Heller Herbarium, is Heller No. 15149, collected June 4, 1938, on a sparsely wooded, dry slope about a mile east of Fredonyer Pass, elevation 5600 ft. Fredonyer Pass is in Lassen Co., California, between Westwood and Susanville, on State Highway No. 36. Arid Transition Life Zone. This species is apparently intermediate between D. nudicaule T. & G. and D. cardinale Hook. Its flower and leaf characters differ greatly from those of D. nudicaule, within whose range it occurs. The name refers to the more or less apricot color of the flowers. So far as observed by me, its root character is peculiar. 220 LEAFLETS OF WESTERN BOTANY [ VOL. II, NO. 13 Polygala lasseniana Heller, spec. nov. Fruticulus subprostratus, 20—30 cm. latus, 10 cm. altus, foliosus omnino, caulibus imbricate impli- catis, dilute viridibus, puberulentibus pilis paucis directis vel partibus superioribus subglabris, in spinas aciculares 5—12 mm. longas terminanti- bus; foliis alternantibus, coloratis similibus caulibus sed minus pubes- centibus, elliptico-oblanceolatis, vix petiolatis, maioribus in medio caulis 12—15 mm. longis, circa 5 mm. latis; floribus dilute roseo-purpurascentibus et luteis, irregulariter racemosis circa terminos caulium, 7—10 mm. longis, glabris preter basin calycis minute pubescentem, pedicellis gracillimis, zquilongis floribus, minute hirtellis; sepalis superioribus et inferioribus 6 vel 7 mm. longis, dilutioribus quam sepala lateralia et petala, sepalis lateralibus saturo-coloratis, 10 mm. longis; petalis circa 8 mm. longis, superiore bifido ad medium, lobis 1 mm. latis, oblongis, saturo-roseo- purpurascentibus, inferiore luteo, cristato; capsulis 8—9 mm. longis, 6 mm. latis, venulosissimis, apice emarginatis, basi breviter stipitatis. The type, in the Heller Herbarium, is Heller No. 15213, collected June 23, 1938, in a stony wash in Balls Canyon south of Secret Valley, Lassen Co., California, along U. S. Highway No. 395. Upper Sonoran Life Zone; a species of sagebrush (Artemisia) the prevailing plant. The relationship of the plant is with P. acanthoclada Gray. PH#osTOMA Spach, Hist. Nat. Vég. Phaner. 4: 392 (1835). About the only thing, as I see it, that this genus has in common with typical Clarkia, C. pulchella Pursh, is the obconic calyx- tube. The two new species described below belong to the Cali- fornian medley long known as Clarkia rhomboidea Dougl., the original of which was collected by Douglas more than a century ago “from the Great Falls of the Columbia to the Rocky Mountains.” Phzostoma atropurpureum Heller, spec. nov. Annuum, circa 35 cm. altum, ramosum prope basin, simplex in juventute; caulibus et ramis pur- purascentibus, adpresso-puberulentibus; foliis tenuibus, elliptico-oblongis vel lanceolato-oblongis, acutis et mucronulatis, tenuiter puberulentibus, laminis circa 13 mm. longis, petiolis gracilibus, infimis laminis zquilongis supremis brevioribus; pedicellis circa 5 mm. longis; tuba calycis 2.5 mm. longa, sepalis lineari-acuminatis, ad 15 mm. longis, circa medium 2 mm. latis; petalis atropurpureis, obovato-cuneatis, 7 mm. longis, 5 mm. latis circa medium, apice 2—3 mm. latis, subacutis; filamentis gracilibus, pur- pureis, 5 mm. longis, squamis basalaribus albo-villosis, 1 mm. longis, basi latis, antheris 5 mm. longis, flavis, curvatis in senectute; stylo purpureo equilongo staminibus; capsulis adpresso-puberulentibus, curvatis, 20 mm. longis. The type, in the Heller Herbarium, is Heller No. 15129, collected May 30, 1938, on a steep slope in coarse granite sand JANUARY, 1940] CALIFORNIAN PLANTS 221 near shrubs, about two miles west of Arch Rock Tunnel, Butte Co., California, on State Highway No. 24. The petals are usually mottled with darker spots near the reddish base. Phzostoma Mildredz Heller, spec. nov. Simile P. atropurpureo sed differt in floribus: tuba calycis 4 mm. longa, segmentis lineari-acuminatis, 2.5 mm. latis sub medio; petalis atropurpureis, lamina flabelliformi, 10 mm. longis, 8 mm. latis circa apicem, sub medio 2.5 mm. latis, basi 4.5 mm. latis ; squamis albo-villosis, 2.5 mm. longis, gracilibus, basi 0.5 mm. latis, apice basi paulum latioribus; filamentis purpureis, 7 mm. longis, antheris rubescentibus, 7 mm. longis, paululum curvatis in senectute ; stylo purpureo, staminibus paulum longiore; capsula crassa, curvata, circa 20 mm. longa, adpresso-puberulenti. The type, in the Heller Herbarium, is Heller No. 15153, collected July 2, 1938, west of Arch Rock Tunnel, Butte Co., California, on State Highway No. 24. It grew in coarse granite sand on a steep slope. It is named in honor of my daughter, Mrs. Mildred G. Pritchett, who was with me when the plants were collected. In general it resembles P. atropurpureum, col- lected six weeks earlier in the same area, but differs greatly in flower characters, as may be noted by comparing the descriptions. Phzostoma modestum (Jep.) Heller, comb. nov. Clarkia modesta Jep., Manual, 673 (1925). May 27, 1914, I found this interesting species about nine miles east of Alder Springs, Glenn Co. It was distributed as a new species of Pheostoma under No. 11458, but was never published. The type, Jepson No. 2690, is from Waltham Creek, San Carlos Range. Gentiana tiogana Heller, spec. nov. Perennis, nana, circa 5 cm. alta, omnino glabra; caulibus simplicibus vel divaricate ramosis ex radice robusta ad 25 mm. longa, foliosis; foliis tenuibus, viridescentibus, oblongo- spatulatis, acutis, infimis 25 mm. longis, supremis linearibus, acutis, sub- connatis basi; pedicellis 1 mm. longis, crassiusculis ; tuba calycis 8—9 mm. longa, viridescenti vel purpureo-lineata apice segmentorum, hyalina inter segmenta, segmentis lineari-acuminatis, 5 mm. longis, 1 mm. latis; corolla 25 mm. longa, tubiformi, 5—6 mm. diametro, lineis obscuro-purpureis 2 mm. latis ad apicem segmentorum apiculatorum 5 mm. longorum lineata, appen- dicibus inter segmenta 1 vel 2, acuminatis, 1 mm. longis ; staminibus insertis 8—9 mm. ex basi corollz, filamentis circa 6 mm. longis, 1 mm. latis, angus- tissimis sub antheris, antheris eburneis, 2 mm. longis; stylo flavescenti, paulo breviore filamentis. The type, in the Heller Herbarium, is Heller No. 15453, collected August 10, 1939, at Tioga Pass, Mono Co., California, elevation 9900 feet. It grew among short grasses about a lakelet 222 LEAFLETS OF WESTERN BOTANY [ VOL. II, NO. 13 only a few yards east of the Yosemite National Park boundary. This delicate little plant is related to Gentiana Newberryi Gray. It probably occurs in meadows within the Park, but being very small and inconspicuous, may easily be overlooked. It was found while collecting G. holopetala Holm which was plentiful about the lakelet. NEW INFORMATION REGARDING CALYPTRIDIUM AND SPRAGUEA BY ROBERT F. HOOVER ter In May, 1938, a plant was found in the Sierra Nevada foothills of Mariposa County, California, which proved to be Spraguea pulchella Eastwood, a species known previously only from J. W. Congdon’s original specimens collected near the “Pea Ridge Road” in Mariposa County. When it was found that this species, although very similar to most species of Calyptridium in general aspect, had never been named under that genus, a study of the nature of the differences between Spraguea Torr. and Calyptridium Nutt. was made, in order that the correct name might be applied to the plant under consideration. By many botanists Spraguea has been combined with Calyp- tridium. This course was first adopted by E. L. Greene (1886), for the reason that ‘‘genera cannot be rested on mere length of styles and filaments.” Such characters, together with number of ovules or seeds, have constituted the principal basis for the recognition of Spraguea as a genus. Most recently the sepa- ration of the two groups as genera has been maintained by Rydberg (1932). Disregarding for the time the question of the generic distinctness of Spraguea, the writer undertook to find how nearly S. pulchella corresponds with other species in the supposed generic characters. In fruit characters, the plant under consideration is certainly a typical member of Spraguea. The capsule is orbicular and strongly flattened in such a manner as to form a broad wing-like margin around the solitary black seed. The flowers, however, are unlike those of Spraguea. The style is elongated but hardly equals the petals and is not at all exserted, thus showing a con- dition intermediate between Spraguea and Calyptridium. The stamens are of the kind found in Calyptridium, the broad anthers JANUARY, 1940] CALYPTRIDIUM AND SPRAGUEA 223 being nearly sessile and considerably shorter than the petals. The petals in fruit are coherent and form a readily deciduous calyptra, as is usual in Calyptridium. From these observations it was concluded that flower and fruit characters are not decisive, and an effort was made to find other possible differences between Spraguea and Calyptridium. Apparently no attention has been given heretofore to the vege- tative organization of the plant. In all the strictly annual species of Calyptridium, the main axis of the plant terminates in flower- ing branches. The large leaves of the basal rosette and the smaller ones on the lower part of the branches may be somewhat crowded, but on examination the structure is always evident. The growth of the plant is thus determinate. In contrast, Spraguea umbellata Torr. and all its segregates bear flowering branches laterally in the axils of the basal leaves, never termi- nally. The continued growth of the basal rosette above the bases of the flowering branches is evident even in plants collected late in the season. Although some forms of Spraguea may grow as