res eqn ey) tht, NA 5 AR wie dt Wy} \ (1) Palit hte i at Mt th Hi Baa ( i bey tytn Gta Tet at PAKS Hae A htt iy! MAAS iv t | 2 Hits ABe i) Hvar INH ih Ps i mi it Hy M My } ie ‘ it eres == SSS a Psat eae i hte? tart v si Athi bi OUD in Ses CA MuiTe! DR Ha a ery SATU vee ‘ cy Wa) ANH ATF ne Me ony aa OiaiU { aR RH — i 625 SoS es Se ee = SSS SSS SS = =: = eRe aes See ete en ES ae ry AN aes f bi ny * Bs) ti SS = By i LS ata rt Sao ee Soar sie ee aS Bere SS ta Se Ss eee oaee ee f) at ae prec Soetecetsr er iat it nae iN i Y Mb Mie Lh : a LONE Wate Man itety tte PA aha aa int TTS BESS = Ne ms i epee = SS = a a i i sa ph BC NG ‘ § ny (14; ¥ Wet 8 an re At Hi} F ih i ss Saat i Aint ey A tb 4 ‘4 it f 4 = === = — a x HK An SASS SES ieteteistecee ot: peyese Ss x Ge =", St Sree cee ose < SS hem (7 SSeS SS aE ates cree eee Stoo sox = Sse e Tetra nahh , i : ute Digitized by the Internet Archive in 2009 with funding from Ontario Council of University Libraries http://www.archive.org/details/s3botproceedings01 cali in if ea. a) RROCEEDINGS es CALIFORNIA ACADEMY OF SCIENCES, S. THIRD SERIES 31) Tee: Aor 1897-1900 SAN FRANCISCO PUBLISHED BY THE ACADEMY INESDEOLENs deve polos conshb chots babe” Dopo oon dbo aaemaeranbon nen Contents areca te iereiisicle sve No. 1. A Morphological Study of Naias and Zannichellia. By Douglas Houghton Campbell. (Plates I-V)..................... (Published June 5, 1897.) No. 2. Studies in the Herbarium and the Field. I. By Alice East- WOOds (Plates) ViL—WiDL) fore laretalenclsierchelalel< STIS eRe cys, « eke elelere (Published November 27, 1897.) No. 3. Studies in the Herbarium and the Field. II. By Alice East- Tporale: (WEES ANE. A) Menon one Onc decrenboonadsoneae (Published September 24, 1898.) No. 4. Phycological Memoirs. By De Alton Saunders. (Plates XII- EONENGTI) Jos oj a) oot Slay shah soVcs«\ [2 SR KPT eoearn = 4 epee ame, en Bor Voi PLATE II. All figures refer to Natas flexilis. Fig. 27. Mature male flower, showing the long pedicel; * 25. Fig. 28. A male flower which has discharged the pollen, showing the lateral rupture of the floral envelope; X 25. Fig. 29. Longitudinal section of a nearly mature flower, showing a com- plete division of the anther into two loculi; x qo. Figs. 30, 31. Germinating pollen-spores; X 400. Figs. 32, 33. Pollen-tubes, showing the granular protoplasm at the ex- tremity. In fig. 33, the two generative nuclei, £, can be seen. Figs. 34-39. Successive stages in the development of the ovule, seen in longitudinal section; x about 400, (except fig. 37, which is magnified about 200). The carpel is not shown except in figs. 34 and 37. The archesporial cells are shaded; ¢, the tapetum, z!., 7., inner and outer integuments. Fig. go. Young female flower, showing the young spiny lobes, 7, and stigmatic lobes, v,; one of the latter is still very small; * about 200. Fig. 41. Section of young ovule, with the archesporium divided by longi- tudinal as well as transverse walls. Figs. 42, 43. Young embryo-sacs, with two and four nuclei; X 450. Fig. 44. A mature female flower; X 25. Fig. 45. One of the stigmatic lobes; 200 Fig. 46. A spiny lobe; x 200. Fig 47. Longitudinal section of the ovary of a mature flower, showing the stylar canal and papilla, 4; about 60. Fig. 48. Young papillz from the upper part of the ovary; X 400. Fig. 49. Older papillze from the same place. Fig. 50. Upper part of mature embryo-sac, showing one of the synergide, s; the egg, 0; and the two polar nuclei in close contact, £; 400. Fig. 51. The two synergidz from a similar one. Fig. 52. Egg apparatus from a slightly younger embryo-sac than that shown in fig. 50. Fig. 53. (a) Upper part of the embryo-sac at the time of fertilization; f¢. the pollen-tube, with two generative nuclei; 0, the egg. (6) The large endosperm nucleus, probably formed by the fusion of the two polar nuclei. f hen ono Ww a , . eh ht at it ea i) iy) Oy f ; fe iyiity f Pah fit) Tiercanyetea tp UOTE Te PLate III. All figures refer to Naias flexilis. Fig. 54. Two sections of an abnormal embryo-sac, in which apparently the upper archesporial cells were not absorbed as is usually the case, and no definite egg apparatus was formed. Fig. 55. Lower part of the embryo-sac shown in fig. 54, with six antipodal cells (four only are shown in the section). Fig. 56. Section of an ovule which apparently had not been fertilized; the embryo-sac is replaced by a mass of elongated cells; X 200. Fig. 57. (@and 6) Two sections through a normal embryo-sac, showing fertilization; #7, pollen-tube; 0, egg; s, synergide; £, endosperm nucleus. (c) Lower part of the same embryo-sac, showing two of the antipodals; x 450 (about). Fig. 58. An abnormal embryo-sac, in which only one cell of the egg apparatus could be certainly distinguished. The end of the pollen-tube was very much enlarged. (6) Antipodal region of the same embryo-sac. Fig. 59. Endosperm nuclei from an embryo-sac containing an advanced embryo. (a) The single large nucleus above the antipodal cells, which under- goes no division; (4) The smaller secondary nuclei from the upper part of the embryo-sac; X 400. Fig. 60. Two sections of the upper part of an embryo-sac, showing the fertilized egg cell, 0, and one of the synergide, s, which is still intact. Fig. 61. (@) Upper part of embryo-sac, with two-celled embryo; X 250. (6) Endosperm nucleus of the same. Figs. 62-66. Successive stages of the developing embryo, in longitudinal section; X 250. ss, primary suspensor cell; £, a secondary endosperm nucleus. Fig. 67. Cross-section of an older embryo. Fig. 68. Longitudinal section of an embryo, showing the primary segments I, 2, 3. Fig. 69. Longitudinal section of an embryo, showing the enlargement at the base of the cotyledon. Fig. 70. Two nearly median, longitudinal sections of an older embryo; X 200; s¢, stem-apex; cof, cotyledon; in a, the boundary of the plerome is indicated by a heavy line. Fig. 71. Median, longitudinal section of an older embryo. Fig. 72. Similar section of a much older one; X 70. Fig. 73. Longitudinal section of an embryo from a nearly ripe seed; X 4o. Fig. 74. Two sections through the apical region of an embryo of the same age as the one shown in fig. 73. Figs. 75, 76. Median, longitudinal sections through the root of advanced embryos; f/, plerome; X 200. Fig. 77. Root of a full-grown embryo, showing the formation of inter- cellular spaces. Figs. 78, 79. Two transverse sections of the root just back of the apex; v, vessel; ev, endodermis; 7, intercellular spaces; * 400 ua regi: hy ee f Nitty aaah ¥ —_—. 4 PT og EE let ~< Pane fawAtan sch a 2eR BOT Vad. Pate IV. All the figures refer to Zannichellia palustris L. Fig. 81. Nearly median longitudinal section through the stem-apex; X 70. The actual apex, 2, was not included in the section, but its position is indi- cated by the dotted line; s#, the sheath below the node; Z!, Z?, the two ‘‘spathe leaves;’’ 7, lacunz in the leaves; 2, female flowers; ¢, male flowers; sg, squamulee intravaginales; a’, apex of a lateral branch. Fig. 82. Transverse section of the terminal buds. The leaves are num- bered; otherwise the lettering as in fig. Sr. Fig. 83. Longitudinal section of stem-apex; X 200. The apex, +, was divided into two parts, only one of which shows in the section. Fig. 84. Transverse section of the staminal primordium; X 4o0. Fig. 85. Longitudinal section of the stem-apex after its division into the carpellary branch, 2, and the secondary stem-apex, 7; X 200. Figs. 86-88. Development of the stamen seen in longitudinal section; x 200. Fig. 89. Transverse section of an older anther. Fig. 90. Longitudinal section of an older stamen, showing the peculiar terminal appendage; X 60. Fig. 91. Two nearly mature pollen-spores. In 4, the primary nucleus is dividing. Two free nuclei, £, are shown, derived from disintegrated arche- sporial cells; X 400. Fig. 92. Longitudinal section of young female inflorescence, showing the dichotomy of the plerome; X 400. Fig. 93. A similar section through the margin of an older inflorescence, showing the ovular rudiment, 0, and the subtending carpellary leaf, c. Figs. 94-96. Early stages in the development of the female flower in longitudinal section; X 4oo. 0, ovular rudiment; cay, carpel. Fig. 97- Section of the young ovule, showing the primary archesporial cell; X 400. Fig. 98. Cross-section of a young flower; X 400. car, Carpellary leaf; ¢, the primary tapetal cell. Fig. 99. Longitudinal section of an older flower; X 150. Figs. roo, ior. Longitudinal section of ovules, showing the later divisions in the archesporium; /, tapetum. Fig. 102. An older ovule, the nucleus of the embryo-sac already divided; X 400. Fig. 103. Longitudinal section through the inflorescence, showing its con- nection with the main axis; sh, the involuere. t eit = i ay (| “it 2 \ Say ASSES PY SH Seu meal wie aes! i tere tien N y 5 ea ont, [ sik a ea Nea Oy ai - PLATE V. All figures except 128 and 129, refer to Zanntchellia. Fig. 104. Longitudinal section of a nearly mature female inflorescence; X 40. Fig. 105. Longitudinal section of a nearly mature ovule; X 200. Fig. 106. Two sections of an embryo-sac after the formation of the egg apparatus and antipodals; X 4oo. Fig. 107. Lower part of a mature embryo-sac, with four antipodal cells; k, the lower polar (?) nucleus; X 350. Fig. 108. Egg apparatus at the time of ‘fertilization; #/, pollen-tube; 0, egg; 5, synergidz; X 4oo. Fig. 109. Antipodal region of mature embryo-sac. Fig. 110. Two sections of an abnormal embryo-sac, with numerous free nuclei; X 400. Fig. 111-116. Successive stages in the development of the young embryo; longitudinal sections X 400. The primary segments are numbered. Figs. 117, 118. Older embryos; X 200; s¢, stem-apex; cof, cotyledon, Fig. 119. Two sections of an abnormal embryo, in which, apparently, one synergid, s, was persistent, and adherent to the embryo; X 4oo. Fig. 120. Section of an advanced embryo; X 70. Fig. 121. Stem-apex of the same embryo; X 200. 7”, the second leaf. Fig. 122. The root of an embryo of the same age. rc. root.cap; ss? secondary suspensor. Fig. 123. Root of an almost full-grown embryo. Fig. 124. Longitudinal sections of nearly mature embryos; X 40; @ shows the stem-apex and root; 4, the hooked cotyledon. Fig. 125. Cross-section of an internode from the stem of a mature plant; X 70. Fig. 126. Longitudinal section of the apex of a root from the mature plant, showing the four primary tissues; plerome, f/; periblem, 4; dermato- gen, d; and calyptrogen, ca/; X 200. Fig. 127. Epidermal cells from near the root-apex, @, surface view; 4, a vertical section; X 400. The cells with dense contents are apparently secre- tory cells. Fig. 128. Embryo-sac of Sparganium eurycarpum Englm, with about ten free nuclui, and four antipodal cells. Fig. 129. Two sections of an embryo-sac of Sparganium eurycarpum Englm, showing extraordinary development of the antipodal cells, av/. Cee i a Laas epee ee aah Nee ‘tone STUDIES IN THE HERBARIUM AND THE FIELD.—No. 1. BY ALICE EASTWOOD, Curator of the Herbartum. CONTENTS. PLates VI-VII. 1G REPORT ON A SMALL COLLECTION OF PLANTS FROM THE WHITE SANDSORVNE We VIEXIC OM rea aiepecremitsatine sas e.ceraversishe tye aigjeseters 71 ee ON: SPURLESS HHORMSHOFP. A QUILEGTA Msc ceraieciecis sieutseisewieece «nee 76 III. THREE UNDESCRIBED CALIFORNIAN PLANTS......... 0.0.0. -0-00% 78 Vee bee VANZANITAS OF Mra oAMATEPATSH eo ccicicicic cc crcis « vfelels «stele eee ove 81 EEXPEANATION ROR) PILATES s ctctercsickerslavoreicvsieloler cleisie acicleveiersisis elais sveeisins 86 I.—REPORT ON ASMALL COLLECTION OF PLANTS FROM THE WHITE SANDS OF NEW MEXICO.! SomeE months ago, Professor T. D. A. Cockerell of the College of Agriculture and Mechanic Arts, New Mexico, sent me a small package of plants for determination. They were collected on the ‘‘ White Sands’’ in August, 1896. They hint of a very interesting flora and indicate by a close alliance with well known species and by marked differences from the same, a peculiar environment and probably some ! The traveler coming down Tularosa Creek, in the Sacramento Mts. of New Mexico, sees before him in the distance what appears to be the sea, with heavy breakers rolling towards the shore. As he descends to the valley, he gradu- ally realizes that the apparent ocean is motionless, and is above, not below, the level of the plain. Coming at length within a few miles of it, he sees before him what are to all appearances great banks of snow; and were it not for the intense heat, the illusion might be complete. Actually arriving at the banks, he finds nothing but pure white sand, piled up perhaps to a height of fifty feet above the plain, from which it rises abruptly, and continuing in un- dulating hillocks as far as the view extends. It is this remarkable formation that is known as the White Sands. Prof. C. H. T. Townsend has described it in similar words in an unpublished paper on the Distribution of Life in the Southwest and Mexico; and I believe almost any one would receive the same impressions on visiting the locality for the first time. L7r] November 23, 1897. 