annuals or at least flower the first year, it seems probable that all are potentially perennial, as are certain usually annual ferms of Eschscholtzia californica Cham. If Spraguea and Calyptridium are to be regarded as distinct, the difference in vegetative structure must be emphasized, as it appears to be more distinctive than features of flower or fruit. Spraguea pulchella is in vegetative organization exactly like Calyptridium monandrum Nutt. and C. quadripetalum Wats., and like those species bears well developed leaves on the branches even to the lower nodes of the inflorescence. It was this vege- tative resemblance which first led the writer in the field to regard the plant as a species of Calyptridium. Spraguea pulchella is thus a typical Calyptridium in habit, a typical Spraguea in fruit, and intermediate, though perhaps somewhat nearer Calyptridium, in flower characters. Such combination in this species of the distinctive features of the two groups seems a fairly conclusive argument against the validity of the genus Spraguea. It has been pointed out previously by Greene (1886) and by Jepson (1914) that certain species connect the two groups, but in no other species is there such a combination of flower, fruit, and vege- tative characters as in S. pulchella. The rediscovery of this rare species thus proved to have an important bearing on the classifi- 224 LEAFLETS OF WESTERN BOTANY [ VOL. II, NO. 13 cation of the entire group. The conclusion seems clear that, since vegetative organization, the most significant difference be- tween Spraguea and Calyptridium, is not strictly correlated with flower and fruit differences, as shown by S. pulchella, Spraguea should be regarded not as a distinct genus but as a section of Calyptridium. However, if Spraguea is to be regarded as a dis- tinct genus, it must be based principally on the vegetative organt- zation of the plant. Therefore, in the opinion of the writer, S. pulchella is properly referred to Calyptridium even if Spraguea be separated from that genus. The necessary combination is accordingly here made. Calyptridium pulchellum (Eastwood) Hoover, comb. nov. Spraguea pulchella Eastwood, Bull. Torr. Club 29:79 (1902). In the original description it is said that this species “comes nearest to Calyptridium monospermum Greene,” to which it has been referred subsequently as a synonym by Rydberg (1932). A study of the type collections of both species and of my col- lection of C. pulchellum (Hoover No. 3442) shows a number of significant differences between the two. The difference in habitat is itself notable. While C. pulchellum is a vernal annual of the foothills at an altitude of 2000 ft., C. monospermum is an alpine plant occurring at 10,500 ft. The latter has slender anthers which equal the petals, a definitely exserted style, and petals which remain distinct at the apex in withering. In all these features of the flower, as well as in the complete absence of foliage leaves from the flowering branches, C. monospermum corresponds with Spraguea and is quite unlike C. pulchellum. However, a much more important difference is found in the vegetative organization of the plant. The condition in C. pul- chellum has been described above as typical of Calyptridium. Calyptridium monospermum, on the other hand, clearly shows a rosette of leaves of indeterminate growth without terminal flowering branches. It is thus a typical member of Spraguea. Although C, monospermum is here discussed as a species, it is doubtful whether it is more than a reduced alpine form of C. umbellatum (Torr.) Greene. The problem of classification of Spraguea can properly be undertaken only by one who has facilities for field study of all the forms in their various habitats, in order that the differences may be properly evaluated. JANUARY, 1940} CALYPTRIDIUM AND SPRAGUEA 225 Calyptridium pulchellum is an attractive little plant with spreading stems not over three inches long and showy rose- colored flowers. The nature of its occurrence is worthy of note. It was found in only one place on a south slope in sandy soil surrounding outcrops of dark-colored stratified rocks, apparently slate. The situation was exposed but surrounded by Quercus Wislizenit A. DC., the only species of tree occurring on the slope. The locality was determined from the topographic map as the hill lying between Mariposa Creek and the source of Humbug Creek, The upper slopes of the hill were covered by large colonies of Lupinus deflexus Congdon and a pale-flowered form of L. Stiversii Kell., which because of the whitish flowers had the appearance of dry straw as seen from a distance. The local occur- rence of L. deflexus is of exceptional interest, because it is an- other rare species collected by Congdon near the “Pea Ridge Road.” It seems probable that the locality where these plants were observed by the writer is the same as Congdon’s Pea Ridge, a name not found on the maps. In addition to the two species of Lupinus previously mentioned, L. nanus Dougl. and L. Ben- thamu Heller were also locally abundant. Because of the un- usually plentiful occurrence of plants of the Pea Family, the name “Pea Ridge” would be quite appropriate. REFERENCES Eastwood, A. Some New Species of Californian Plants, Bull. Torr. Club 29: 75—82 (1902). Greene, E. L. Some Californian Polypetalae, Bull. Torr. Club 13: 141—144 (1886). Jepson, W. L. Calyptridium, FI. Cal. 1:463—465 (1914). Rydberg, P. A. Spraguea, Calyptridium, N. Am. FI. 21: 316— 320 (1932). ANTIRRHINUM OrontTiIuM L. In April, 1939, Mr. C. A. Reed discovered the Lesser Snapdragon in Santa Cruz Co. about 10 miles north of Santa Cruz. Although we do not know of its occurrence elsewhere in California, it is a weed of Eurasian and Mediterranean lands that is likely to spread and become common if once it is established. To the French it is known as Téte-de- Mort and to the Italians, Gallinella—John Thomas Howell. 226 LEAFLETS OF WESTERN BOTANY [VOL. II, NO. 13 NEW SPECIES OF LUPINUS BY ALICE EASTWOOD lay Lupinus angustiflorus Eastwood, spec. nov. Caules solidi, erecti, straminei, costati, leviter adpresso-pubescentes, ramosi supra; ramis erectis, terminantibus racemis, brevi-pedunculatis laxe floriferis; foliolis 5 ad 7, lineari-oblongis, obtusis, apiculatis, 2—9 cm. longis, circa 5 mm. latis, supra glabris, infra leviter adpresso-pubescentibus, villis brevibus, tenuibus ; petiolis zqualibus vel brevioribus foliolis, leviter pubescentibus; stipulis subulatis brevibus, paulo basi conjunctis; floribus albis (?), circa 12 mm. longis; pedicellis erectis, filiformibus, 5—10 mm. longis, bracteis persistentibus, lineari-acuminatis, petiolis brevioribus; calyce basi in alabastro cuneato, segmento superiore 2-fisso, inferiore apice truncato, 3-dentato, leviter pubescente; vexillo zquali alis, suborbiculato, circa 1 cm. lato, coartante ad unguem circa 1 mm. latum, sinu supra unguem, superiore parte dilatato, dorso medio villoso; carina tecta alis oblongis, 1 cm. longis, 5 mm. latis, ungue 1—2 mm. longo; carina glabra, prope stricta, apice acuta. Type: Herb. Calif. Acad. Sci. No. 189454, collected by Mrs. E. C. Van Dyke, June 13, 1931, at Kelly Camp, Mt. Lassen National Park, California. This belongs perhaps to the L. laxiflorus group but is en- tirely without either the spurred or saccate base of the calyx. The pubescence is similar on stems, upper leaf-surface, pedicels and calyx, consisting of fine short sparse appressed hairs. The lower part of the plant is unknown, the type sheet has three erect stems with many erect branches, all yellowish and rather slender and ending in laxly flowered, short-stemmed racemes about 1 dm. long. The flowers which appear to be white taper from a rather broad top to the base of the calyx and are longer than the calyx and erect. The bracts are persistent and shorter than the pedicels. The fruit is unknown. I have seen no other lupine with flowers like this. Lupinus indigoticus Eastwood, spec. nov. Caules ex rhizomate ligneo, fistulosi, 5 cm. diametro, 6—8 dm. alti, purpurascentes, puberulentes, infra sine foliis, supra ramosi et foliosi; racemis terminalibus, pedunculatis et floribundis ; floribus densis, diffusis; bracteis filiformibus, alabastra super- antibus, mature deciduis; foliolis 6 ad 9, plerumque 8, anguste oblanceo- latis, apice acutis obtusisve, mucronatis, 2—5 cm. longis, 3—5 mm. latis, sericeis et argenteis villis adpressis, densioribus infra; petiolis commune circa zequilongis foliolis, stipulis filiformibus, adnatis ad medium; calycibus circa 6 mm. longis, basi minimis, lobis obtusis, integris prope zequalibus ; pedicellis strigosis, brevioribus calycibus; corolla circa 7—8 mm. longa, indigotica, vexillo glabro sessili, brevioribus alis ; carina alisque cohzrenti- JANUARY, 1940] NEW SPECIES OF LUPINUS 227 bus basi, unguiculatis, vexillo et alis margine undulatis; alis 5 mm. latis, carina glabra, tecta alis, preter apicem purpureum; leguminibus ignotis. Type: Herb. Calif. Acad. Sci. No. 216318, collected April 17, 1934, three miles east of Walterville, Lane County, Oregon, in the Mackenzie River Valley by Eastwood and Howell, No. 1588. This is a tall bushy perennial herb with stout, hollow, purplish stems, leafless at base, branching above, each branch ending in a long raceme on peduncles about as long. The leaves are rather pale green from the appressed short silvery hairs on both sides, the lower denser than the upper. The attenuate bracts can be seen only on quite young racemes where they surpass the buds. They fall before the flowers open. This perhaps comes nearest to L. Andersoni Watson. The flowers are much smaller and the