42 CALIFORNIA ACADEMY OF SCIENCES. [PRoc. 3D SER. degree of isolation. The interesting description of the country, which Professor Cockerell has written, throws considerable light on the probable history of these plants. Cnothera tubicula filifolia, var. nov. PLATE Vi, Fic: 2. Stems slender, suffruticose; older parts with grayish brown, shreddy epi- dermis; younger parts and leaves slightly glandular, hoary-puberulent, with spreading short white hairs; leaves sessile, becoming almost terete with the involute margins, 5-15 mm. long, 1-14 mm. or less wide, crowded at the ends of the branches and in small rosulate clusters in the axils; calyx-tube 3-34cm. long, funnel-form lobes spotted with dark purple, 1 cm. long, 4 mm. broad, with pointed tip 1 mm. long; corolla 34 cm. in diameter, yellow, tinged with white when withered, petals rhomboidal, slightly acuminate, upper margin wavy; stamens with anthers 1mm. broad, 7 mm. long, equalling the filaments; The exact extent of the White Sands I do not know, but they are probably about forty miles long and a good many miles across. They lie to the east of the San Andreas rangeof mountains. They are composed of pure gypsum, undoubtedly deposited from a salt water lake, which must have been shut off from the sea and by degrees have dried up. Prof. A. Goss, of the N. M. Agricultural Experiment Station, has pointed out to me, that as the lake dried up, the gypsum would be precipitated early, being comparatively insol- uble. Eventually the other salts would be deposited on the top of it. The more soluble salts have long ago been washed away, and we have remaining the beds of gypsum, which now rise considerably above the surrounding plain, the latter doubtless having been lowered by gradual denudation. While the plain itself contains a great quantity of gypsum, the banks are perfectly dis- tinct and well defined—as well defined as a miner’s dump. Somewhere in the sands, I am informed, there is a spring, and water is nowhere far from the surface. It might be thought that no vegetation would grow on pure gypsum sand, but there is a scattered growth, consisting of various shrubby and herbaceous plants. I even founda small poplar, which looked to me like Populus tremu- Joides, though not typical; the poplar of the surrounding country, at least of the Rio Grande Valley, across the mountains, is ?. /vemonfit, never P. tremuloides, at so low a level. One of the commonest shrubby plants on the sands is a Rhus, while a Bigelovia grows to a considerable size. When, last autumn, I visited this locality with Prof. C. H. T. Townsend, I was able to collect a few plants and insects; and my companion went somewhat further than I did, with the result of collecting two species of bees in quantity, and some other insects, which I had missed. Unfortunately, I was anything but well at the time, and we could not delay more than a short while, so that what was obtained was a mere fragmentary sample of the actual fauna and flora of the sands. T. D. A. COCKERELL. MEsILLa, N. M., Feb. 28, 1897. Bot.—VoL. 1.] ZEASTWOOD—STUDIES FROM THE HERBARIUM. 73 pistil extending 2 or 3 mm. beyond the stamens, stigma round and thick, more than 2 mm. in diameter; ovary 4-toothed at summit; capsule fusiform, contracted under the spreading, 4-toothed, reddish summit, 4-ribbed, splitting into four valves to within mm. of the base and connivent at the top, persistent and becoming woody; seeds dark brown, irregular in outline, becoming mucilaginous when moistened; one row in each cell. This differs from typical @. tub7cula chiefly in the very narrow leaves and a more condensed habit of growth, the slender branches becoming almost fasciculate. The pecu- liarities are probably due to poverty in the environment, which by slow starvation would bring about a reduction in the size and form of its organs. Cnothera albicaulis gypsophila, var. nov. PLATE VI, Fic. 2. Stems woody, corymbosely branched, canescent throughout, with dense, closely appressed pubescence; leaves narrowly oblong to lanceolate, acute, 8-20 mm. long, 2-5 mm. wide, cuneate at base, margin sinuate-dentate, with 3 or 4 short teeth on each side, midrib conspicuous, petioles 2 mm. or less in length; flowers axillary, 3 cm. in diameter; calyx with tube 3 cm. long, slender, divisions 15 mm. long, united in one or two sets, free at the tips and bases; petals white, rhomboidal, as broad as long, narrowed at base toa broad claw, slightly wavy on the upper margin; stamens shorter than the petals, about equalling the style; capsule 3 cm. long, almost perpendicular to the stem, variously curved or even inclined to coil, splitting almost to the base into four narrow valves; seeds 1-2 mm. long, ovate-lanceolate, acute, obtusely angled, usually mottled with purple, minutely tuberculate. The description and figure were drawn from specimens without roots. The older parts of the stems present a ragged appearance due to the shreddy epidermis and the narrow spreading valves of the empty capsules. The desert environment is suggested by stiff, dry stems, close branches and leaves, and rapid development of the fruit. The flowers show but little tendency to become pink as they fade. @nothera albicaulis is a variable species, easily recog- nized amid all its forms. Its most constant and distinctive characters are the white, shreddy epidermis, the capsules sessile by a hard, broad base, spreading from the axis so strongly as to make the stem somewhat zig-zig, valves sep- arating almost to the base and widely spreading. 74 CALIFORNIA ACADEMY OF SCIENCES. [PRoc. 3D SER. This variety is nearest to § ¢7¢chocalyxw Engelmann,! which has been named Anogra pallida Engelmanni Small.” The description of this variety is meager, but the erect and nu- merous branches of my specimen do not suggest ‘‘ parce ramosa,’’ nor are the seeds obtuse, as stated on page 334, Am. Journ. Sci. (II), Vol. XXXIV, under @. albicaulis. The leaves approach those of var. v drevzfolza in size. The plate was made by tracing after a photograph. It has, therefore, greater accuracy than would be possible with an ordinary drawing. The flower, however, was drawn from a dried specimen soaked and spread so as to show its shape. Bigelovia graveolens appendiculata, var. nov. PLaTE VI, Fie. 3. Stems light green, slightly tomentose except at the densely tomentose leat axils; leaves involute, becoming terete, mucronate; heads in rather few- flowered, corymbiform cymes on pedicels, often divaricately spreading; bracts of the involucre carinate, ciliate, and scarious on the margins, obtuse or acute at the woolly apex; corolla with from one to four yellow, linear appendages of various lengths on the tube, and a few cobwebby hairs on the border. This differs from all other forms of this variable species in the peculiar appendages of the corolla-tube. They are not transformed pappus bristles, being too far up on the tube, rather appearing like an outside corolla, as if the corolla were trying to become double. In other respects it approaches var. hololeuca Gray. Thelesperma gracile Gray. Bidens gracilis, ToRREY, Ann. Lyc. N. Y., Vol. II, p. 215. ffTabitat.—On the Canadian River. The fragmentary specimen from the ‘White Sands”’ shows but little variation from the type; the heads are smaller than those of any of the species represented in the Herbarium of the California Academy of Sciences. 1Am. Journ. Sci. (II), Vol. XXXIV, p. 335. 2Torr. Bull., Vol. XXIIT, p. 176. Bort.—VoL. I.] ZEASTWOOD—STUDIES FROM THE HERBARIUM. 75 Muhlenbergia pungens 7hurder.! This is, according to C. J. Croft,’ a valuable forage plant, and is known in Arizonaas Black Gramaand Grama China. It was determined by Prof. Scribner, who reports it as ‘¢ without unusual characters or special peculiarities.’’ There is also a branch of a shrubby Labiate, whose leaves in form and flavor suggest an Audzbertia; but as it is without flowers or fruit, it cannot be identified. 1Proc. Phil. Acad. Sci., 1863, p. 78. 2 Proc, Cal. Acad. Sci. (I), Vol. III, p. 205. 76 CALIFORNIA ACADEMY OF SCIENCES. [PRoc. 3D SER. IIL.—ON SPURLESS FORMS OF AQUILEGIA. Aquilegia differs from allied genera chiefly in having five spurred petals and a column of staminodia surrounding the pistils. In some species a tendency to degenerate from the markedly specialized type exists, and the varieties are characterized by spurless petals. These cause the flowers to superficially resemble those of /sopyrum, Clematis, or Anemone, more than Aguzlegia. Double-flowered forms of Aguz/egia are quite common in some species under cultivation, the stamens either becom- ing distinctly petaloid with spurs, or, together with the pet- als, becoming sepaloid, flat, and destitute of spurs. Agquilegia vulgaris, which is widely spread through Eu- rope and has, probably, been more extensively cultivated than any other species, has several varieties with ecalcarate petals. Two are given in De Candolle’s Prodromus, Vol. I, p. 50: A. stellata, with double flowers, petals flat, spur- less, and colored; A. degener, with double flowers, petals and sepals flat, spurless, and green. These are both figured in Clusius’ ‘‘ Historia Rarorium Plantarum,’’ according to De Candolle. In the ‘‘ Index Kewensis ’’ another variety, A. ecalcarata, is given, which the name indicates to be spur- less; but it is regarded by horticulturists as identical with A. stellata. The attention of the writer has been called to these pecu- liarities by the occurrence of two species in Colorado, which have ecalcarate varieties. The first of these belongs to A. cerulea. Aquilegia cerulea Daileye, var. nov. PLATE VII, Fic. 1. This is in all respects similar to the type except that the spurs of the petals are entirely wanting, sepals and petals flat, blue. The plate indicates the peculiarities of the variety. The plant figured was drawn from specimens sent me by Miss Bot.—Vot. I.] ZEASTWOOD—STUDIES FROM THE HERBARIUM, "77 Anna L. Dailey. They were collected at Evergreen, Col- orado, where Miss Dailey noted them during several years. It was her opinion that the plants had become more numer- ous since she first noticed them. Mr. George E. Osterhout of New Windsor, Colorado, reports this form as frequent in Estes Park, Colorado. Some years ago, the writer found several plants with the spurless petals growing among the typical form in Platte Canon, Colorado. In the ‘‘ Gardener’s Chronicle,’ Vol. XVI, 1881, p. 16, the following note occurs concerning A. cerulea: “Mr. E. G. Loder sends us a spurless flower of A guzlegia ceru/ea quite similar to the European stellate Columbines. The long, white, spurred petals are, in this case, absent and replaced by an additional whorl of ovate, acute, blue sepals. A few years since, Mr. Loder tells us he ‘ collected seeds of A. cerulea at an elevation of 10,300 feet, not far from South Park, Colorado.’ The seeds were sown in North- amptonshire in 1879. BR.)in < «2 « w, «= cl oss) | s “3 13. Brodiwa cafitata BENTH. ....-...-..+425 Sent [Es lee bas = is = 14. Partetaria debilis FORSTER. . 2 2. ee ee ee | Re Tt ae ee * 15. Chenopodium Californicum WATS... +. +++ +++++| * | ance: * * oy 16. ‘Chenopodium tur ale Tis. «is, sw ss 0) 2s 0 6 est ease * |) ee x 17. Atriplex: Caltjornicum MOO... 23 012s oe 0s ss * * * | * aa! 18. Atriplex decumbens WATS. « ss 22 3s 1s 2 oo 0 + | * y 19. Suedasp.? (S. Torreyana ?) foundon .....+..4. ¥ sed * * | * 20. Aphanisma biitotdes NUTT... . ~~. 2 ee eee eee | + [| 21. Abronia maritima NUIT. « . 55 6.6 a es ns sd = * | * | Bor.—Vot. I.] EASTWOOD—STUDIES FROM THE HERBARIUM. 119 2 e 2 [3 Qa SPECIES. 8 Ss $ £ a B/E) 8 /E/8 a |e AZ A OY ONS MINDEU ALA PLZ AM Ate) oy (o)i@) eles) (6) 616) 6), © iv! well © * 4 3 23h A 07-2 %a: ZIOAiSPAMON hea; «(ey «: “ev ie) (sss. oe) s/s : : | Oy QUE? (CoHem oo. 8) Gp ole co decen. os Geotoucee Geoet ts les x by} * 25. Spergularia macrotheca HEYNH. (Tissa.). . es = * * * 26. Platystemon Californicus BENTH......-.-+-.+..-.-| * a * * * 27. Lepidium bipinnatifidum DESV. (L. Menztesii of Bot. Cal.) * BONE LEPIGLUME HIM AMINZNULT 11 fo tals) encase fers csi Fa) hs) fe Memes! 