pubescence different. Lupinus mariposanus Eastwood, spec. nov. Perennis, circa 9 dm. altus ; caulibus solidis, infra sine foliis, supra divaricate ramosis, purpureis, strigosis; foliolis 7 ad 9, oblanceolatis, circa 2 cm. longis, 5 mm. latis, apice mucronatis, adpresso-pubescentibus, villis sericeis; petiolis nonnumquam longioribus foliolis, strigosis; stipulis lineari-attenuatis ; racemis peduncu- latis, 1.5—2 dm. longis, floribus verticillatis vel diffusis, circa 1 cm. longis, 9 mm. latis, violaceis; pedicellis 2 mm. longis; calyce bracteolato, labio superiore 2-dentato, labio inferiore integro, obtuso, paulo longiore et latiore superiore; corolla violacea, circa 1 cm. longa; vexillo orbiculato, 7 mm. diametro, reflexo, basi saccato; alis 9 mm. longis, 4 mm. latis, basi auricu- latis ; carina curvata, glabra, paulo tecta alis, apice purpurea, ungue brevi; bracteis ovatis, preecoce caducis ; leguminibus ignotis. Type: Herb. Calif. Acad. Sci. No. 262645, collected May 29, 1938, 4 miles from Nipinnawasee on the road to Mariposa, Mari- posa County, California, by John Thomas Howell, No. 13880. This is a tall bushy perennial herb, about 9 dm. high, with stems naked below, branching widely above, purplish, pubescent with short close spreading hairs; leaflets 7 to 9, oblanceolate, about 2 cm. long, 5 mm. wide, mucronate, clothed with silvery silky hairs closely appressed but not dense; petioles pubescent like the stems, a little longer than the leaflets; stipules sepa- rate, linear-attenuate ; racemes 1.5—2 dm. long, terminating the branches ; flowers violet-blue, verticillate or diffuse, about 1 cm. long, 7 mm. wide, on pedicels 2 mm. long; calyx bracteolate, upper lip 2-toothed, lower entire, obtuse, a little longer and broader than the upper; corolla about 1 cm. long, 7 mm. wide; banner roundish, 7 mm. across, strongly reflexed, the claw sac- 228 LEAFLETS OF WESTERN BOTANY _ [VOL. II, NO. 13 cate; wings 9 mm. long, 4 mm. wide, auriculate and with a short claw ; keel curved, smooth, white at tip, scarcely covered by the wings; bracts ovate, acuminate, evident only on the racemes in bud ; pod unknown. This is one of the species aggregated under L. albicaulis Dougl. This perplexing group is in general characterized by the glabrous, strongly curved keel, covered but little by the wings, branching stems, leaves with petioles equaling or shorter than the leaflets, pubescence various but stems and foliage never smooth. STUDIES IN CEANOTHUS—III BY JOHN THOMAS EN a The following notes on the species and varieties of Ceanothus included in my recent key to the subgenus Cerastes on the Pacific Coast (Leafl. West. Bot. 2:204—207) are presented chiefly for the enlightenment of those who may erroneously believe that all problems in the group have been solved. 1. CEANOTHUS VESTITUS Greene, Pitt. 2:101 (1890). Following Trelease (1897, p. 416), Standley (1923, p. 722), and Munz (1935, p. 302), I am not recognizing C. Greggit Gray as a member of the Californian flora. I have not seen the type of Gray’s species from the “Battlefield of Buena Vista,’ Mexico, but from the original description and from the examination of specimens collected in Mexico, it would appear quite probable that C. Greggii, at least in its typical form, is strictly Mexican. No specimens from the United States have been seen with pedicels pubescent as in the Mexican specimens I have examined and as described by Gray, “‘pedicellis pubescentibus (demum glabratis)” (Pl. Wright. 2:28,—1853). In the United States, all the material from Arizona to Texas which is generally referred to C. Greggt does not seem to represent a single entity, but study in the field will probably be needed in order to interpret properly the variants which I have seen. Although the whole C. Greggu complex is still poorly understood, I give the follow- ing tentative range for C. vestitus: in California, from the southern Sierra Nevada in Inyo and Kern counties to the northern slopes of the San Bernardino Mts., east to the Charles- JANUARY, 1940] STUDIES IN CEANOTHUS 229 ton Mts., Nevada, and perhaps to northern Arizona. In north- western Arizona at Hackberry, a plant probably referable to C. vestitus has been named C. Greggti var. orbicularis (E. Kelso, Rhodora 39:151,—1937); and more recently McMinn has treated the plants from the western part of the range as C. Greggu var. vestitus (1939, p. 312). Some Californian variations are to be noted in C. vestitus. In the original description, the leaves and branchlets are de- scribed as ashy-tomentulose and the leaves “4 to 6 lines long, round-obovate, . . . sharply spinulose-dentate all around.” In most of the specimens that have been examined, the leaves and branchlets are tomentulose only when young and are early glabrescent. The leaves are generally more elongate, oblong- oblanceolate to elliptic or obovate and the margins are as frequently entire as they are minutely to more coarsely spinulose- dentate. In the Kern River Canyon a variant has been collected which is intermediate between C. vestitus and C. cuneatus. In this the hairs of the pubescence are straight or nearly so and appressed as in C. cuneatus, the leaves are uniformly entire and are larger and oblanceolate, but the fruit has horns decidedly lateral and midway down the sides. 2. CEANOTHUS PERPLEXANS Trelease in Gray Syn. FI. N. A. 1, pt. 1: 417 (1897). C. Goldmanii Rose, Contrib. U. S. Nat. Herb. 12: 284 (1909). C. Greggii var. perplexans (Tre- lease) Jepson Man. Fl. Pl. Calif. 623 (1925). With Trelease, I believe that C. perplexans is closely related to C. crassifolius, more closely perhaps than to C. vestitus with which it has been frequently confused. The broader and mostly larger leaves and the larger stipules seem to be characters which definitely sepa- rate it from C. vestitus. It appears to be common enough in the mountains west of the Colorado Desert from the southern slope of the San Bernardino Mts. to the mountains of northern Lower California where it has been named C. Goldmanii. Trelease also reports C. perplexans from Guadalupe Island and Arizona. I have not seen the insular material (probably C. crassifolius of various authors) nor have I seen any specimen referable to C. perplexans from Arizona. For the privilege of examining the type of C. Goldmanit, I am indebted to Dr. W. R. Maxon. 230 LEAFLETS OF WESTERN BOTANY [ VOL. II, NO. 13 3. CEANOTHUS cCRASSIFOLIUS Torr., Pacif. RR. Rept. 4:75 (1857). The tomentum which clothes so densely the branchlets and lower side of leaves in this species has been found to be the same in character as that in the two preceding species and in representatives of the C. Greggi complex found in states to the eastward, but more or less different from the vesture in all other species in the subgenus under consideration. I believe that this alignment based on the character of the pubescence may indicate a fundamental phylogenetic grouping, and perhaps a primitive one, when one considers the wide distribution of the C. Greggu group in the Sonoran regions of the mountains and plateaus in the United States and Mexico (cf. Standley, 1923, p. 722). Ceanothus crassifolius is found in the chaparral of the lower mountains from Santa Barbara Co., California, southward to northern Lower California. The variety, var. planus Abrams, well deserving nominal recognition, is known only from Santa Barbara and Ventura counties, California. 4. CEANOTHUS MEGACARPUS Nutt. N. A. Syl. 2:46 (1846). C. macrocarpus Nutt. in T. & G. Fl. N. A. 1:267 (1838), not Cav. This species occurs in the lower coastal mountains from San Diego Co. to Santa Barbara Co. Although it is usually a shrub less than 3 m. tall, I have seen arborescent specimens in the Santa Ana Mts. up to 5 m. tall with slender trunks covered with a finely furrowed dark grayish-brown bark. In the mountains of Santa Barbara Co. is a distinctive entity with opposite or occasionally alternate leaves and with large fruit and conspicuous horns. Attention has been called to this plant (K. Brandegee, 1894, p. 206; McMinn, 1930, p. 145), but no data on its occurrence in the field have been seen. A possible relationship between it and the peculiar form of C. cuneatus growing in the Pecho district of San Luis Obispo Co. should perhaps be sought. 5. CEANOTHUS INSULARIS Eastw., Proc. Calif. Acad. Sci., ser. 4, 16: 362 (1927). An insular species closely related to the preceding, C. insularis is found on Santa Catalina, Santa Cruz, and Santa Rosa islands. In the type specimens which were col- lected on Santa Cruz Island, the leaves are uniformly opposite ; but in other specimens from that island as well as in specimens from the other islands, the leaves are occasionally alternate JANUARY, 1940| STUDIES IN CEANOTHUS 231 (cf. Trelease, 1897, p. 416; Munz, 1935, p. 302). The close relationship between the island and mainland plants is expressed by the name C. megacarpus var. insularis (Eastw.) Munz (Bull. S. Calif. Acad. Sci. 31:68,—1932). In referring C. insularis to synonomy under C. crassifolius, Jepson (1938, p. 480) dis- poses of the island plant under the name by which it was reported many years ago (Zoe 1: 134; Bull. Calif. Acad. Sci. 2: 393) and fails to recognize as representative of C. imsularis certain island collections which he cites as “a hornless variety” of C. megacarpus (ibid., 476). 