3 * * * 29. Dithyrea Caltfornica var. marttima DAVIDSON (Biscutella) 30.1 Oligomerts glaucescens CAMBESS. (O. subulata WEBB.) .. * * * * Bl melee PMU INICTARERUS DOUGIi. ovici oe +) ee) vi «sis | = i! 32. Lupinus albtfrons BENTH. (L.Chamissonts)........) * | * | Vs 33. Trtfoltum microdon H. & A. var. pilosum var. nov.. . RETO L/OLEMINE PRICVOROIE <) otto (3) ro) aula) otc) “2 Iw (evs: Koy loo sinters: “atc tA * SA Liot/ OLSMIN per Ct eMt eM, We SC Gis) leivasie) We ls) aice sce) ie) | * B50 Lrsfoltum: stenophyllum NUTT. s 6 <3 6 3% 55 ots BOL ELiet (OMIM BIMEZE VATS ric) eitcewietaulel cits) fie) ov ee) ie 37-4 Trifolium dichotomum H.& A... 6. 2s vee ee ws | Bom CAicaponSsatioailgsje ts centedc| ollsliel 'silevsyitesiie vo) telisi-a veo] eS 39: -WMedwago-denticulata WILLD...) 2. 1s se ces es | a ey ed 40. Astragalus didymocarpus H. & A... 1... 2.2.2.2 ae | ¥ < Alot SEZ GSAS 1 7ASKEE SD NOW) ie weve) et) ioe hci vs sens AQ: LL OSACKLEG VENUSTA'SP.;TOVwie;< sts csv sis: © eee) #1) we AS Lerodiumn Ciculartum TELER ss) s\ sos $66.5) see 1c «1 ee ne * z * 44. Malva pusilla SMITH (M. borealis)... 2... eee ee * - LEI * A5n \ Unothera viridescenS HOOK. « oi vs + «0 0 6 6s tg * 46. Mesembryanthemum nodifiorum We. 1. 1 ee ee es * * 46a.5 Mesembryanthemum crystallinum\.. 2... 2. see ee hie x eee |e * 47. Opuntia Engelmanni SALM. var. dittoralis ENGELM?. . . | * i “ud * 47a5 Opuntia prolifera ENGELM.......-..2..-.% A ABs) AHLUS SE PUSTILUS” MIICEEE. jel tote ce) wie) fee] io els ft oe) ot ele * = a * * 49. Apiastrum angustifolium NUTIT..............| * = ze Osten S AMICMLA IM enatest, FLY Stes. |.) s) s/s) ovo) ves stn erieice * Slee eucedansrge t#S1lare Sp. MOVs. «= 2 se fs 6 0 so i Sasa GtiLaPN CULNIEAG RAT Ure aiycti ticie acts. [or ci saeaemonen ee bd 5 * * 53-1 Convoluulus macrostegius GREENE. .... ...... es * bet ta 54. Heltotropium Curassavicum\,.. .. 1... 2... 220 A * - * SMM ECLOLRP-VA VENER IS IEG a) se) eye) 2 aos) Wiel «elle! 6) ele) eh eh | “purymey 120 CALIFORNIA ACADEMY OF SCIENCES, [Proc. 3D SER. Sl Be) Bre Wie SPECIES. Beil ae || 1S £ g | 5 B/E) 1E]8 56.1 Cryptanthe marttima GREENE... +++ s+ eee ees 57. Cryptanthe Torreyana GREENE .....-.-: - beret * io 58. Amsinckia St. Nicolatsp.NOV......+-++-+s | 59. Amsinckta maritima Sp. NOV... 66 6 eee ee ee eae Jase, 60. Lyctum verrucosum Sp.NOV. ... ++ ++. Prteign ie yon 61. Lycium Californicum NUIT... - + +++ eee ee eee * * che 62. Orthocarpus purpurascens BENTH... . +. ++ +s 2 ee * * me 63. Plantago insularisSp.NOV.. ~~... ++ 2 eee eee 64. Malacothrix indecora GREENE... 1 eee ee ee eee Be 65. Malacothrix implicatasp.N0V... 2... eee eee ee] * * = 66. Microserts lineartfolia GRAY. ..-....++-.+-. ga tc = * * * Oy, Sonchus aster’ VOL). yah aos a fms, Seren ee ee | | ba! * 3 68. Samchus: demerrr7rg2es! Tye xn co: is te qa) wl se fo ta ah tee wt ve) | | | it GQ. Sonchus oleracens: Ty. xcs. .c: eile ©)/s0,s) sl ose poysi leis | * * * 70. Centaurea MelttensisL,.?. 2. 21 eee eee ee ne * TMi, AQLEMEESTE. Broa oe © Feel a we '8y os seks ve) | ami ote, os Ted ta ioe 92, Achillea Millefoliurie Via cs ee est atta ceneetptan | Ue * * © * * 73.2 Amblyopappus pusillus H.& A... we ee ee a Fe fay ensaes| ieee * * * * 74. Baeria gracilis GRAY. ........ See ee ce ree * cp * * 75° Hemisonia Streets GRAY. . 00 6 5 6 eS ee tw * * 76.1 Leptosyne gigantea KELLOGG. ... 3 +222+2+-| * * € * ¥ 77. Franseria Chamissonts LESS. var.visctda var. nov. . . * Rranseria: Chamissonts WESSi«, osc) os a ee ee ee] | * 78. Franseria bipinnatifida NUTT. var. dubia var.nov. ... . | | Eo PL Omtanaitjida sa aw ean Se ee: teu eons | eee * | * 79. Baccharis consanguinea DC. ? (BK. pilularis). . . 6... * * fos * 80. Bigelovia veneta Or Grindelia?....... a ahert ace *Closely related species found on some of the other islands but different from the form on San Nicolas. 1 Guadalupe Island, Baja California. ® Cedros Island, Baja California and S. America. 8 San Benito Island and Anacapa Island, Calif. 4 This might more correctly be considered a form of 7. Macref with the heads on long peduncles. 5 Species reported but not collected. SUMMARY. 82. Species recorded. 64. In common with the mainland, of which about 30 are Californian. 53. In common with Santa Catalina Island. 48. Incommon with Santa Cruz Island. 31. In common with Santa Rosa Island. 31. In common with San Miguel Island. 6. In common with Guadalupe Island; 1 with Cedros Island, 1 with San Benito Island and 1 with Anacapa Island ; 7 species and 3 or 4 varieties peculiar to San Nicolas. Bot.—Vot. I.] EASTWOOD—STUDIES FROM THE HERBARIUM, 121 II.—NEW SPECIES OF CNICUS FROM SOUTHERN COLORADO AND UTAH. 1. Cnicus bipinnatus, sp. nov. Cnicus Drummondii var. bipinnatus Eastwoop, Zoe, Vol. IV, p. 8. Glaucous and glabrous except for some slight arachnoid tomentum on the stems, petioles, and involucral bracts: stems stout, erect, leafy, 6 dm. or more high, branching from the base and also above; leaves with numerous linear-lanceolate divisions which are 2-6 cm. long, 5 mm. wide and irregularly parted near the base, generally on one side only, into similar lobes varying in length and sometimes as long as the main division, margin laciniate-dentate, spiny, lateral spines 2 mm. long, terminal 4-5 mm. ; radical leaves petiolate, 1%-3 dm. long, 6-10 cm. wide; cauline, sessile, 10-15 cm. long, 5-6 cm. wide: heads corymbose at the ends of the leafy branches, almost sessile, narrow, cylindrical, 4% cm. long, 1 cm. or more wide; involucre of appressed, imbricated bracts, successively shorter, in seven ranks, the lower ones pointed with a weak prickle 3-5 mm. long, the upper attenuate to a scarious tip, minutely puberulent; flowers purple; corolla with tube about half as long as the throat and divisions, throat about one-third as long as the linear, clavate-tipped divisions ; stamens surpassing the corolla; style straight with the node 2% mm. from the tip; akenes glabrous, shining, flattened, obovate- oblong, 6 mm. long, 3 mm. wide, surmounted with a yellow ring. This is nearer to Cuzcus Reothrocki than to Cnicus Drum- mondi but differs from the former in its foliage, narrower heads, short, weak spines, and general appearance. It formed a clump two feet or more in height and almost two feet in diameter. ‘Collected in Colorado in Johnston Cafion, near where it joins the Mancos River, in a locality but rarely visited by white men. Closely related to this is the plant which the writer! described under the name C. Hothrockit var. diffusus. I take this opportunity of giving this, too, specific rank. 2. Cnicus diffusus, sp. nov. Similar to the above in habit and surface: leaves narrower, less deeply divided, with more regular and triangular lobes; spines at the tips of the lobes 1 cm. long, stiff, yellow, those along the margins 1-3 mm. long: heads somewhat broader than the last, with outer involucral bracts tipped with stiff spines from 1-2 cm. long, deflexed-spreading in fruit, inner bracts attenuate 1 Proc. Cal. Acad. Sci., Ser. 2, Vol. VI, p. 303. 3 Sept. 13, 1898. I22 CALIFORNIA ACADEMY OF SCIENCES. [PRoc. 3D SER. to along spiny point; flowers purple; corolla with tube four-fifths of the throat and divisions ; throat about one-third of the tube, divisions linear, abruptly acuminate, 18 mm. long; stamens surpassing the corolla; node of the stigma 2 mm. from the apex ; coma 2% cm. long. Willow Creek, San Juan Co., Utah, Aug. 13, 1895. 3. Cnicus Hesperius, sp. nov. Stem stout, simple, erect, leafy from the base, ribbed, almost glabrous, 3-4 dm. in height, 2 cm. in diameter : leaves ro-15 cm. long, 1%-2 cm. wide, upper surface glabrous, lower tomentose, ascending, linear-lanceolate, with numerous rounded lobes, spiny-margined with two to three large, yellow, subulate spines, 5 mm. long, and several shorter ones: heads sessile, crowded, in an oblong, terminal, erect, leafy, glomerule; involucre 2 cm. long, 2% cm. wide, with bracts 3 mm. broad at the yellow, glabrous, ovate base, tapering to a long brown-purple spine 1 cm. long, arachnoid with silky wool except at the glabrous yellow apex: flowers light purple ; corolla tube almost equalling the throat and divisions ; throat fusiform, contracted under the divisions which are linear with thickened apex and about half as long as the throat; anthers surpassing the petals by 3 mm., sparingly arachnoid except at the pointed tips ; style with inconspicuous node concealed by the anthers and stigma, exserted 2 mm. ; coma 1 cm. long. Mt. Hesperus on the Bear Creek Divide above timber line, La Plata Mountains, southwestern Colorado, Aug., 1892. Collected by the writer. This Cnzcus is nearest to C. erzvocephalus Gray, under which it was placed by the writer in *‘ Zoe,’’ Vol. IV, p. 8. A recent inspection of the specimens of Cyzzcus in the Her- barium of the Academy has convinced me that this deserves to have specific rank and that it is not a hybrid with Cuzcus Parryz, as | had formerly conjectured. The involucral bracts are less densely tomentose than in C. ertocephalus, the glomerule erect, the flowers light pur- ple or pink, the entire plant less arachnoid, and the stamens have not only the filaments wooly but also the anthers. I have named it in honor of the mountain on which it is found, the highest in the La Plata Range. There were few individuals growing along the trail leading to the sum- mit of the ridge. Bot.—Vo.t. I.] ZASTWOOD—STUDIES FROM THE HERBARIUM. 123 III.—THE COLORADO ALPINE SPECIES OF SYNTHYRIS. One alpine species of Syuthyrzs is now recognized from the mountains of Colorado. This has been aptly named S. alpina. 1. Synthyris alpina Gray. PLATE IX, Fics. 1a-1d. This is found on most of the high peaks of Colorado, growing above timber line, in loose, rocky soil. Its low stature and short spike of dark purple flowers superficially distinguish it from the other species. There are always four calyx divisions, variable in shape and size even in flowers from the same spike, and conspicuously fringed with long white hairs. The corolla consists of two parts, united at the base; the upper broadly obovate, entire, and slightly concave; the lower two or three cleft, with laciniate or entire divisions varying in length and breadth. The sta- mens and style are moderately exserted and the flowers are erect. In the little-explored mountains of southwestern Colorado two different species are found above the tree limit. One of these is so distinct that in a group so closely related as that to which S'yxthyr7s belongs, it might be taken as the type of a new genus; the other approaches S. a@/pzna. The placing of these two species in Synthyr7s necessi- tates a change in the generic characterization of the calyx. Instead of ‘‘calyx with four divisions’’ it must become ‘‘calyx with two, three, or four divisions.’’ This difference in the number of calyx divisions arises probably from a union of parts not infrequent in other genera. 2. Synthyris Ritteriana, sp. nov. PLATE IX, FIGs. 2a-2e. Sparingly pubescent with short, scattered hairs ; scape stout, nearly 3 dm. high, closely covered from the middle of the stem to the spike with foliaceous bracts : leaves radical, 8-10 cm. long, 344 cm. wide, oblong-elliptical, obtuse, 124 CALIFORNIA ACADEMY OF SCIENCES. [PrRoc. 3D SER. cuneate and decurrent at base, crenate ; petioles stout, 8-12 cm. long ; bracts of the scape alternate, 114-24 cm. long, 10-18 mm. wide, ovate-acuminate, sessile by a subcordate base, crenate-dentate near the apex; floral bracts rhomboidal, acuminate, tapering at base to a short petiole which surpasses the pedicels : flowers white, erect on short pedicels in a spike 5 cm. long; calyx of three divisions varying in shape and size in flowers from the same spike, 4mm. long, orbicular to obovate, entire, toothed or cleft, obtuse or acute, fringed with white hairs less dense and shorter than those on the pre- ceding species ; corolla of two divisions, surpassing the calyx by 2 mm.; upper part broadly obovate, acute, indistinctly spurred at base, ciliate; lower variously cleft with two to three laciniate, ciliate divisions ; stamens with orbicular-ovate anthers I mm. broad and filaments inserted at the base of the corolla, surpassing it by 2 mm.; fruit unknown. Collected by the writer in Cumberland Basin, La Plata Mountains, Aug., 1892. It grew in the alpine meadow where moisture was abundant. It is named in honor of Mr. and Mrs. B. W. Ritter of Durango, Colo., to whose kindness I owe the opportunity of visiting these mountains. The plate shows the differences in the floral organs of the three species, all drawn to the same scale. 