6. CEANOTHUS VERRUCOSUS Nutt. in T. & G. Fl. N. A. 1: 267 (1838). This species with its alternate leaves and small generally hornless fruits is indicated as a probable base in a phylogenetic scheme for the subgenus Cerastes by McMinn (1930, chart on p. 123). I agree that there is much weight to be placed on the evidence he presents, but, as I have suggested above, the widely dispersed C. Greggii group may be more primitive. Ceanothus verrucosus, restricted to the arid coastal mesas and slopes from San Diego, California, to Ensenada, Lower California, does not seem to occupy a primitive geographical or geological position unless its present limited and specialized distribution is inter- preted as highly relictual. 7. CEANOTHUS SUBMONTANUS Rose, Contrib. U. S. Nat. Herb. 12: 284 (1909). Although this Lower Californian plant bears a striking resemblance to C. cuneatus in general appear- ance and was treated as a part of that species by Standley (1923, p. 721), the subvillous pubescence of twigs and leaves and the lateral horns on the fruits seem to mark it as a distinct entity. However, until further field work has been done and extensive collections have been studied, I believe that a proper systematic estimate of C. submontanus cannot be given. A direct relation between it and C. cuneatus should perhaps be sought through those subtomentose-pubescent forms of the latter species which occur in southern California. The attachment of the horns on the fruit is of critical importance in C. submontanus and it will be necessary to investigate the stability and value of the character. In the type specimen, which I was privileged to examine through the courtesy of Dr. W. R. Maxon, the stems are slender and appear as if they might have been rather pliant. The hairs 232 LEAFLETS OF WESTERN BOTANY [ VOL. II, NO. 13 on the young stems are dense, short, and a little crisped, and, although some are distinctly spreading, mostly they are sub- appressed but not uniformly directed upward. The inflores- cences are shortly but definitely pedunculate. The fruits are 5 mm. in diameter, dark brown, the horns are short, about 1 mm. long, spreading, slender, and the intermediate crests are repre- sented by low wrinkled folds. 8. CEANOTHUS FRESNENSIS Dudley ex Abrams, Bot. Gaz. 53:68 (1912). Although I once wrote something about C. fres- nensis as a montane homologue of C. cuneatus (Leafl. West. Bot. 1:71), in this treatment I have placed it in the “Greggu- verrucosus’ group because the attachment of the slender spread- ing horns is generally distinctly lateral. On the other hand, the prostrate habit, the apically denticulate leaves, and the blue flowers seem to indicate a definite relationship to C. prostratus. Perhaps this attractive plant is descended from the offspring of a fertile cross between C. prostratus and C. vestitus, a cross which would combine those characters that mark this plant as a very distinct entity with a natural distribution in the middle altitudes of the Sierra Nevada of California from Plumas Co. (McMinn, 1939, p. 309) southward to Fresno Co. (See also remarks under C. connivens Greene.) 9. CEANOTHUS RAMULOSUS (Greene) McMinn, Madrofio 5:14 (1939). Although I am well acquainted with the much- branched, dark green, blue-flowered shrub as it occurs in the Mt. Tamalpais region of Marin Co. and in the Calistoga region of Napa Co., I do not know it as it grows in the Santa Cruz Mts. which is the locality first cited by Greene in the original description of C. cuneatus var. ramulosus (FI. Francis. 86, 1891). Whereas in Greene’s original description the leaves are described as “narrower and longer’ than in the species, this is a distinction not met with in the plants I know in which the leaves are frequently smaller and broader than in C. cuneatus. Usually the shrub is erect, but McMinn (Madrofio 5:15) calls attention to a subprostrate form on the bluffs above Pt. Sal, Santa Barbara Co. As I understand C. ramulosus, it is intermediate in character between C. rigidus and C. cuneatus, but since it maintains a certain distinctness through a natural geographic distribution, JANUARY, 1940] STUDIES IN CEANOTHUS 233 I believe it is well regarded as a separate specific entity. It is perhaps a matter of significance that C. ramulosus has been found chiefly in areas where C. rigidus and related species are adjacent to C. cuneatus. A real problem centers in this species which must be known more adequately both geographically and systemati- cally before it can be properly interpreted in the evolutionary fabric of which the subgenus Cerastes seems to be constituted. 10. CEANOoTHUS CUNEATUS (Hook.) Nutt. in T. & G. FI. N. A. 1:267 (1838). Rhamnus cuneata Hook. Fl. Bor. Amer. 1:124 (1829). This widespread species, although generally readily recognizable, is almost as variable as the diverse floral provinces it inhabits and may at length prove to be a complex of several smaller specific entities when it is studied in more critical detail. Or it may be that part of the variation within the specific lines set here is the effect of hybridization between C. cuneatus and other species in the subgenus Cerastes. This species has sometimes been regarded as the most widely distributed species in the subgenus Cerastes (cf. McMinn, 1930, p. 124), but its range is not so great as that of C. Greggu as that species has been accepted by Standley (1923, p. 722), McMinn (1939, p. 311), and others. Ceanothus cuneatus ranges from middle western Linn Co., Oregon, southward through the hills and lower mountains of California where it is the most widely dispersed species in the genus. Neither Mrs. Brandegee (1894, p. 205) nor Munz (1935, p. 303) reports C. cuneatus south of California, but both Standley (1923, p. 721) and McMinn (1939, p. 311) extend its distribution into northern Lower California. Perhaps Standley included C. cuneatus as a Lower Californian shrub because he treated C. submontanus Rose as that species. I have not seen material collected by Palmer on Guadalupe Island so I cannot say whether his collection re- ported by Watson (Proc. Am. Acad. 11:114) as C. cuneatus would at present be referred to this species. There is some variation in the character of the pubescence of young stems which seems to be correlated with geographic distribution. In Oregon near the northern limit of its range, the pubescence is subvillous and consists of loose upwardly directed hairs*; in southern Oregon and throughout most of 1 The original collection was made near the northern limit of the range and the branches were described by Hooker as “subferrugineo-pubescenti- bus,” “clothed with a rusty-colored down.” 234 LEAFLETS OF WESTERN BOTANY [ VOL. II, NO. 13 California the pubescence of upwardly appressed hairs is dis- tinctly sericeous; in southern California the pubescence is fre- quently villous or subtomentulose and is reminiscent of the pubescence of C. submontanus in northern Lower California. Although I have not seen the type of C. oblanceolatus Davidson (Bull. S. Calif. Acad. Sci. 20: 53,—1921), a specimen I collected at the type locality in Bouquet Canyon, Los Angeles Co., Cali- fornia, has the appressed pubescence of northern Californian plants rather than the subtomentulose pubescence of many southern Californian plants. In the characters of the fruit there are some variations to be noted. In some forms the intermediate crests which are always low and inconspicuous are obsolete and in northern California occurs a variant with fruits larger than in the more common widespread form. In the South Coast Ranges are plants which are related to C. cuneatus which vary in foliage, flowers or fruits, but they cannot be properly treated until the limits of C. ramu- losus have been more clearly defined. Most of these variants have been indicated by McMinn (1930, p. 144). And already, under C. vestitus, I have indicated an intermediate form between C. cuneatus and that species in the Kern River Canyon. Although the occurrence of C. cuneatus in the flora of Arizona has not been suggested heretofore, some specimens from the central part of the state with wrinkled apical horns and low wrinkled intermediate crests seem to be more definitely allied to the C. cuneatus group than to the C. Greggu group to which they have always been referred. Possible relationship between the Arizona plant and the subtomentose form of C. cuneatus of southern California or C. submontanus of Lower California should perhaps be looked for. 11. CEANOTHUS CONNIVENS Greene, Pitt. 2:16(1889). The few times that this plant has been noted in the literature, it has been treated as a probable hybrid of C. cuneatus and C. prostratus (K. Brandegee, 1894, p. 216; Trelease, 1897, p. 416) since in habit and dentation of leaves it partakes of the latter, in fruit and leaf-shape of the former. However, since it has been collected only in Calaveras Co., California, entirely within the range of C. fresnensis as recently extended by McMinn (1939, p. 309), a possible relationship between C. connivens and C. fresnensis will JANUARY, 1940] STUDIES IN CEANOTHUS 235 have to be considered. The nearly apical erect horns and the more elongate leaves (mostly 1.5—2 cm. long and 0.5—1 cm. wide) seem to mark this plant as distinct from C. fresnensis. However, I have examined only two fruiting specimens of C. connivens, both in Herb. Univ. Calif., the type collection which was made “between Murphy’s and the Big Trees” by Greene and a second collection which was made at Sheep Ranch by T. S. Brandegee. It was with some hesitancy that I included C. connivens in my key because it is much too little known to be treated properly, but the very fact that it has been neglected prompted me so to emphasize it that the attention of botanists and botanical collectors may be directed to it. 12. CEANoTHUs FERRIS® McMinn, Madrojio 2:89 (1933). Although this species was originally reported as a variation of C. cuneatus (McMinn, 1930, p. 145) and was related to that species when it was described (1933, p. 90), I am here regarding it as more closely related to the C. rigidus group, not only because of the frequently dentate margin of the leaves but chiefly because of their texture, venation, and conformation; and in his latest work, McMinn interprets the relationship of C. Ferrise in this way too, placing it between C. gloriosus and C. purpureus (1939, p. 317). In the shape and size of the stipules and in the absence of intermediate crests, C. Ferris@ is like the broad-leaved Ceanothus of the Ben Lomond district in the Santa Cruz Mts. which was mentioned by McMinn (1933, p. 90) and which has since been definitely reported as C. Ferrise (Wieslander and Schreiber, Madrono 5:40,—1939) ; but I do not regard the plants as the same although the Ben Lomond plants may prove to be only varietally distinct. Whereas I have seen C. Ferrise@ only on the serpentine ridge in the vicinity of the type locality in the Mt. Hamilton Range, I have seen the plant of the Santa Cruz Mts. only in areas of sedimentary rocks, either sandstone or shale. According to McMinn (1939), C. Ferrise “occurs . . . occasion- ally in the Santa Cruz Mountains.” 