3. Synthyris reflexa, sp. nov. PLaTeE IX, Fics. 3a-3d. Glaucous and somewhat viscid; scape erect, 1%-2 dm. high, clothed below the spike with broad, foliaceous bracts: leaves radical, 7 cm. long, elliptical-oblong, obtuse, truncate or cuneate at base, finely crenate, thin in texture ; petiole 3-5 cm. long: bracts of the scape alternate, crowded, 2%4 cm. long, 2 cm. wide, sessile by a broad subcordate or truncate base, broadly ovate, acute, entire or serrulate near the apex, diminishing upwards; floral bracts ovate to linear, fringed with long white hairs : flowers. greenish white, reflexed on pedicels 1 mm. long; calyx of two divisions 5 mm. in diameter, orbicular, fringéd like the bracts; corolla of two parts united at base, the upper broad, somewhat hood-shaped, 5 mm. broad, 8 mm. long, sparingly fringed ; the lower two-cleft 5% mm. long, 4mm. wide, divisions laciniate, 2 mm. long; stamens two, anthers cordate, with cells not con- fluent, filaments inserted at the base of the corolla, surpassing it by 4 mm.; style shorter than the filaments, stigma capitate ; fruit unknown. Collected by the author in Kendall Basin, in the San Juan Mountains, near Silverton, Aug., 1890. It was quite rare. This is the only Syzthyr7s known to have reflexed flowers and two divisions to the calyx. Bot.—Vot. I.] EASTWOOD—STUDIES FROM THE HERBARIUM. 128 It is very doubtful if the genus Syxthyr7s will stand in the future as it is now set forth in the Synoptical Flora; nor is it more certain that Professor Greene’s transferrence of all the species to Wadfenza will be final.t In a group of genera so closely related and so polymorphous as those included under subtribe Veroniceze in Bentham and Hooker’s Genera Plantarum, botanists will always differ in regard to generic limits. A difference in the arrangement of the leaves and habit _of growth primarily separated Walfenza from Veronica; the form of the corolla, the number of calyx divisions, the dehiscence of the anthers and the shape of the seeds sep- arated Synthyris from Wulfenta. Professor Greene has shown the worthlessness of considering the number of calyx divisions as a generic character, and his position is reinforced by the two new species described in this paper and by two described under Wudfenza by Aven Nelson. Professor Greene, however, does not give a diagnosis of Walfenta nor tell why it, also, should not be included under Veronica. If habit of growth is taken as a generic distinction, then Synthyris naturally falls into two genera, one containing S. rotundifolia and S'. reniformis; the other, the remain- ing species and perhaps the original Walfenca. If other characters are to be taken, many genera would result and the synonymy become much involved. It seems simplest to follow Bentham and Hooker and Gray and leave these uncertain problems to the future; so the new species of the Colorado mountains have been described as Synthyris. 1 Erythea, Vol. IT, p. 80. 2 Torr. Bull., Vol. XXV, pp. 281, 282. 126 CALIFORNIA ACADEMY OF SCIENCES. [Proc. 3D SER. IV.—FURTHER OBSERVATIONS ON THE MANZANITAS OF MT. TAMALPAIS. Since the publication of the article on the ‘‘ Manzanitas of Mt. Tamalpais,’*! I have been able to explore parts of the mountain and the neighboring hills hitherto unknown to me. Scarcely a week has passed that I have not spent Sunday on the hills of Marin County. As a result of these frequent visits and the new territory explored, new facts concerning the distribution, time of blooming, and characteristics of the different species of Arcto- staphylos have been observed. 1. Arctostaphylos nummularia Gray.— This I have found only on Mt. Tamalpais proper. It does not seem to grow on the hills around Fairfax, where the Big Carson and the Little Carson Creeks take their rise, nor on the ridge between Bear Valley and Bolinas. It is abundant on the Boot-jack and Throckmorton Trails, also on the trail leading directly from Mill Valley up the mountain and on the trail from the Potrero to the head of Cataract Gulch. While well formed fruit can be found at almost any time of the year, ripe fruit is seldom seen and never persists as does the fruit of all the other species. 2. Arctostaphylos canescens /ustwood.— This, too, seems to be confined to Mt. Tamalpais and apparently grows only on the southern slope. It loves bleak, gravelly hillsides, where it often holds exclusive possession. On the Throckmorton Trail a few bushes are to be seen, a few more on the Bill Williams Trail; but on the West Point Trail it is abundant, adorning the slopes during the winter with its blooming bushes. On November 7th, the first flowers were seen. The plants on the Throckmorton Trail were beginning to bloom. The delicate pink blossoms have their beauty much enhanced by the gray-green foliage. From this date until See this Vol., p. 81. Bot.—Vot. I.] ZASTWOOD—STUDIES FROM THE HERBARIUM, 124 February 20th, it was seen in bloom every Sunday. Well formed fruit was observed on the bushes on the Throck- morton Trail before those on the higher West Point Trail had ceased blooming. Some plants only a few inches high, easily pulled up by the roots, and with but one or two branches, were laden with blossoms. The number of these small plants indicates a species full of vigor, seeking new territory for habitation. The roots of the larger plants are spreading rather than deep and are easily pulled up. Dur- ing May the berries become ripe, and before the other spe- cies show ripe fruit this has usually lost all its berries. 3. Arctostaphylos glandulosa astwood.—On Decem- ber roth the first plant of this species was found in bloom. It may have been blooming also elsewhere on the mountain; but two weeks earlier flowers were not seen on the plants near the foot of the Boot-jack Trail, where I have always found them first in bloom. On May rst, fully grown fruit was observed on bushes on the Boot-jack Trail. This species is the most abundant, found everywhere on the slopes and ridges of Mt. Tamalpais and the Fairfax hills. I am still uncertain whether two species ought to be recog- nized where I have described but one or whether, with all its forms, it ought not to be considered a variety of A. tomentosa. 4. Arctostaphylos montana Fastwood.—This has a wide range in Marin County. It is usually found only on the uplands and is especially partial to the bluish gray vol- canic areas. Wherever Cupressus Goveniana and Quercus dumosa var. bullata are to be found this also will be seen. Besides Mt. Tamalpais, it is found on the Fairfax hills, where it covers large sections of country, being the most abundant manzanita there. It was first seen in flower February twenty-second, on the trail to the source of the Big Carson. Only one bush was in bloom, situated in a warm and sheltered spot. On May rst it was still blooming 128 CALIFORNIA ACADEMY OF SCIENCES. [Proc. 3D SER. on Mt. Tamalpais, but evidently nearing the end of its season. Having had an opportunity of examining fresh specimens of A. Hooker? in bloom, sent me from Monterey by Miss Marion Rouse, I am confirmed in my, belief in the identity of this species as distinct from A. Hookerz. The flowers of the latter are much smaller, the leaves not so thick, and the fruit, as noted before, has much thinner pulp. Bot.—VOL. I.] EASTWOOD—STUDIES FROM THE HERBARIUM. 1 29 V.—TWO SPECIES OF ERIODICTYON HERETOFORE INCLUDED UNDER ERIODICTYON TOMENTOSUM. The genus Hriodictyon was founded on &. crassifolium, collected near San Diego, Calif., and described by Ben- tham in the Botany of the Sulphur, page 37. Previous to this, another species had been described and figured by Hooker and Arnott as Wigandia Californica.' In the Bot- any of the Sulphur, Bentham transferred this to Zrzodictyon, giving it the appropriate name £. glutinosum. It again suffered a change of name when Torrey? restored its earli- est specific name, calling it &. Californicum. Bentham described still another species in the Botany of the Sulphur as &. tomentosum, differing from £. crassz- folium in having pedunculate dense cymes, broader leaves, and very numerous, smaller flowers. This Dr. Torrey took as the type under which &. crasszfolium should be included, and suppressed the former name, giving, as his reason, the numerous intermediates observed by Dr. Parry and him- self around San Diego and elsewhere. Gray accepted Torrey’s view with the remark: ‘* Z. cras- sifolium Benth. was doubtless rightly united with this by Dr. Torrey, and this name should have been preferred, but the other is good and of the same date.’’$ Greene described a plant from Monterey County, Calif., collected by Mr. Brandegee in 1885, as true &. tomentosum, and restored the name £. crasstfolium to the plant of Southern California.* The attention of the writer was attracted to this puzzle by the rediscovery of Mr. Brandegee’s plant near Jolon, in the San Antonio Valley, Monterey County, Calif. Truly it seemed that Professor Greene had solved the mystery; but to be certain, specimens of this from Monterey County and of the large flowered &. crasszfolium from San Diego were sent to Kew for comparison with the types. The following 1 Bot. Beech., Pl. LXXXVIII, p. 364. 2 Bot. Mex. Bound. Sur., p. 148. 3 Proc. Am. Acad., Vol. X, p. 331- 4Bull, Cal. Acad. Sci., Vol. I, p. 201. 130 CALIFORNIA ACADEMY OF SCIENCES. [PRoc. 3D SER. reply was received: ‘* Eriodictyon crassifolium and E£. tomentosum are conspecific and your plant named /omen- tosum is apparently an undescribed species which we also possess from J. G. Lemmon without locality, and from G. R. Vasey, Monterey, 438, coll. July, 1880.”’ Since Lemmon, Vasey, and Brandegee have all passed through Jolon when on their way to the Santa Lucia Mountains in search of Adbzes bracteata, doubtless the region of the writer’s collection of this species coincides with theirs and will be considered as the type locality of the new species. 1. Eriodictyon niveum, sp. nov. PLATE X, Fics. 3a-3d, Densely white-tomentose or flavescent; stems three or four feet high, growing in clumps, very leafy below the inflorescence: leaves thick, elliptical- ovate or obovate, 4-6 cm. long, I-3 cm. wide, entire or crenate, except on the cuneate base, apex acute or obtuse, lower surface reticulate-rugose, upper with veins scarcely discernible; petiole broad, 5-1o mm. long: panicle terminating a long, naked peduncle, compactly or widely branched, the stout branches varying much in length on different plants; cymes densely flowered, with the lower bracts obovate or spatulate, tapering to broad petioles, the upper oblanceolate to linear; flowers small, almost sessile; calyx equalling the corolla tube, divisions linear-subulate, densely clothed with white silky hairs; corolla white or tinged with lavender, 4 mm. long, urceolate, glandu- lar-hirsute externally, glabrous within, tube furrowed longitudinally, slightly contracted under the five small spreading lobes; stamens with the free portion short, inserted below the throat, anthers oval, 1 mm. long; styles shorter than the sepals; capsules orbicular, obtusely 5-angled, tomentose, especially on the angles; seeds four or sometimes five, brown, minutely favose, variable in shape, often keeled, more than 1 mm. long. Collected by the writer near Jolon, Monterey County, Calif., in flower June 1, 1893, and in fruit Sept. 22, 1894. The following specimens, besides, are in the Herbarium of the Academy :— Lobb, San Antonio Valley (date not given); T. S. Bran- degee, Monterey County, 1885; Dr. Palmer, ‘‘ Perhaps from Monterey County,’’ 1876. There is also a specimen without flowers or fruit collected by T. S. Brandegee at Zapato, Fresno County, Calif. The foliage and pubescence are exactly of this species. Bot.—Vot. I.] EASTWOOD—STUDIES FROM THE HERBARIUM. 131 2. Eriodictyon Traskiz, sp. nov. PLATE X, Fics. 2a-2c. Densely white-tomentose, except the dark-colored, glandular-hirsute calyx: leaves elliptical, 5 cm. long, 15 mm. wide, acute at apex, the base narrowed to a petiole 5mm. long, margins dentate except near the base, veins distinct on the lower surface, barely evident on the upper: panicle slightly surpassing the leaves, with branches spreading or curving upwards, rather slender, glandular and tomentose, bracts from elliptical with dentate margins to lan- ceolate or linear with margins entire; cymes densely flowered: flowers small, on pedicels 1 mm. or more in length; calyx divisions five, narrowly linear, not uniform in length, 4-5 mm. long; corolla purple, the tube equalling the calyx, 5 mm. long, contracted at base and throat, furrowed longitudinally, divisions of the limb irregularly orbicular, not uniform in size, glandular- hirsute externally as well as the upper part of the tube; stamens inserted half way down the tube, almost sessile; style branches glabrous, 114 mm. long; ovary ovoid, glandular-hirsute; fruit unknown. This was discovered May, 1897, on one volcanic upland on Santa Catalina Island, Calif., at an elevation of about 1500 feet, by Mrs. Blanche Trask, the indefatigable local botanist in whose honor it is named. Probably, this is the plant collected by Lyon on Santa Catalina, referred by Dr. Gray to E&. tomentosum.’ It approaches &. nzveum but is undoubtedly a distinct species. A plant slightly differing from the above was collected in the Santa Inez Mountains, Calif., by T. S. Brandegee, in 1888. The peduncles are stout, with thicker and more spread- ing branches, leaves larger and coarser, pedicels much longer, 4 mm. long, while the filaments are short but dis- tinctly evident, the corolla has the same shape, the lobes of the border being not quite so broad. As this, too, is in flower only and very young, comparisons of the fruit and seed cannot be made. The plate shows the floral organs of 7. crass¢folium from San Diego, part of the collection sent to Kew, and also those of &. niveum and &. Traskie, all drawn to the same scale, five times the actual size. 1 Suppl. Synop. FI., p. 420. 132 CALIFORNIA ACADEMY OF SCIENCES. [PRoc. 3D SER. VI.—NEW SPECIES OF PACIFIC COAST PLANTS. 1. Campanula angustiflora, sp. nov. PLATE XI, FIGs. 2a-2¢. Annual, scabrous-hispid; stems slender, angled, 1-3 dm. high, branching diffusely from the base to the top with upwardly spreading branches: leaves sessile, ovate to orbicular, acuminate or acute, deeply dentate, 5-15 mm. long, 3-10 mm. wide: flowers axillary on stout, upwardly spreading pedun- cles, twice to four times the length of the flower; divisions of the calyx linear- subulate, almost equalling the corolla, connivent in fruit; corolla tubular, with five triangular lobes; stamens included, the anthers linear, longer than the thin, broadly triangular-subulate filaments; ovary obovoid, slightly con- stricted at the apex, ribbed; style short, thick, with three revolute stigma lobes; fruit strongly ribbed, irregularly humped with the three valves above the middle; seeds numerous, minute, shining, light brown, with a small darker spot at one end, 3-sided or keeled. This has been included under C. exzgua Rattan by Mrs. Brandegee, who first discovered it on Mt. Tamalpais, July 5, 1886, collecting it again in the same locality June, 1890, and July, 1893. She also found it on Mt. St. Helena, May to July, 1889.! It was rediscovered on Mt. Tamalpais by Mr. J. W. Congdon, and by the writer near the water tank at the head of the East Fork of Sequoia Canon on the rail- road track. Besides the points of difference shown by the figures of the two species, there are differences in habit of growth and general appearance. C. exzguwa is lower, more slender, less branched, and with the branches divaricately spreading ; the leaves are smaller and narrower, and almost hug the stem. The figures are drawn from a specimen collected by Volney Rattan on Mt. Diablo, Calif. It is probably part of the type. In the Herbarium of the Academy there are specimens from Mt. Hamilton, Calif., collected by W. W. Price, similar to those of the Mt. Diablo Campanula. From Priest Valley in Monterey County, Calif., very young specimens of an annual Campanula were collected by the writer, May 12, 1893, resembling C’. angustiflora in 1 Zoe, Vol. I, p. 83. Bot.—Vot. I.] ZASTWOOD—STUDIES FROM THE HERBARIUM. 133 habit of growth, shape of corolla, and character of stamens and style. The leaves are much narrower, and the calyx divisions vary in length in the same flower and surpass the corolla. This I leave under C’.. angustzflora, as these small annual Campanule may be more common than is now supposed and the two species may vary a good deal among themselves. Their season is short, they grow in out-of-the- way places, and they are inconspicuous; so the chances of their discovery and collection are small. 2. Romneya trichocalyx, sp. nov. PLATE XI, Fics. 4a-4c. Perennial, glaucous and glabrous except for the scattered, spreading setze on the peduncles, rhachis, petioles, and lower margins of the leaves; stems many, suffruticose, laxly spreading from the base as they grow older, leafy, branched: leaves rather thin, with conspicuous venation, ovate-orbicular in outline, variable in size and divisions, pinnately 3-5 parted, the lower divi- sions entire, toothed or lobed, the terminal larger, cuneate, 3-5 cleft; leaves on the peduncle closely surrounding the bud, much smaller, the divisions linear, narrower, more numerous, and more setose-ciliate; petioles flat, 5 mm. to 2 cm, long: calyx of three imbricated sepals covered with upwardly appressed, scabrous setee, except near the margins and on the underlapping parts; corolla white, texture crape-like, 8-15 cm. in diameter, variable in the shape, size, and number of the petals; stamens numerous, with linear-oblong anthers and slender filaments, the lower half brownish purple, the upper yellow; styles ro-11, viscid, incurved; ovary ovoid, densely setose; capsule oblong-ovate, the walls breaking irregularly from the stout, straight ribs of the framework; seeds not seen. This has long been included under Romneya Coulter?, from which it is most markedly distinguished by the setose calyx. There is no doubt of the plant with smooth calyx being true 7. Cow/ter7, since the description and figure of the type confirm it.! Miss Kate E. Cole of Oakland first drew my attention to the fact that there were two kinds of Homneya in cultiva- tion, describing the marked differences between them; but it was not until last fall when I myself saw the two kinds growing side by side in Golden Gate Park, San Francisco, that I began to look up the matter. 1 Lond. Journ. Bot., Vol. IV, p. 75, Tab. II. 134 CALIFORNIA ACADEMY OF SCIENCES. [Proc. 3D SER. So tar as I have been able to discover, there are but two plants of /t. trichocalyx in the Park. Both grow back of the statue of Halleck where two paths meet. . Coulter grows there too; but the best specimen is to be found near the Haight St. entrance. It is also very abundant in other parts of the Park, being more desirable as a cul- tivated plant owing to its greater size and compact habit. Taking the plants in the Park for comparison, supple- mented by the specimens in the Herbarium of the Cali- fornia Academy of Sciences, the following are the points of differences :— Flabit.—R. Coulteri forms large, erect, close clumps, with many strong branches from along the stems. /. ¢richocalyx does not form close clumps because the stems as they grow tall have a tendency to lean over as if too weak to stand; they are more leafy, less branched, and neither so tall nor so stout. Leaves.—The leaves of R. Coulteri are thicker in texture, with fewer and larger divisions, becoming simple on the peduncle but never growing close under the flower, thus leaving the upper part of the peduncle naked. The differences in the leaves, however, are not always to be depended on as the leaves of 2. trichocalyx are so variable, often closely approaching those of R. Coultert, The upper leaves of R. trichocalyx however always become more dissected on the peduncle and grow close under the flower. Inflorescence.—The peduncles of R. Coulter are stouter and more spread- ing than are those of 2. drichocalyx. Calyx.—. Coulteri has a smooth calyx; that of 2. trichocalya is setose. Corolla.—In the specimens in the park, A. Cowl/eri has larger flowers, with the texture less crape-like than in 2. frichocalyx. We have specimens of the latter in the Herbarium with corollas fully as large as any of 2. Coultert. Fruit.—When the walls of the capsule break away leaving the skeleton of the pod, the ribs of the pod of 2. Coulteri are more slender, becoming atten- uated towards the apex and convolute; those of A. ¢richocalyx are stouter, uniform, and do not twist around. The figures show the differences in the buds and the leaves of the peduncle. They were drawn from fresh specimens from Golden Gate Park, smaller than ordi- narily because they were the last of the season. Both spe- cies bore fruit, but as it did not ripen, I was unable to com- pare the seeds. The pods from which the comparison was made came from herbarium specimens. Bot.—Vot. I.] EASTWOOD—STUDIES FROM THE HERBARIUM. 135 The following specimens of each species are in the Her- barium of the California Academy of Sciences :— Romneya Coulteri, Anaheim, July, 1885, M. K. Curran, (with fruit and buds). Baja California, W. G. Wright, (in flower). Golden Gate Park, San Francisco, (cultivated). Santiago Creek, near Orange, Orange Co., Calif., Miss Agnes Bowman, June, 1898, (with flowers and buds). Romneya trichocalyx, Aliso, Baja Calif., T. S. Brandegee, May 30, 1893, (in flower and dry fruit). Sausal, Baja Calif, T. S. Brandegee, June 4, 1893, (in flower). Cafion de Gato, Baja Calif., T. S. Brandegee, June 5, 1893, (stem only). Near Temecula, Riverside Co., Calif., No. 393, S. B. Parish, Oct., 1882, (in flower). Matilija Cafion, Ventura Co., Calif., F. W. Hubby, May 18, 1895, (in flower). Santa Maria R., Santa Barbara Co., Calif., Mrs. Ida E. Blochman, (a bud only). Golden Gate Park, San Francisco, (cultivated). 3. Sedum Congdoni, sp. nov. PLaTE XI, Fics. 5a-5d. Stem 1-6 cm. high, simple or branched from near the base, with slender, erect, tortuous branches: leaves alternate, 2-4 mm. long, 1-2 mm. wide: very fleshy, ovate, obtuse, sessile, the place of insertion above the base; flowers yellow tinged with red, sessile in sparingly branched, few-flowered cymes terminating the branches; calyx with five short, broadly triangular divisions, acute and red-tipped; petals five, ovate-lanceolate, less than 2 mm. long, red at the apex; stamens ten, with thread-like filaments shorter than the petals; anthers kidney-shaped; ovaries five, tuberculate near the apex, 1-ovuled; styles curved outwards; fruit unknown. This might be mistaken for Sedum pumilum, since both are small and have one-seeded follicles. The latter has much larger yellow flowers, linear-lanceolate petals, erect styles, and glabrous ovaries. The fringe of hairs at the suture of the follicle is much longer and finer in SS’. Aemezlem than in S. Congdonz. The former is farinose when young, becoming glabrous with age. The figures are designed to show the differences in regard to the petals and pistils of the two species. 136 CALIFORNIA ACADEMY OF SCIENCES. [Proc. 3D SER. This was discovered by Mr. J. W. Congdon, at Grant’s Springs, Mariposa County, Calif., and collected April 9, 1898. It adds another to his namesakes in Mariposa County, appropriately associating his name with the flora which he has done so much to make known. 4. Cercocarpus Traskie, sp. nov. PLATE XI, Fics. 7a-7e. Tree, 10 to 25 feet high; trunk 2 to 1o inches in diameter, 6 to 8 feet to the lowest branches; bark rough, grayish brown externally, reddish on the inside; upper branches covered with a thin, downy tomentum: leaves orbic- ular to oval, 2-6 cm. long, I-5 cm wide, with obtuse or acute apex, subcor- date, truncate, or cuneate base; margin revolute from deeply dentate to entire; upper surface dark green, glossy, glabrous, except the downy young leaves; lower surface densely white-tomentose, veins large and conspicuous on both sides; petiole stout, about 5 mm. long. Inflorescence androgynous, the polygamous flowers numerous in axillary umbels; calyx white-tomentose, with tube 1 cm. long and border 5-toothed, open campanulate, 5-8 mm. in diameter, glabrous within; stamens numerous, anthers tomentose, with two linear-oblong cells united only at the insertion of the slender filament; perfect flowers with stigmas curved like shepherd’s crooks, style exserted; akenes 1 cm. long, linear-oblong, covered with upwardly appressed silky hairs, tipped by the circinate, persistent style, about 5 cm. long, clothed with long, fine, silky hairs spreading horizontally. This, the most beautiful of the Pacific Coast Cercocarpz, was discovered by Mrs. Blanche Trask at the southern part of the island in a volcanic region known as ‘‘ Salte Verde.”’ It is a wild place, too rough for men on horses, with no trails but those made by the goats. Even in winter the heat is great. She writes as follows concerning the place and the trees: ‘‘There are about forty or fifty trees in an arroya so small that there is but room to squeeze through, a southern exposure where Ruin and Earthquake have passed and in whose footprints but few plants have dared to rise.”’ The sea dashes at the base of this arroya, the walls of which rise to a height of from 100 to 500 feet. The trees are all isolated, not at all forming thickets. That any one should have found a new tree on an island that so many botanists have visited is surprising; but it is due to the great enthusiasm, the wonderful power of Bot.—VOL. I.] ZASTWOOD—STUDIES FROM THE HERBARIUM. 137 exploration, and the intense love for Santa Catalina Island and its flowers which Mrs. Trask possesses. It is with pleasure that I give her name to this tree. Photographs of several of these trees taken recently by Mrs. Trask show them to have widely spreading branches and graceful habit, and to be well worthy of cultivation. The branches are abundantly adorned with rosettes of white tomentose flowers at all the leaf axils, the contrast of which against the dark, glossy green of the upper leaf sur- face is striking and beautiful. The same contrast occurs between the upper and lower surfaces of the leaves, the beauty of which is enhanced by the strong, even venation and revolute margins of the leaves. This tree is unlike the other Pacific coast species and perhaps approaches C. fothergillordes H. B. K., the Mexi- can species, more nearly than any other. It seems to bea type of Cercocarpus isolated and distinct. 5. Calochortus Purdyi, sp. nov. PLATE XI, Fics. 8a-8/. Glabrous and glaucous; stem 2-3 dm. high, rather stout, erect, branching, two to many-flowered, not bulbiferous at base: radical leaf solitary, sheathing the stem, linear-lanceolate, acuminate, 2 dm. long, 1 cm. wide, the upper surface bright green, the lower glaucous and ribbed with the filiform nerves; bracts foliaceous, lanceolate-acuminate, amplexicaul, upper ones opposite; pedicels equalling or slightly surpassing the bracts, erect in flower, recurved in fruit: flowers broadly open-campanulate; sepals from elliptical to narrowly ovate, abruptly acuminate, tinged with purple on the outer surface, purple- veined on the inner, two-thirds as long as the petals; petals broadly obovate- cuneate, acute or rounded at apex, creamy white or tinged with purple, bearded all over the inner surface with long hairs which are white on the upper half of the petals, purple on the lower, somewhat arched by the nar- row, transverse, semicircular, conspicuous gland, the shallow pit of which is covered by a densely hairy narrow scale; anthers lanceolate, abruptly acumi- nate, cream color or purplish, shorter than the filaments, which broaden to the base; capsule 3 cm. long, 2 cm. wide, broadly elliptical, with the three thin wing-like valves transversely veined. This belongs to the § Hucalychortus according to Wat- son’s arrangement in the Botany of California. In habit it resembles C. a/bus, but in general is more like a giant C. 138 CALIFORNIA ACADEMY OF SCIENCES. [PRroc. 3D SER. Maweanus. Its nearest relative is, however, C. Zolmzez, which belongs to the same region and from which it can readily be distinguished by the absence of the scale cover- ing the gland of the latter. Most of the C. Zolmzez in herbaria is probably this species. It was compared with the original specimen of C’. Zo/m7ez at Kew by J. G. Baker who says: ‘* The Willamette plant differs a good deal from the original C. Tolmiec. C. Tolmzez has pale lilac petals bearded all over the face, no spot, no scale, no obtuse anthers.”’ There is no true C’. Zo/mzez in the Herbarium of the Academy. We have specimens collected by Thos. Howell, from Grant’s Pass, Oregon, from Prairies, West- ern Oregon, April 1881, and from Hillsboro, Oregon, May 1881, all marked C. Zolmzez, but each with a scale covering the gland and with flowers creamy rather than blue. It would seem as if the anthers were variable, since all of C. Purdyz that I have examined have acuminate anthers; but evidently the specimens sent to Kew had obtuse anthers. C. Purdyt grows in the Willamette Valley, in the toot- hills on dry gravelly soil. It is never found in shaded woods. It is named in honor of Carl Purdy who knows Calo- chort? more intimately than anyone, and whose work on the genus in the garden, the field, and the study has accom- plished so much towards determining the true specific limits in this difficult and variable genus.! 1 Since the above description was written and the drawings made, a beautiful figure of this species has appeared in the ‘‘Gardner’s Chronicle,” Vol. I, 1898, p. 305, fig. 147. The floweris much larger than that of any specimen seen by me. It is not unusual, however, that plants have larger flowers under the more favorable surroundings that cultivation produces. It isa well known fact that Pacific coast species vary consider- ably in the size and vigor of individuals according to the amount of moisture and fertility of the soil. I40 CALIFORNIA ACADEMY OF SCIENCES. [Proc. 3D SER. EXPLANATION OF PLATE VIII. Fig. 1. Peucedanum insulare, sp. nov. a. and 6 Parts of the same leaf, actual size. External view of carpels, magnified as shown in figure. é. d. Cross-section of a carpel, magnified as shown in figure. Fig. 2. Hosackia venusta, sp. nov. a, Calyx. 6, Standard. c, Keel. d, Wing. Figures mag- nified 5 times. Fig. 3. Hosackia argophylla GRAY. a, Calyx. 6, Standard. c, Keel. d@, Wing. Figures mag- nified 5 times. Fig. 4. Abronia alba, sp. nov. a, Flower. 6, Immature fruit. Figures magnified 5 times. Lycium verrucosnm, sp. nov. Tip of branch, actual size. a. 6. Flower magnified 5 times. Interior of corolla showing the position of the stamens. cq Fig. 5. d. Interior of the calyx showing the pistil. Astragalus Traskie, sp. nov. Fruiting branch, actual size. Flower, twice the actual size. c. Cross-section of the pod. d. Mature pod. Fig. 7. Amsinckia St. Nicolai, sp. nov. a, Calyx. 6, Corolla spread open. c, Outline of corolla. d and ¢, two views of the akene. Figures about 5 times the Fig. 6. a. b. actual size. Amsinckia maritima, sp. nov. a, Calyx. 6, Corolla spread open, Enlarged as in fig. 7. Fig. 8. c, Outline of corolla. [Easrwoon | Pure VIL Bape CaLAcanScr." oer. Bar. VorI. cin eo om i ate ip vay a he 53) ; ‘ i ; ; ey al A 4 i : 1 | — ‘ ia Se toa 7 i Al in f Ng Vion) i ii i} ‘ i} ) Y Unis i \ i i} th i i i \ . i t i} i fi ‘ H i R ; , OSC olin J et ens it SoG ee crits | OHIO, ant 142 CALIFORNIA ACADEMY OF SCIENCES. EXPLANATION OF PLATE IX. All parts of the flowers are 5 times actual size Synthyris alpina GRAY. a. Flower. 6. Calyx. c. Lower part of corolla. ad. Upper part of corolla. Fig. 2. Syuthyris Ritteriana, sp. nov. a. Flower. b. Calyx. cand d. Lower part of two different corollas. e. Upper part of corolla. Syuthyris reflexa, sp. nov. a. Flower. 6. One of the equal divisions of the calyx. c. Lower part of corolla. ad. Upper part of corolla. igs t. Fig. 3. [PRroc. 3D SER. Are IX J py [Eastwoon | Proc. CatAcanScr.3" Ser. Bor 144 CALIFORNIA ACADEMY OF SCIENCES. [PRoc. 3D SER. EXPLANATION OF PLATE X. All figures are 5 times actual size. Fig. 1. Evtodictyon crassifolium BENTHAM. a. Part of corolla showing the stamens. 6. Corolla. c. Part of calyx showing the pistil. Fig. 2. Eriodictyon Traskice, sp. nov. a. Interior of corolla showing the stamens. 6. Corolla. c. Calyx spread open showing the pistil. Fig. 3. Lriodictyon niveum, sp. nov. a. Interior of corolla showing the stamens, 6. Corolla of a young flower. c. Corolla with lobes fully expanded. d. Calyx spread open showing the pistil. LERIOQDICTYON CRASSIFU evel AIO fastwoon, DEL Proc. LatAcan Ser. 3° Ser. Bor. Vocl. ; . [ EASTWwoou | Plate X LIUM FEVTHAM, ec ERIODICTYON TRASKIA. A4S7woan. ICTYON NIVEUM 4457woa7 LOW BRITTON & REY, SF 146 Fig. ) CALIFORNIA ACADEMY OF SCIENCES [Proc. 3D SER. EXPLANATION OF PLATE XI. Campanula exigua RATTAN. a, Flower fully expanded. 6, Fruit. c, Pistil and stamen. Figures 5 times actual size. Campanula angustifiora, sp. nov. a, Flower expanded. 6, Corolla spread open. c, Fruit. d, Pistil and stamen. Figures 5 times actual size. Romneya Coultert HARVEY. a, Bud. 6, Immature pod. Figures actual size. Romuneya trichocalyx, sp. nov. a, Bud. 6, Immature pod. c, Leaf. Figures actual size. Sedum Congdon, sp. nov. a, Plant actual size. 6, Flower. c, Follicle. @, Petal and stamen. The flower and its parts enlarged to times. Sedum pumilum BENTHAM. a, Petaland stamen. 4, Follicle. Enlarged to times. Cercocarpus Traski@, sp. nov. a. Tip of a twig, actual size. 6. One of the larger leaves, showing the lower surface. c. Anther enlarged. da. Fruit. e. Flower. Calochortus Purdyi, sp. nov. a. Plant, actual size. b. Petal. c. Scale on petal enlarged. dad. Sepal. e. Stamen enlarged. J. Ripe pod. ‘| ~ PLATE ] J "wuOoD EAST [ L PHYCOLOGICAL MEMOIRS. BY DE ALTON SAUNDERS. CONTENTS. Pirates XII-XXXII. PAGE [ep SOME) PAGIFIG“COAST (ECTOCARPAGEDE:. cfaic rh E rr = rs re SI ri Ff ~ = 5 ts a § 4 — B KJ po ij BY a) rst on q ak Re S fq S 4 =I a S aa — =) xq oi = S a) CHITONICOLU THEARPDOS Ey EF E Pips l= 4 FiesS—4. ELToGARPUS CONPERVOMES Fy Sto ae as a ve er ee sc ee ANG ST] CR CAS a Proc [arAcan.Scr.3* Ser. Bor Van] ‘Val ER. Bor) q Proc CanAcan Scr." ¢ EETOCARPUS CYLINDRICUS, SP NOV. 3 al = i= o 4 jaw fo Cha Ee a Proc CaLAcan Scr. ae Gs pis S eS ——— ee eee ELH Det. . ETOCARPUS 1 : : 7 Oy, (or 7 oo ~ i . i) 7 7 is oa i ig “s 7 7 ~ i F: 7 i" = 5 — 5 Ls = = § & $e} Ss = ae : pb ea aay eet —— 5 Gi es yy ae ei fa Oe md pa (cai = ae ea +P mi pea BS < Bb ia ia SS me =e ea toes | io < a =| iS} [ap ay Ie a) te 4 i a, | al 2 as Se fa FA jaa fap) wm RTOCAR ECTOLARP Proc CarAcanScr.3* Ser. Bor Vol. Froc Ca AcanScr.a* Ser Bor Vowl. [SAUNDERS ERD (i Was ‘ E26 SouTMessvweew ai) fe) fs ees Oe ELA Ten EETOCARFUS MUGRONATUS, SP NOV Paoe.CarAcan Scr." Ser. Rot Vol. ——— ECTOLAREUS CORTICULATUS, By : 5 ] Paoe.CavAcan.scr.3° SER. Bor Vol. [SAUNDERS | PLATE XX] t) o & aoe SS qi eee tS ALITY BRITTON & REY, SF. Fires. 1- Z. EGTOCARPUS MITEHEILA AARV jf Fiss.4-4 ECTocaARPUS PENICILLATUS £.AG a1 ; =A ie - a“ ‘ 2 Proc.CanAcan Scr 3° ger. Bor.Vor.]. [SAUNDERS | PLATE XXII. ELFL Ten, ECTOCARPIS oo ILIGUDLGSUS PARVIS, VAR. NOV. ~- ine ER. HoT Voc. 5 Proc. CatAcan Scr 37 sa0g5Se5 oe 23' aoe, ee 8 aa "aa ReEDESS ELSES See Semereens Ze Ee Ene PED —— Re. § vif LIT» =p Ess eg ge SESS EE ee 2 é Ss yo ES NOV. RIABING FORM, TA VE COUNFERVOIES EETOCAREDS t a) a SE LTY BRITON & REY, we Bar] } Ly A ( / Proc. CacAcan 5cr.3°'Ser. Bor Vol. woo =a) =} im ow = ae a reals 2 B&B ) : Up wu ida ari we V2 a Sn Mi GOA a, fy SEN he ha iS z y By > BN | WES Sim ets P= -ee cd jel Ge =P Sea de oT Isis te Nun eo dl (ai Weal i tA =) (aa Be Bey ELA fen. : " oy Re a : i” : ; a = ni v , ' 7 : a - + ; | - & ' : : P } ; a / ’ 7 VV {LN ea | PLATE 7 a |SAUNDER Proc CavAcanScr.3" Ser. Bor Vol. IN REY, SF LEY BRIE ELA TED, El ey Eigsl=A-co8-. FYLATELEA LITTORALIS, 11 NBA, FORM. NOV. LLA LITTORALIS, VAR. =] GB 4 a pL, SER. Bor. Vol, | LAL { PRO. say aes rats Tapeh wm oon els me oO fe eet mg 4 3 Ee Ss 4 ree a ee =a auc fy ne 63 ~ tn | tic juplei—! oy ok (eal Ww ra pe AUNDERS | FLATE XXXII eC je ul Proc. CatAcan5cr.3* Ser. Kor Voz!. ITTON & REY, SF: LTH ER VW V N op or REULATA , oa) OLPOMENIA TUBI =| Ly ieuieiey Ib eh Fies.4- 6. COLFOMENIA SINUOSA EXPANSA, FORM. NOV. ERA & 7}; Ly CT TEL | figs. /7- &. COLPOMENIA sinuosa /A SOME OBSERVATIONS ON THE DEVELOP- MENT OF THE KARYOKINETIC SPINDLE IN THE POLLEN-MOTHER-CELLS OF COBAZA SCANDENS CAV.} BY ANSTRUTHER A. LAWSON. PLATES XXXIII-XXXVI. From the recent investigations of Strasburger (1897), Farmer (1895, 6), Swingle (1897), Harper (1897), Mottier (1898), and others, we learn that centrosomes are present and take an active part in the formation of the karyokinetic spindle in certain of the lower plants. The most complete series of stages yet published showing the behavior of centrosomes in plants is found in the papers of Swingle (1897) and Harper (1897). Swingle investigated the apical cells of Styfocaulon, Harper the developing asco- spores in the ascus of Hyryszphe. In both of these cases, when the spindle is about to be developed, there is a body present which is surrounded by a system of kinoplasmic radiations. This body divides into two and the daughter- centrosomes thus formed migrate to opposite sides of the nucleus and form the spindle in much the same manner as in animals. Strasburger (1897, @) has described and fig- ured well defined centrosomes in the odgonium of Fucus, and Mottier (1898) has also found them in the tetraspo- rangia of Dictyota. In all of these cases it will be observed that the centro- some is described and figured as taking an active and very essential part in the development of the spindle. We have thus a very striking parallel between the formation of the spindle in animals and in the lower plants; but when we attempt to extend this parallel to the higher plants we meet 1Contributions from the Botanical Laboratories of the University of California, No. 3. Presented for the degree of Master of Science. Prepared under the direction of Mr. W. J. V. Osterhout. [169| November 15, 1898. 170 CALIFORNIA ACADEMY OF SCIENCES. [PRroc. 