13. CEANOTHUSs RIGIDUS Nutt. in T. & G. Fl. N. A. 1:268 (1838). In this treatment, C. rigidus is restricted to the region of the Monterey Peninsula where it was originally collected. It 236 LEAFLETS OF WESTERN BOTANY [VOL. II, NO. 13 may be necessary eventually to include some forms from as far south as Santa Barbara Co. as has been done by Jepson (1936, p. 478) ; but I am reluctant to do it now although I have no name for the related plant in southern California.2, The plant in Marin Co. which has for many years been referred to C. rigidus does simulate that species to a remarkable degree ; but field study has convinced me that this plant is to be definitely associated with C. gloriosus and is perhaps the result of hybridization between C. gloriosus and C. ramulosus where they occur together in the Mt. Tamalpais region. Even on the Monterey Peninsula C. rigidus is not entirely uniform and one variant with broadly cuneate leaves and shortly pedunculate inflorescences has been distinguished as var. pallens Sprague (Kew Bull., 1915, p. 380). 14. CEANOTHUS PUMILUS Greene, Erythea 1: 149 (1893). Although this was originally described from Waldo, Josephine Co., Oregon, and has doubtless been regarded as an Oregonian species by Californian botanists, it occurs in California in Del Norte Co., and, in the Red Mt. region in northern Mendocino Co., it has been named as C. prostratus var. profugus Jepson (1936, p. 479).° Trelease (1897, p. 416) considered C. pumilus to be a hybrid between C. cuneatus and C. prostratus, but the species has distinctive characters which are maintained through a natural range in the western part of the Siskiyou area nearly corresponding to the range of that remarkable endemic, Garrya buxifolia Gray. 15. CEANoTHUs conFusuUs J. T. Howell, Leafl. West. Bot. 2:160 (1939). As this was originally understood and described, it was believed to be most nearly related to C. divergens Parry with which it had been long confused ; but with the preparation of these notes, C. confusus has seemed to be about as nearly related to C. pumilus of which it may be regarded a derivative in the southern North Coast Ranges of California. 16. CEANOTHUS DIVERGENS Parry, Proc. Davenport Acad. Sci. 5:173 (1889). Although specimens of C. divergens simu- late C. purpureus Jepson, the two are very easily distinguished 2McMinn (1939, pp. 309, 315) refers this variant in Santa Barbara Co. to C. ramulosus. 3 Mature fruits have not been seen from the Red Mt. region, but im habit foliage and flowers the plants are like those studied in Del Norte Co., Cali- fornia, and Josephine Co., Oregon (cf. Leafil. West. Bot. 2:162). JANUARY, 1940| STUDIES IN CEANOTHUS 237 by excellent characters of foliage and fruits. How variable C. divergens is in the field and what possible relation it may bear to C. purpureus or to C. confusus are questions which await critical field study in the narrow area about the Napa Valley where it is endemic. (Cf. Leafl. West. Bot. 2: 159, 165.) 17. CEANOTHUS SONOMENSIS J. T. Howell, Leafl. West. Bot. 2: 162 (1939). Restricted to the chaparral-covered slopes of the Hood Mt. Range on the east side of the Sonoma Valley, C. sonomensis seems to be as narrow an endemic as C. purpureus Jepson in the Napa Range and typical C. Ferrise McMinn in the Mt. Hamilton Range. Data have been collected in an attempt to show that C. sonomensis is probably a hybrid of relatively recent origin and it is planned to present the evidence in a later paper. 18. CEANOTHUs GLorIosus J. T. Howell, Leafl. West. Bot. 2:43 (1937). C. rigidus var. grandifolius Torr., Pac. RR. Re- port 4:75 (1857). On coastal flats and mesas from Marin Co. to Mendocino Co., the typical prostrate aspect of this species grows, but back from the coast the erect form, var. exaltatus J. T. Howell, ibid., 44, flourishes in brush and forest. Near Tomales Point, Marin Co., a form with nearly entire leaves has been collected, but it is readily recognized as this species by the texture and venation of the leaves and by the stipules. Wherever C. gloriosus and another species of the subgenus Cerastes grow in proximity, puzzling hybrid-like intermediates occur. Such have been found in the Santa Rosa region near the type locality of C. sonomensis and on Bolinas Ridge where C. ramulosus also occurs. 19. CEANOTHUS PURPUREUS Jepson FI. W. Mid. Calif. 258 (1901). This very narrow and distinct endemic is found only in the southern part of the Napa Range in Napa and Solano counties where it grows in the chaparral on the bedded volcanic rocks so characteristic of the region. The very large stipules in this species reach a maximum size for the genus Ceanothus. They were found to be so decisive in separating C. purpureus from species which it resembles that I have given considerable attention in this study to their relative development in all parts of the subgenus Cerastes. Although the characters of the stipules and of the fruit seem to indicate a direct relation between 238 LEAFLETS OF WESTERN BOTANY _ [VOL, II, NO. 13 C. purpureus and the C. rigidus group (cf. Howell, Leafl. West. Bot. 2:45), C. purpureus bears a strong resemblance to C. Jep- soni in habit and foliage, a resemblance which has prompted the combination C. Jepsoni var. purpureus Jepson (Man, FI. Pl. Calif. 624, 1925). 20. CEANOTHUS PROSTRATUS Benth., Pl. Hartw. 302 (1848). As in the case of the other widely distributed species in the sub- genus Cerastes, such as C. Greggu and C. cuneatus, C. prostratus is the center of a number of variations which are as yet only partly understood. The number of “prostratus” entries made in northern and central eastern California by McMinn (1930, p. 123) in his geographic and phylogenetic chart of the subgenus Cerastes is indicative of the variations of which I speak. In many instances these variations appear to have originated through hybridization and several have been named, such as C. fresnensis and C. connivens. But other hybrid-like variants besides these are yet to be studied and recorded before we can properly under- stand C. prostratus and delimit it from the hybrid-like variants which still confuse the concept. It is of interest to note that, in those suspected hybrid-derivatives that have been segregated from C. prostratus, the character of the fruit partakes not of C. prostratus but of the other parent-suspect.* For that reason, in this study where primary emphasis has been placed on charac- ters of fruit whenever possible, these species have been treated in those groups where their fruits are more nearly typical. In the Lake Tahoe district two species have been named which are apparent hybrids between C. prostratus and some member of the subgenus Euceanothus: C. rugosus Greene® (FI. Francis. 88,—1891) from near Truckee, and C. serrulatus McMinn (Madrofio 2: 89,—1933) from near Cascade Lake. The other parent of the suspected hybrids might be C. cordulatus Kell. or C. velutinus Dougl., both of which occur in the region. Ceanothus prostratus is the most northerly ranging species of the subgenus Cerastes and is found from southern Wash- ington (Piper, Fl. Wash., p. 387) southward through the Cas- 4 Thus the fruit of C. fresnensis approaches that of C. vestitus, the fruit of C. connivens approaches that of C. cuneatus, and the fruit of OC. pumilus (if that species should be considered here at all) approaches that of C. rigidus. 5 In the original description Greene suggests that one parent may have been C. cuneatus, but as Mrs. Brandegee (1894, p. 217) points out, it is “without doubt a cross between C. velutinus and C. ‘prostratus.” JANUARY, 1940] STUDIES IN CEANOTHUS 239 cades of Oregon to California and Nevada. Inthe Sierra Nevada of eastern California and western Nevada the range of C. pro- stratus is definite enough but its detailed distribution is not yet known in the high North Coast Ranges of California where it perhaps extends southward to Mendocino Co. (McMinn, 1930, p. 146). The pale form from the eastern slope of the Sierra Nevada in Nevada was named by Greene in honor of C. F. Baker, but I have not found the name properly published with a descrip- tion and I do not see in the plant a distinctive character which would warrant specific recognition. In both the Coast Ranges and Sierra Nevada in California, just to the south of C. prostratus, are to be found two remark- able species, each of which I regard as outlying derivatives of that species: C. pinetorum in the Sierra Nevada and C. Jepsont in the Coast Ranges. Unlike those derivatives of suspected hybrid origin which are noted above, these two species are characterized by fruits as large as or larger than the fruit of C. prostratus. 