3D SER. with difficulties. In spite of the fact that Guignard (1891) and other writers have described centrosomes in the higher plants, recent investigations have rendered their existence very improbable. These investigations agree in showing that the spindle in its earlier stages possesses several poles, and later, by the fusion of these, becomes bipolar. In such a process of spindle-formation it is difficult to understand how a centrosome can have any part, nor has any observer, Guignard excepted, claimed to discover centrosomes in connection with it. Guignard (1898), however, in a recent article, while admitting that the spindle goes through a multipolar stage in the course of its development, maintains that this does not prove the non-existence of the centrosome. In the pollen-mother-cells of Wymphea and Wuphar he describes centrosomes situated at the apices of the cones in the multipolar figures, and also at the poles of the mature spindle. But unfortunately he does not describe or figure a series of stages that would illustrate the behavior of these bodies. Moreover, Strasburger (1897, 6) has already dis- cussed this idea and concludes that it is highly improbable that a centrosome plays any part in this process. Although multipolar spindles were previously described by Belajeff (1894), Farmer (1893), and Strasburger (1896), their significance was not fully understood until Osterhout’s (1897) observations on Lguzsetum were made known. In this paper, which furnishes us with the most complete series of stages of spindle-formation yet worked out, the process is described as follows :— The first thing to be observed in the formation of the spindle in Hguzsetum is a felted zone of kinoplasmic fibres surrounding the nucleus. These fibres grow out from the nuclear wall and take on a radial arrangement. By the coming together of their free ends these threads form a series of cones. The nuclear wall now breaks down and the fibres composing the cones grow in and become attached to the linin and chromosomes. The apices of the cones now approach each other and arrange themselves in Bor.—VOL. I.] LAWSON—COBA‘A SCANDENS. LT two groups and then fuse to form a bipolar spindle. No bodies or granules that could be identified as centrosomes were observed in any stage of the process. It seems very improbable that the occurrence of multi- polar figures is an abnormal phenomenon. They have been found to occur in Le/zum by Farmer (1893), in Larix by Belajeff (1894), and Strasburger (1896). They have been found in Lilium, Fretellaria, Helleborus, Podophyllum, and Pinus by Mottier (1897, @ and 4), in Hemerocallis by Juel (1897), in Chara by Debski (1897), and in Zama by Webber (1897.) The writer has observed them in Aes- perle, Hedera, Disporum, Smilacina, Gladiolus, Irts, Cobea, and other genera. From this we can only conclude that multipolar spindles are of very general occurrence, and this throws considerable doubt upon the existence of cen- trosomes in the higher plants. It must be observed, however, that in all cases where multipolar figures have been described, very little has been observed upon the very earliest stages in the development of the kinoplasmic fibres. It is with the hope of throwing some light on these stages that the following observations upon the pollen-mother-cells of Cob@a scandens have been made. About two years ago Mr. Osterhout suggested that the writer undertake some investigations upon spindle-forma- tion in anthers of different orders of flowering plants. In undertaking such a task it is very desirable to obtain material with large anthers and large pollen-mother-cells as well as large nuclei. The anthers of Cob@a are very good in all of these respects. The anthers were gathered and immediately fixed in the field. The following fixing fluids were used: Alcohol, 95 per cent. Chromic acid, 1 per cent. Flemming’s mixture, chromic-osmic-acetic, strong solution. Wilson's sublimate-acetic. Boveri's picro-acetic. Corrosive sublimate, saturated solution in 95 per cent. alcohol. I72 CALIFORNIA ACADEMY OF SCIENCES. [PRoc. 3D SER. The best results were obtained by using Flemming’s strong solution diluted with one volume of water. The material was washed in running water from six to eight hours. It was then carried through different grades of alcohol by means of an apparatus consisting of a tumbler with a cover and a glass funnel. The funnel was of such a size as to be supported in the mouth of the tumbler. A piece of parchment paper was folded and placed in the fun- nel in the same manner as for filtering. 10 per cent. solution of alcohol was placed in the tumbler and the anthers were placed in water in the funnel. The mouth of the tumbler was covered and the material was thus allowed to remain for an hour or two. The alcohol in the tumbler was changed at intervals to 25 per cent., 50 per cent. and 95 per cent. solutions. By this means the effect of the rapid change from a weak to a strong solu- tion of alcohol was obviated. The anthers were then thoroughly dehydrated in absolute alcohol. They were then placed in a mixture of bergamot oil and alcohol and then in pure bergamot oil. From the bergamot oil they were transferred to a mixture of bergamot oil and paraffin, and from this to pure paraffin, where they remained at a temperature of 55°C. for twenty-four hours. Microtome sections of 3% in thickness were used. Many stains were tried; especially iron-hamatoxylin and Bordeau red, ruthenium red and thionin, etc., but the best results were obtained from Flemming’s triple stain, safranin, gentian violet, and orange G. In the resting condition of the pollen-mother-cell the nucleus is quite large. It contains one or two large nucleoli which stain very readily with safranin and some- times appear to be vacuolated. The chromatin, which is in the skein stage, stains blue with the gentian violet. As soon as the chromatin breaks up and forms the chromo- somes it stains red with safranin. The chromosomes appear as small oval bodies which are invariably situated in contact with the nuclear wall. The largest number that was observed in polar view was twelve. On the chromosomes Bot.—VOL. I.] LAWSON—COBA\A SCANDENS. L732 the writer has made very few observations. They will therefore not be discussed farther. The linin now appears in the form of a lumpy or granular thread; it stains blue and is invariably connected with the chromosomes. The cytoplasm now appears in the form of a clear retic- ulum, as shown in fig. 1. The meshes of this network, which can be traced from the nuclear wall to the cell-wall, appear to be smaller and radially elongated in the imme- diate neighborhood of the nucleus; but as one follows them outwards they increase in size and are comparatively large towards the cell-wall. Scattered irregularly through the cytoplasm are numerous small spherical bodies. These bodies have the appearance of oil-globules in the living cell, but after the cell has been killed in Flemming’s fixing fluid they appear quite black. What the chemical nature or function of these bodies is, the writer is unable to state. The cytoplasm does not maintain its clear, uniform appearance for a very long period, but soon undergoes a remarkable differentiation. This differentiation is shown in figs. 2-5. It is brought about by the gradual accumula- tion of a granular substance which: forms a complete zone about the nucleus. The minute structure of this substance is dificult to make out, but it appears to consist of granules which vary in size. While some are very minute, others are comparatively large. They appear to be arranged in such a manner as to give the impression of a foam-structure. This granular zone is so constant in Codea and in several other genera observed by the writer, that for con- venience it will be called perikaryoplasm. While it is accumulating, the black bodies, which were previously scattered irregularly through the cytoplasm, take up a definite position. They arrange themselves along the outer margin of the perikaryoplasm, and in section present the appearance of a complete ring, as shown in fig. 5. It is at this stage that the contrast between the outer cytoplasmic reticulum and the perikaryoplasm is greatest. The threads of the outer reticulum now appear to have no 174 CALIFORNIA ACADEMY OF SCIENCES. [Proc. 3D SER. connection with the nucleus. They can no longer be traced through the perikaryoplasm. Where the outer reticulum abuts on the perikaryoplasm the meshes of the former are so compressed as to give the impression of a loose membrane. These two constituents of the cytoplasm not only differ from one another as regards their structure but their staining properties stand in great contrast. | While the outer reticu- lum stains a light or gray-blue with the gentian violet, the perikaryoplasm stains a decided orange with the orange G. The entire cytoplasm thus presents the appearance of two distinct and sharply differentiated zones, differing from one another as regards structure and staining properties. These zones are seen in fig. 5. In case it should be thought that this is due to artifact, it might be well to state here that a large number of living cells were examined, and this striking differentiation of the cytoplasm and the ring of spherical bodies, together with the threads of the outer reticulum, could be readily seen. This, in the writer’s opinion, is sufficient evidence to prove that the phenomenon is a normal one. There seems little doubt that it is the first step towards the formation of the spindle. In fig. 6 we see the first indication of the breaking down of the nuclear wall. It evidently does not break down all at once. It commences to do so at one or more points where it becomes lost in a network which is now developed out of the linin and perikaryoplasm. Figs. 6-8 show stages in the breaking down of the nuclear membrane. It will be seen from these figures that the perikaryoplasm, where it is in the immediate vicinity of the breaks in the nuclear wall, is now undergoing a change. The linin has lost its lumpy appearance and now appears as fine, delicate threads (fig. 6) which are in direct communication with the perikaryoplasm. The latter is losing its granular nature in these places and is taking on the form of a net- work. In fig. 8 we see this network taking on a very definite form. It no longer stains orange, like the Bot.—Vot. 1.] LAWSON—COB4A SCANDENS. 175 perikaryoplasm from which it develops, but stains deeper and deeper violet like the linin. At the points where the threads intersect granules are seen. In fig. 8 the nuclear wall is no longer continuous but is in the form of fragments. To what extent the linin enters into the formation of the network, the writer is at present unable to say, but it cer- tainly appears to begin its formation. The amount of linin present when the nuclear wall breaks down is not sufficient to form the large central network that we see in fig. 9. This network increases in size and apparently grows at the expense of the perikaryoplasm. In fig. g the nuclear wall has entirely disappeared and the area once occupied by the nucleus is now filled by a perfect network which supports the chromosomes within its meshes. These latter bodies appear to be more frequently present in those places where the network is growing outward in the form of projections. The contents of the cell is now beautifully differentiated into three parts: the outer cytoplasmic reticulum which stains light blue, the perikaryoplasm which stains orange, and the central network which stains violet. Between these latter two there is a transition place where the threads of the central network become granular and stain less deeply violet, until finally they become orange and pass over into the granular substance of the perikaryoplasm, where they can no longer be followed. In fig. 9 we see that the central network has at three places formed definite projections. How many of these are generally formed it is difficult to make out from sec- tions, but the number is nearly always more than two. In fig. 10 we see that the meshes of the network have become elongated in the directions in which the projections extended. In fig. 12 we see that the meshes have a drawn- out appearance, and we have the outline of a multipolar spindle. Figure 11 is probably a later stage than fig. 12, inasmuch as the meshes are drawn out to such an extent as to present the appearance of distinct fibres, and we now have the first spindle-fibres formed. In this figure it will 176 CALIFORNIA ACADEMY OF SCIENCES. [PRoc. 3D SER. also be observed that the nucleolus still persists, which shows that it has not been used to form the central network. This is the latest stage in which the nucleolus was seen, but what eventually becomes of it was not observed. In the next figure we have a characteristic multipolar spindle. Here the fibres are straight and form cones which terminate in sharp points. We see extending from the apices of the cones fibres which have free ends. These fibres eventually become the mantle-fibres which are so characteristic of the mature spindle. The multipolar figure becomes bipolar by the fusion of the cones in much the same manner as it does in Aguzsetum, as shown in figs. 14 and 15, but during the process the mantle-fibres have developed to a considerable length. From the time of the breaking down of the nuclear wall the chromosomes have been in connection with the network from which are developed the spindle-fibres; so that this process differs materially from those cases where the spindle-fibres have been described as growing in and becoming attached to the chromosomes. As shown in figs. 15 and 16, the mature spindle termi- nates in sharp points; there are three kinds of fibres present: The contractile fibres, which are in connection with the chromosomes, are compound in nature, being made up of a number of fine fibres. The continuous fibres extend from pole to pole uninterrupted. The mantle-fibres are remark- able for their great length and early development; they consist of very delicate straight threads; those from one pole cross those from the opposite pole in the manner shown in figs. 15 and 16. It should be observed that although the mantle-fibres are of great length they never extend beyond the perikaryoplasm. After the mantle-fibres have reached their maximum development the contractile fibres commence to draw the chromosomes to the poles. This takes place in the usual way. It will be noticed, however, as shown in fig. 17, that during this process the mantle-fibres appear more divergent from their respective poles. When the chromosomes have Bot.