21. CEANOTHUS PINETORUM Coville, Contrib. U. S. Nat. Herb. 4:80 (1893). This species, which grows in the southern Sierra Nevada of California in Kern and Tulare counties at middle altitudes and higher, is much nearer C. prostratus than C. Jepsoni with which most workers have more or less confused it and from which it has been separated heretofore with difficulty if at all.© The resemblance between the two is striking, but I believe it to be a simulation, not indicative of a real or direct relationship which would be extraordinary if one considers geo- graphic distribution. From a taxonomic point of view it might be added that if all species of Ceanothus were as easily separated as C. pinetorum and C. Jepsoni, the genus would not be the truly difficult one that it is in many parts. In reviewing these species for the present study, I prepared the following diagnoses : Ceanothus pinetorum. Branchlets somewhat pliant, internodes glabrous ; leaves mostly spreading and plane with 5 to 8 teeth on the sides, stipules conspicuous and sometimes making a corky ring at the nodes; peduncle glabrous; fruit 6—8 mm. long, overtopped by the slender horns, the horns wrinkled or a little knobbed, nearly round, or if broader, then flattened 6 According to his chart, McMinn (1930, p. 123) would have C. pinetorum derived from CG. cuneatus and Jepson (1936, p. 476) suggests a relation between C. pinetorwm and OC. perplexans; but in this work. where I have placed emphasis on characters of fruit and of pubescence, I am unable to bring these entities into close relationship. 240 LEAFLETS OF WESTERN BOTANY [ VOL. II, NO. 13 tangentially, intermediate crests of inconspicuous low wrinkles; base of style much swollen ; seed 4 mm. long, 3 mm. wide, oblong, truncately obtuse. Ceanothus Jepsoni. Stems rigid, pubescent when young ; leaves deflexed, variously folded or undulate, with 4 to 6 teeth on the sides, stipules usually smaller; peduncle pubescent; fruit 8—10 mm. long, crowned with horns and crests, the horns coxcomb-like and flattened radially, the intermediate crests usually more than one, much-wrinkled with knob-like processes that are more or less confluent with the lower part of the horns; seed 4.5 mm. long, 2.75 mm. wide, obovate. 22. CEANOTHUS JEPSONI Greene Man. Bay Region Bot. 78 (1894). This remarkable species is an inhabitant of serpentine areas, occurring diagonally across the southern part of the North Coast Ranges of California from Marin and Sonoma counties (where the flowers are light to dark purplish-blue) to the inner- most ranges of Napa, Lake, and Tehama counties (where the flowers are whitish). Although in typical form it is one of the most distinctive species of Ceanothus, puzzling aspects are found occasionally which may have originated through hybridization. At times these intermediate forms are like C. Jepsoni in appear- ance but differ in some character of the fruit ; at other times the habit and foliage are quite unlike those of C. Jepsoni but the fruit discloses the marked and unmistakable “influence” of C. Jepsoni. As in all instances of suspected hybridization in this genus, where hybridization is generally acknowledged as not uncommon, one can only properly interpret such “intermediates” with critical and extensive field observation. REFERENCES Brandegee, K. Proc. Calif. Acad. Sci., ser. 2, 4:204—212, 216—218 (1894). Jepson. Fl. Calif. 2:475—480 (1936). McMinn. Contrib. Dudley Herb. 1: 121—147 (1930). Ill. Man. Calif. Shrubs 305—320 (19397). Munz. Man. S. Calif. Bot. 302—303 (1935). Standley. Contrib. U. S. Nat. Herb. 23: 720—722 (1923). Trelease. In Gray Synop. Fl. N. A. 1, pt. 1:416—417 (1897). 7 Prof. McMinn’s book on Californian shrubs does not bear a date of issue. From Mr. J. W. Stacey, who published the work, we learn that it was first offered for sale on Oct. 6, 1939, the official date of publication. The date of publication thus established is especially important in the genus Ceanothus because on several occasions in 1939 nomenclatorial additions and changes have been offered in the genus. Vot. Il No. 14 LEAFLETS of WESTERN BOTANY Y CONTENTS PAGE EOS ISSESS LTT AAA a eel Pe aaa 7s TORUS PRY A.ice Eastwoop What Is Lepidium Menziesii? . . . . . . « 245 C. Leo HitcHcock The Lupinus Breweri Aggregate. . . . . : « 249 ALicEe Eastwoop New Western Plants Aa eC te caatin gor Be ole (his oe JoHn THomas Howe. SAN FrANcisco, CALIFORNIA Aprit 12, 1940 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 NU A INCHES pevgaonyenay tony ovunpna gga pony aca nage ecg nena ny Owned and published by A.ice Eastwoop and JoHN THomAs Howe. APRIL, 1940| STUDIES IN CASTILLEJA 241 STUDIES IN CASTILLEJA BY ALICE EASTWOOD 1. CASTILLEJA IN THE MARBLE MOUNTAINS, Siskiyou County, CALIFORNIA In August, 1939, Mr. John Thomas Howell made a collecting trip to the Marble Mountains in Siskiyou County and brought back a most interesting collection of Castilleja. Some were readily identified, two being endemic, namely, C. arachnoidea Greenman and C. schizotricha Greenman. Another was the widely distributed and variable C. miniata Dougl. and the fourth was doubtfully referred to C. pruinosa Fernald. The others proved to be most perplexing and to the collector appeared so different that while they grew apparently in the same environment the different appearance of each marked them as distinct. This series of specimens is the subject of the present article. Castilleja excelsa Eastwood, spec. nov. Circa 8 dm. alta, supra ramosa, ramis ascendentibus; caule infra simplici, circa 3 dm. alto, puberulo, cos- tato; foliis viridibus, separatis, lanceolatis vel ovato-lanceolatis, sessilibus, basi lata, 4—6 cm. longis, 6—15 mm. latis, 3-costatis, glanduloso-puberu- lis; spicis erectis, elongatis, 1—3 dm. longis, glanduloso-puberulis et glanduloso-villosis, bracteis latis, 3—5-lobatis, lobis rubris, lateralibus divaricatis, multo angustioribus medio lobo; calyce 2 cm. longo, fisso ex basi 9 mm., segmentis loborum 3, inzequalibus, linearibus ; corolla angusta, circa 3 cm. longa, exserta ex calyce 12—15 cm., galea 16 mm. longa, dorso viridi, glanduloso-puberula, ventre rubra, membranacea, labio dentibus attenuatis, incurvatis, terminantibus tumores; testa seminorum membra- nacea, brunnea, scrobiculata. Type: Herb. Calif. Acad. Sci. No. 273635, collected Aug. 4, 1939, near Spirit Lake, Marble Mts., Siskiyou County, Cali- fornia, elev. about 6000 ft., by John Thomas Howell, No. 15058. This seems quite different from the other species of Castilleja collected at the same time by Mr. Howell. The foliage is green, not in the least cinereous, the spikes are much longer and the flowers smaller. The seeds are covered with the transparent scrobiculate testa, but are rounded at top, tapering at base where a short curve gives them the appearance of short, stout commas. It is in general, also, more glandular and also captures tiny insects. Leafi. West. Bot., Vol. Il, pp. 241-256, April 12, 1940, LIBR Saw BOTA! GAR 242 _ LEAFLETS OF WESTERN BOTANY [VOL. II, NO. 14 *. Castilleja globosa Eastwood, spec. nov. Herba perennis, globosa, circa 6 dm. lata, 3 dm. alta; caulibus diffusis, 4—5 dm. longis, ramis multis, gracilibus, leviter tomentulosis et vestitis villis dendroideis tenuibus ; foliis linearibus vel oblongo-linearibus, longissimis foliis 7 cm. longis, 5—8 mm. latis, sessilibus, 3-costatis, supremis raro trisectis; floribus primo erectis in spicis latis et brevibus, in fructo floribus exsertis distichis in spicis longioribus; apicibus bractearum rubris, gracilibus, acuminatis, superanti- bus gemmas, bracteis infimis viridibus, trisectis, segmentis linearibus, acutis ; calyce circa 2 cm. longo, segmentis 7—8 mm. longis, laciniis 4 mm. longis, anguste linearibus; corolla gracili, circa 3 cm. longa, galea 15—20 mm. longa, dorso glanduloso-puberula, ventre rubra, membranacea ; labio parvo, medio dente multo breviore duobus lateralibus obtusis dentibus. Type: Herb. Calif. Acad. Sci. No. 273645, collected Aug. 8, 1939, Black Mt., Marble Mts., Siskiyou County, California, by John Thomas Howell, No. 15159. Castilleja muscipula Eastwood, spec. nov. Omnino cinerea, ramosa basi ex radice lignea, circa 3—4 dm. alta, glanduloso-puberula et vestita villis albis inzequalibus et simplicibus et dendroideis ; foliis anguste oblongis vel lineari-lanceolatis, 3—4 cm. longis, 3—8 mm. latis, integris vel 1- vel 2-lobatis, lobis angustis; spicis terminalibus; bracteis rubris, zquilongis calycibus, trisectis, segmento medio oblongo, latiore segmentis lateralibus linearibus; calyce circa 2 cm. longo, fisso 1 cm. ex basi, laciniis rubris, lineari-attenuatis, circa 3 mm. longis; corolla 3—5 cm. longa, galea 2.2 cm. longa, gracili, conspicue exserta, disticha et curvata extra, dorso viridi et glanduloso-puberula, ventre rubra et membranacea, labio protuberante 3 dentibus incurvis. Type:.Herb. Calif. Acad. Sci. No. 273638, collected Aug. 4, 1939, near Spirit Lake, Marble Mts., Siskiyou County, Cali- fornia, at about 6000 ft. elevation, by John Thomas Howell, No. 15060. This may be related to C. pruinosa Fernald, but differs most conspicuously in the long galea curving outwards in two ranks. The name is given to the group on account of the many small, dead insects found among the glandular spikes and amid the hairs on leaves and stems. Castilleja muscipula var. armeniaca Eastwood, var. nov. Differt: foliis viridioribus, bracteis et segmentis calycis armeniacis, omnino magis glandulosis et villis brevioribus. Type: Herb. Calif. Acad. Sci. No. 273637, collected the same day and place, Howell’s No. 15061. This variety is the greatest insect catcher, the very glandular spike of flowers being full of dead ones. APRIL, 1940| STUDIES IN CASTILLEJA 243 Castilleja muscipula var. angustifolia Eastwood, var. nov. Differt: foliis angustioribus, magis cinereis et foliis supremis trisectis. Type: Herb. Calif. Acad. Sci. No. 273649, collected the same day and place, Howell’s No. 15059. 2. MISCELLANEOUS NEW SPECIES Castilleja filifolia Eastwood, spec. nov. Caules basi ramosi, erecti, tenues, purpurascentes, glabrati vel leviter arachnoideo-floccosi, circa 2 dm. alti; foliis inzqualiter pinnatisectis vel bipinnatisectis, puberulis, seg- mentis filiformibus vel anguste linearibus, rachidi 1 mm. lata; spicis longis, tenuibus, interruptis, flavo-viridescentibus, pedicellis brevissimis, bracteis foliis supremis similibus, lobo medio bractearum superiorum oblongo vel spatulato, 4—5 mm. lato, superante flores; calyce circa 15 mm. longo, tuba 6 mm. longa, lobis segmentorum linearibus, 7 mm. longis, 1 mm. latis, viridi-nervatis ; galea obtusa, 4 mm. longa, dorso puberula, labio et galea prope zquali, labio trisaccato, segmentis erectis oblongis obtusis, circa 1 mm. longis; capsula elliptica, circa 5 mm. longa; seminibus in sicco membranaceo-alatis, in aqua mucosis. Type: Herb. Calif. Acad. Sci., No. 202286, collected June 22, 1931, in open forest, 8 miles north of Diamond Lake, Douglas County, Oregon, by John Thomas Howell, No. 6900. This species is distinguished by the long slender greenish- yellow spikes which appear smooth except under a lens. The divisions of the leaves are thread-like. The whole plant seems smooth, but under a lens is slightly floccose and puberulent. The lobes of the bracts become broad, especially the middle one which almost covers the flowers. The two lips of the corolla are almost equal and the three sacs of the lower lip are tipped with purple and have three erect obtuse divisions. The upper lip or galea is obtuse and about 4 mm. long. It belongs to the group between Orthocarpus and Castilleja. The corolla is that of Orthocarpus with the short galea and 3-saccate lower lip almost as long. Castilleja Jusselii Eastwood, spec. nov. Caules erecti, ramosi vel simplices ex radice simplici, circa 2 dm. alti, virides, subpilosi; foliis inferi- oribus erectis, anguste linearibus, attenuatis, foliis superioribus trisectis, segmentis erectis, longis, linearibus, lateralibus segmentis zqualibus vel brevioribus medio; bracteis flores superantibus, lobo medio spatulato, conspicuo in longis spicis, lobis lateralibus angustis, lineari-oblanceolatis, obtusis; calycis segmentis corollam superantibus, laciniis oblanceolatis, longioribus tuba; galea basi lata, apice acuta, costata, labio inferiore zequali vel paulo breviore galea, lobis 3, obtusis. 244 LEAFLETS OF WESTERN BOTANY _ [VOL. II, NO. 14 Stems erect, simple or with few, erect branches, several from a tap- root, about 2 dm. high, green but with some spreading white hairs; leaves erect, the lower narrowly linear, attenuate, upper with 3 long linear divisions, ribbed, the lateral equaling or shorter than the middle; bracts surpassing the flowers, the middle lobe spatulate, lateral half as wide, linear- oblanceolate, obtuse ; calyx equaling or surpassing the corolla, the divisions of the segments oblanceolate, longer than the tube; galea broad at base, ribbed, pointed, lower lip almost equaling the galea, with 3 obtuse lobes. Type: Herb. Calif. Acad. Sci. No. 165665, collected July 12, 1928, on Star Lake Trail, Lake Tahoe region, California, by M. S. Jussel in whose honor it is named. This belongs to the Castilleja pilosa group, differing from the type in the green hue of the plants, the sparse hairy pubes- cence and the parts of the flowers. It resembles C. pilosa in habit, erect leaves with erect divisions and the broad middle lobe of the bracts. These are imbricated in the young spikes. The flowers seem to be reddish. Castilleja lassenensis Eastwood, spec. nov. Caules czspitosi, simplices, graciles, 10—15 cm. alti, primo nutantes, demum erecti, glanduloso-pilosi ; foliis linearibus vel lanceolatis, acuminatis, 1—2 cm. longis, 1—2 mm. latis, erectis, foliis supremis brevi-trilobatis ; spicis purpureis, bracteis trilobatis, floribus brevioribus, lobis lateralibus linearibus, 3 mm. longis, lobis mediis latioribus, 4 mm. longis, apice truncatis, obtusis vel 3-crenatis; calyce circa 11 mm. longo, segmentis bilobatis, lobis linearibus, obtusis, 2—5 mm. longis ; corolla 14 mm. longa, galea 6 mm. longa, acuminata, exserta, labio circa 4 mm. longo, trilobato, lobis lanceolatis, obtusis, 2 mm. longis. This low subalpine Castilleja grew in dense mats from thick rootstocks clothed with dense, rope-like rootlets; sterile stems much shorter than the flowering ones and often densely clustered at their base. All parts of the plant are clothed with short, glandular hairs. When young the spikes nod, later becoming erect and lengthen in fruit. The narrow, linear or lanceo- late acuminate leaves are erect and hug the stems, only those under the short, purple spikes are lobed and merge into the bracts. The calyx is about 11 mm. long, the two divisions with 2 linear obtuse lobes, 2—5 mm. long. The corolla is conspicuously exserted, the galea acuminate, 6 mm. long, the lip 4 mm. long with 3 lanceolate, erect lobes 2 mm. long. Both galea and lower lip surpass the calyx. The large stigma can be seen just above the apex of the galea and is obscurely lobed. Type: Herb. Calif. Acad. Sci. No. 187874, collected Sept. 7, 1931, on the trail to Bumpus Hell, Mt. Lassen, alt. 7000 ft., by Mr. M. S. Jussel. Other specimens in the Herb. Calif. Acad. Sci. are as follows: one collected by the author in meadows at foot of Lassen Peak, Aug. 26, 1912, No. 1843, Herb. No. 28672, APRIL, 1940] LEPIDIUM MENZIESII 245 another collected by A. H. Kramer, Lassen National Park, summer of 1933, Herb. No. 216887. Castilleja Paynez Eastwood, spec. nov. Herbacea, perennis, basi cespitosa, crassa et squamosa, cinereo-puberula et arachnoideo-villosa; caulibus multis, circa 6—10 cm. altis, simplicibus, erectis; foliis inferi- oribus linearibus, integris, circa 1 cm. longis, 1 mm. latis, foliis superioribus tri- vel bi-sectis, segmentis anguste linearibus, divaricatis, puberulis; spicis 1—3 cm. longis, porphyreis, arachnoideo-villosis, bracteis similibus foliis supremis, segmentis latioribus, flores superantibus, pedicellis 1—1.5 mm. longis; calycibus circa 15 mm. longis, tuba 6 mm. longa, segmentis anguste linearibus, corallam superantibus ; corolla circa 13 mm. longa, galea lata, 5 mm. longa, apice tridentata, interiore villosa, labio trisaccato, 2 mm. longo, segmentis linearibus, 1 mm. longis; capsulis oblongis, 5 mm. longis, 2 mm. latis. Type: Herb. Calif. Acad. Sci. No. 219403, collected July 26, 1929, on Mt. Warren, Warner Mts., Modoc County, California, 9660 ft. alt., by Miss Frances Payne, No. 126, in whose honor it is named. This is related to the group connecting Castilleja and Ortho- carpus. It is a pretty alpine species, the greyish stems clothed with soft, white, cobwebby hairs. The flowers and bracts are reddish-brown, the yellow anthers and capitate stigma are ex- serted and contrast beautifully with the terra-cotta-colored parts of the inflorescence. WHAT IS LEPIDIUM MENZIESIT? BY C. LEO HITCHCOCK University of W ashington, Seattle Lepidium Menzies DC. (Syst. 2: 539,—1821) has long been a source of taxonomic confusion, chiefly because of the fact that the type of the species is in England and therefore students of our western flora have usually found it impossible to study the plant and to compare more recent collections with it. Of the various attempts that have been made to find a match for the Menzies plant, none has been successful. In the writer’s recent paper on the genus, the species was with reluctance reduced to synonymy under L. virginicum var. pubescens (Greene) C. L. Hitchcock (Madrono 3: 283,—1936). As mentioned in that paper (p. 284), Mr. John Thomas Howell took the trouble to examine the DeCandollean type at the 246 LEAFLETS OF WESTERN BOTANY [ VOL. II, NO. 14 British Museum, and compared with the Menzies plant several collections of Lepidium which I sent him. Of these specimens he thought that one from Nanaimo, B. C., Eastwood No. 9767, came closest to matching the plant at the British Museum. His notes on the type, itself, are as follows : “The plant is low (1 dm.) and branched from the base or a little above. The basal leaves are distinctly pinnate and each leaflet again deeply lobed on the upper side two or three times. The pubescence of the leaves is cinereous-subhirsutulous—the hairs coarse and somewhat spread- ing. Above the base the leaves are entire or with one or more lobes or teeth, and linear-lanceolate. The stem and inflorescence are finely puberulent, the same type of pubescence from base to tip, but not on the leaves. The petals are about as large as in your specimen (7. e., Eastwood No. 9767) but not so conspicuous ; certainly the petals in the type collection are not ‘more developed,’ but rather less. The silicles seem to be nearly, if not quite, the same. There is a tendency for yours to be broadest below the middle, at least in some; but in Menziesii the silicles are round. In yours, the ‘wings’ and notch seem to be a trifle more con- spicuous than in the type collection. The type collection has a very sparse crispy puberulence on the silicles.” The presence of this puberulence on the fruits was puzzling, as none of the material I have seen from this region was of this nature, although I suspected that the Menzies plant might be merely a pubescent form of our common littoral plant of the Puget Sound area. Such pubescent-fruited plants occasionally are to be found among the glabrous-fruited plants of practically all our species of Lepidium and Draba. During the fall of 1937 and the spring of 1938 I took pains to examine hundreds of plants of L. Menzies in the field, thinking that I might find some pubes- cent silicles. They were all glabrous, however. While studying the material in the private herbarium of Mr. J. W. Thompson, I noticed a collection, Thompson No. 6036, “on rocky open slopes” at Deception Pass, Island County, Washington, April 25, 1931, represented by two plants about 4 and 6 cm. tall. Both of these plants have a “sparse crispy puberulence on the silicles.” Although they lack basal leaves, they seem to fit Mr. Howell’s description of the Menzies plant quite well. Since other plants from the same vicinity, as represented by several collections, APRIL, 1940| LEPIDIUM MENZIESII 247 have the type of leaf that is characteristic of L. Menztesit, it seems to me that Mr. Thompson’s collection probably can be considered to be “typical” L. Menziesit. During the spring of 1939, I searched for plants like those found by Mr. Thompson in the same general region where he made his collection and finally found a colony of the plants just west of Cranberry Lake, Deception Pass State Park, Whidby Island, Island County (Hitchcock & Martin No. 4658). The plants were limited to a small patch of perhaps one hundred indi- viduals and were found nowhere else,. although the glabrous- fruited form was rather abundant nearby, as it had been the year before (Hitchcock No. 3455). It seems certain, therefore, that L. Menziesu usually has glabrous fruits and that Menzies just happened to collect the rare form with puberulent silicles. Since studying the plants of these two collections as well as many others, both in the field and the herbarium, I have come to believe that I erred in treating the Puget Sound plants as identical with L. virginicum var. pubescens. In comparison with the several varieties of L. virginicum the following character- istics of “L. Menzies” deserve emphasis. It is restricted mainly to a narrow zone near the beaches of Puget Sound, although I have seen it growing on the campus of the University of Wash- ington. There is no difference in shape between its fruits and those of the various varieties of L. virginicum; likewise the flowers show no distinctive peculiarities as some have supposed. The cotyledons are sometimes incumbent, but more frequently they are oblique; in this respect the form is similar to the vars. pubescens and medium (although there may be a general tend- ency for the cotyledons to be more nearly incumbent than in those two varieties, the difference, if there is such, is scarcely detectable). There is no constant difference in the size, number, or shape of the cauline leaves, the leaves in general being most similar to those of the variety medium. The plants blossom from May to October (and probably later) but are usually sturdy biennials or winter annuals with very thick basal rosettes of leaves. Hitchcock No. 3455, from sand dunes west of Cranberry Lake, Whidby Island, May 15, 1938, includes plants in all stages of development, but most of them were depauperate seedlings of 248 LEAFLETS OF WESTERN BOTANY _ [VOL. II, NO. 14 1938 which were in full flower. Likewise, Hitchcock & Martin No. 4658 consists of tiny plants not over an inch in height; although the plants surely are seedlings of 1939, they are all in flower and fruit. It is therefore evident that the plant may behave as an annual, winter annual, or biennial, whereas the other varieties are commonly supposed to be annuals. Local con- ditions play a large part in determining the duration of these weedy pepper grasses, however. Recently I found, on the Uni- versity of Washington Campus, several plants of L. virginicum var. typicum that had blossomed during the spring and summer of 1938 that started to bloom prolifically again in January of 1939, the fall of 1938 having been a very mild one in Seattle. Nevertheless, I do not believe that the plants are identical with any of the other varieties of L. virginicum although too closely related to that complex to merit more than varietal rank there. The main difference is to be found in the pubescence and shape of the basal leaves. As Howell pointed out, they are “pinnate and each leaflet again deeply lobed on the upper side 2 or 3 times.” This type of leaf is occasionally found in the other varieties, especially the var. medium, but it is the usual type in this entity. The pubescence of these leaves and of the basal portion of the stem is rather crisp, coarse, and often curled; it is much more dense and the hairs are coarser than they are in the other varieties. Another general difference that is noticeable is that the plants tend to be more freely branched. These charac- teristics, then, the compact basal rosettes of pinnate-pinnatifid leaves, the coarser pubescence, freely branching habit, and coastal distribution, are thought to be sufficient to warrant recognizing the plant as Lepidium virginicum L. var. Menziesii (DC.) C. L. Hitch- cock, comb. nov. L. Menzies DC. Syst. 2: 539,—1821; L. vir- ginicum subsp. Menziesit (DC.) Thell. Monog. Lepid. 225, 230,—1906, in large part ; L. virginicum var. pubescens (Greene) C. L. Hitchcock, Madrofio 3:283,—1936, in part. Chiefly littoral plants of the Puget Sound region, represented by such numbers as Zeller & Zeller No. 938, Peck No. 13142, Piper No. 444, Otis No. 1640, Hitchcock No. 3455, Hitchcock & Martin No. 4658, and Thompson No. 6036, 10628, and 5226. APRIL, 1940| LUPINUS BREWERI AGGREGATE 249 THE LUPINUS BREWERI AGGREGATE BY ALICE EASTWOOD In the Sierra Nevada of California two types of low perennial lupines inhabit the upper elevations. One may be typified by Lupinus Danaus Gray and the other by L. Breweri Gray. While the former has many stems springing from a taproot, it does not form mats as does the latter. In L. Danaus the banner is strongly reflexed and flares back from the wings, while in L. Breweri it is but little folded back and almost approximate to the wings. The angle between the banner and wings is much wider in L. Danaus than in L. Breweri. In L. Breweri the entire plant is whiter on account of the thick vesture of silvery, silky pubescence, which is generally both appressed and spreading. Herein I am endeavoring to make it possible to discriminate between variants of L. Breweri but am not using the varietal names that have been given as some would be homonyms of already published species of lupines and also because from the meager varietal descriptions it is impossible for me to ascertain what particular plant is meant. Key To SPECIES 1. Stems trailing ; flowering stems leafy.................2.---.-.-0--se+-000-- L. Breweri ie otems matted: floweritig- stems. SCApOSE:....-5. = st ssshcs ence 2 PEPEOWETSCADOUE Omit: LONG = oa ee eS 3 Pea Ss APOUSEA eek TELL II ON cic cote es ass emeee on ae eae eee ee 4 matannen bairyon back, keel ciliate -..c....c-ccssccesce ' * Cw Paredes, bap odt ‘ee 8) 8 aoe , ae : . f - ree Peohesi el OOM) me Met it peer ee ‘ ‘ a . b. bine " - Mel ae Utb enter wetred ‘ ' ‘ pied 7 ; : ‘ 1 UM at MN enebad PIF rh Rat F® Réaat oot Bot ' ou i ; wad eg . sheet ee yh a . ms ‘ f wo , ‘ ‘ j : oh ml, * qe Haye * i eo =e " fps tae . ‘ ae 2 , by ‘ ‘ ‘ ‘ * ‘ L y , privet 4s) ae Que teens ‘ . 4 hap rue em Rarerrrertary rire thi tey ne ao " set . ‘ inale . ‘ ; hat | hy ‘ : hi ' ’ - “ 7 ‘ ' Wide . ‘ i y don et eae aett A me oe 4 ‘ e. h Pane Tiny reer eer aaa ’ , ’ 7 fel ope Care tr tt MEAL Mt de rty ge a ee Pe yr er ee ee ‘? tira . 4 ‘ ‘ ' " My pied pa pahaneh Mv abel Lg ope ved =p tw . ¢ ‘ i « ARI GL hid heeds u el orn oh eo hs ts + on ma adie HUMOR TL Pogeg indy t ey ey ' ; ‘ of tee ; PveTiPrin ee it Toten ty) code ; was , . ‘ : i Pt ae an ee , . . \ gtew ii re ' ‘ } , ' 4 ; Wt " st he f ‘ \ ‘ “ ; . 7?’ i ‘ { dy ' eee! | ‘ ‘ yt pet ts ' ‘ ‘ ' pr ere > ve , ‘ re eee Ey, toate , 6 ely Sebel > MCs tt He’ w . We hee me # LV seh Ar ‘ ‘ " i j ‘ F Te ke ’ : ' Panay ; k : ‘ Coe . , : , ‘ r ds ‘ ' ‘ ; ' - 7 i } “ar . ; ey bide PeaiaM ion i 9K Mae toe Wea! ‘oleay . H t \ an 4 ; ny ‘ ; f P A x \ ‘ ' or ‘ “ ” ‘ - ' ‘ 7 | ‘ " ‘ inv { : ’ wai , . . : VaRe RL byt, EAN 1 P yl is pest hens beeen, Cem itrn tert Heme py . ‘ . ‘ ‘ ‘ Pre ae ye Pepe LEN 9 . por boner teetre wah pee = oa es i ‘ t ( eon een erin ah ae * nk ee) + Fi ‘ \ ’ Mer ac ar ter “” i. . ie ue H * : ho net roe 7 ” ; ’ A « ‘ ‘ ‘ p 4 s ‘ ot ‘ . + j chp i ‘ Me Hit eee ae é , " cet poten © ‘ ‘ poi BZ . " ‘ = ron ye we tg ue . ‘ ‘ A ’ TAT hic hha eae " 7 3 vs ‘ MIE een ate Me a Ae a ; y Tidta ke bah Ae a dahl ee Le , 4 , : ‘ y Meee Ce ce ells! { Iv UNIKTGa beg H) ABLE DEA ! } A) . ; le $ ' Dche: BAbh i 10) eg ee ‘ ’ . bem 4 ' WAT ae a ibe, tone F, : , _— } ave WORE eT ay i ke ra ‘ ‘ ie . a ee a Ae 7 ed at ' an Assen ites - mr t Aiea : ‘ ‘4 eves en es it nates reriris a V Fefig -] ‘ . sh. ; Medea haven : ‘ ¢ aid HH es ae tee PT malt pe ‘ ‘ ' n * eobretees cheer g inh Heme iaee . y Hirsh : : ‘ . ue +6 Gan OH HEE eer ‘ ; ‘ « . ies “ ‘ ‘ “4 a) ; : Ce alee “er ‘ ’ : * lewis beh tretre od . ; ‘ * fier ae bine oe) ; Cites Peat platy thes Tevatelp Fyet ib be aot areas ein ' Oe Ue er 4 they at ety ‘ Hehe : vt ’ ‘ ' ’ y wae : WW : a. ' been . i ‘ Views sont sree i 4 > Pete ad |