—VOL. I.] LAWSON—COBAZA SCANDENS. ui 7 reached the poles the mantle-fibres immediately lose their straight appearance and hang loosely alongside of the remaining continuous fibres, as shown in fig. 18. They soon after disappear. Although the continuous fibres now curve out towards the cell-wall as shown in fig. 19, there is no cell-plate formed until after the second division. When the daughter-nuclei have surrounded themselves with a membrane the continuous fibres still persist, being attached to the daughter-nuclei as shown in fig. 19. In this figure we also see that the perikaryoplasm appears to be more plentiful at the ends of the cell where the nuclei lie. It surrounds the nuclei except where the continuous fibres are in connection with the nuclear membrane. ‘The peri- karyoplasm gradually surrounds the nuclei and apparently cuts off the continuous fibres. As shown in fig. 20, they can no longer be traced to the nuclear wall, and the perikaryoplasm has formed a complete zone about each nucleus. The development of the spindle of the second division is identical with that of the first division, but on account of its small size, the stages in the process are much more difficult to work out. In fig. 21 we see the last traces ot the continuous fibres, and each of the nuclei is surrounded by a granular zone. It will also be noticed that the black bodies have formed rings at the outer margin of the zones of perikaryoplasm, just as they did in the case of the first division. These zones are as well defined and differentiated from the rest of the cytoplasm in structure and staining properties as in the first division. In fig. 22 we see the two spindles of the second division lying at right angles to one another. In one the spindle is represented only by the cross-sections of its fibres. The other shows the entire spindle with the characteristic mantle- fibres, and the chromosomes on their way to the poles. This figure also illustrates the fact that the perikaryoplasm accommodates itself to the shape of the spindle. In one I 78 CALIFORNIA ACADEMY OF SCIENCES. [Proc. 3D SER. view it appears round and in the other oblong in outline, corresponding to the position and shape of the spindle. In the next figure we see that the spindles are more nearly in the same plane. The mantle-fibres of the two spindles have reached over and have apparently united with one another. In fig. 24 we see three daughter-nuclei in the same plane. By means of the continuous fibres and mantle-fibres they are connected with one another. These fibres now occupy almost the entire cell-cavity. Cell- plates are now formed in the usual way. Swellings appear on the connecting fibres; these increase in size, and finally result in forming cell-walls which separate the daughter- cells from one another. During this stage there appears to be very little of the perikaryoplasm left and what is present is scattered irregularly through the cell. Whether the method of spindle-formation observed in Cobea is of general or exceptional occurrence must be left for future investigation to decide. From Belajeff’s (1894) description of Zarzx it may be supposed that a similar pro- cess takes place here, but the origin of the network and the manner in which the spindle-fibres subsequently arise from it were not sufficiently investigated. The develop- ment of spindle-fibres out of a network, as shown in figs. 9-13, recalls Wilson’s (1895) description of the formation of astral rays and spindle-fibres in Toxopneustes. This method of spindle-formation differs decidedly in its earlier stages from that observed by Osterhout (1897) in E-quisetum, although both agree in the formation of a num- ber of poles which subsequently fuse to form the bipolar spindle. We have in Codea a zone of granular substance, the perikaryoplasm, whose function it is to take part in the formation of a network from which the spindle-fibres are developed; no such zone, however, is present in Hguzse- tum. Similar zones have been figured by Belajeff (1894) (fig. 6) in Laréx and Mottier (1897, @) in Lekum, but they have not been described as taking any part in the formation of kinoplasmic fibres. Bot.—VOL. I.] LAWSON—COBAZA SCA NDENS, 179 In all other cases where multipolar spindles have been observed the earlier stages have not been sufficiently inves- tigated to warrant any statement as to whether they follow the Hguzsetum type or that here described for Cobea. From the series of stages here figured the writer can only conclude that centrosomes or directive spheres can take no part in the formation of the spindle, and they con- firm the idea that has already been expressed that in the vascular plants the method of spindle-formation is entirely different from that which prevails in the lower plants and animals. In conclusion I wish to acknowledge my indebtedness to Mr. Osterhout for many valuable suggestions in the prep- aration of this paper. SUMMARY. The observations made upon the formation of the spindle in Cobea may be briefly stated as follows: A granular substance gradually accumulates and forms a complete zone around the nucleus. This zone is designated perikaryoplasm. Upon the breaking down of the nuclear wall the linin of the nucleus and the perikaryoplasm form a network which occupies the central portion of the cell. This network grows out into several projections which become the cones in the multipolar figures. The spindle-fibres are formed by the elongation of the meshes of the network in the direction of the projections. The cones elongate and become sharply pointed. They fuse in two groups and form the bipolar spindle in the same manner as that observed by Osterhout in Aguzsetum. The mature spindle is characterized by the great length and crossing of the mantle-fibres. The spindle-formation of the second division is identical with that of the first division. No bodies that could be identified as centrosomes were found in any stage of the process. BOTANICAL LABORATORY, UNIVERSITY OF CALIFORNIA, April, 1898. 180 18g95a. 18958. 1891. 1898. 1897. 1897. 18974. 18976. 1898. 1897. 1896. 18974. 18978. 1897. 1895. 1897. CALIFORNIA ACADEMY OF SCIENCES. [Proc. 3D SER. BIBLIOGRAPHY. BELAJEFF, W. Zur Kenntniss der Karyokinese bei den Pflanzen. Flora, Bd. LXXIX, 1894, p. 430. Desski, B. Beobachtungen tiber Kerntheilung bei Chara fragilis. Jahrb, f. wiss Bot., Bd. XXX., 1897, p. 227. FARMER, J. B. On nuclear division in the pollen-mother-cell of Lilium Martagon. Ann. of Bot., Vol. VII, 1893, p. 392. Ueber Kerntheilung in Lilium-Antheren besonders in Bezug auf die Centrosomenfrage. Flora, Bd. LXXX, 1895, p. 56. On spore-formation and nuclear division in the Hep- atice. Ann. of Bot., Vol. 1X, 1895, p. 469. GUIGNARD, L. Nouvelles études sur la fécondation. Ann. des Sci. Nat. Bot., Serie 7, Tome XIV, 1891, p. 163. Centrosomes in Plants. ot, Gaz., Vol. XXV, 1898, p. 158. Harper, R. A. Kerntheilung und freie Zellbildung im Ascus. Jahrb, f. wiss. Bot., Bd. XXX, 1897, p. 249, JueLt, H. O. Die Kerntheilungen in den Pollenmutterzellen von Hemerocallis fulva und die bei denselben auftretenden Unregel- missigkeiten. /ahrb. /. wiss. Bot., Bd. XXX, 1897, p. 205- Mortier, D. M. Ueber das Verhalten der Kerne bei der Entwickelung des Embryosacks und die Vorginge bei der Befruchtung. /ahrb. /. wiss. Bot., Bd. SC 1897, p. 125. Beitrage zur Kenntniss der Kerntheilung in den Pollen- mutterzellen einiger Dicotylen und Monocotylen. /ahrb. f. wiss. Bot., Bd. XXX, 1897, p. 169. Das Centrosome bei Dictyota. Aerichte der deutschen Bot. Gesell., Bd. XVI, 1898, Heft. 5. OsTERHOUT, W. J. V. Ueber Entstehung der karyokinetischen Spindle bei Equisetum. /ahrb. f. wiss. Bot., Bd. XXX., 1897, p. 159. STRASBURGER, E. Karyokinetische Probleme. /ahrd. /. wiss. Bot., Bd. XXVIII, 1896, p. 151. Kerntheilung und Befruchtung bei Fucus. /ahrd. f. wiss. Bot., Bd. XXX, 1897, p. 351. Ueber Cytoplasmastructuren, Kern- und Zelltheilung. /ahré. J. wiss. Bot., Bd. XXX, 1897, p. 375. Wesser, H. J. Peculiar structures occurring in the pollen-tube of Zamia. Sot. Gaz., Vol. XXIII, 1897, p. 453. Witson, E. B. Archoplasm, Centrosome, and Chromatin in the Sea- Urchin Egg. /ourn. of Morph., Vol. XI, 1895, p- 443. SWINGLE, W. T. Zur Kenntniss der Kern- und Zelltheilung bei den Sphacelariaceen. /ahrb. /. wiss. Bot., Bd. XXX, 1897, p. 297. Fig. 1. Fig. 2. Fig. 3. Fig. 4. Fig. 5. Fig. 6. CALIFORNIA ACADEMY OF SCIENCES. [Proc. 3D SER. EXPLANATION OF PLATE XXXIII. Figures drawn with Abbe’s camera lucida. Zeiss, Apochromatic Immersion Obj. 12 mm. 1.30 Ap. Compensating Ocular No. 6. A young pollen-mother-cell. The cytoplasm is a clear network which stains a light or gray-biue. The meshes of the network appear to be smaller and radially elongated towards the nucleus. There are numerous spherical bodies scattered through the cytoplasm. They are stained black, probably by the fixing fluid. The nucleus is quite large and shows a large nucleolus and small oval chromosomes which lie immediately upon the nuclear wall. The linin, which stains deep blue, is of a lumpy or granular nature. This figure shows the commencement of a remarkable differentia- tion which takes place in the cytoplasm which immediately surrounds the nucleus. A zone of granular substance, the perikaryoplasm, forms around the nucleus, which differs decid- edly both in structure and staining qualities from the sur- rounding network. While the outer cytoplasmic reticulum stains a light or gray-blue, the granular substance stains orange. This figure shows a later stage with an increase in the quantity of perikaryoplasm about the nucleus, otherwise it does not differ essentially from fig. 2. Shows a still greater increase in the quantity of perikaryoplasm The about the nucleus. The black bodies are now beginning to form a ring at the outer margin of the granular zone. perikaryoplasm has now reached its maximum and stands out in the greatest contrast from the outer cytoplasm. The meshes of the outer reticulum which abut on the perikaryoplasm are small and apparently compressed, giving the impression of a loose membrane which separates the two zones from one another. It is impossible to trace any of the outer meshes into the granular zone. They appear to have no connection with the nuclear wall. The black bodies have now taken up their position at the outer margin of the perikaryoplasm and form a complete ring about it. It will be noticed that in the preceding figures the linin is more or less of the nature of a granular thread. It now appears to have lost its granular structure and appears in the form of very fine threads. We now have the first indication of the breaking down of the nuclear wall, and at that part where the nuclear wall is broken, the perikaryoplasm is commencing to take on the form of a network and stains violet. r wb I. VAX is ] LA pt B [ Lawenn i y Var Inq Ro oe SEP qe 184 Fig. 7. Fig. 8. Fig. 9. CALIFORNIA ACADEMY OF SCIENCES. [PRoc. 3D SER. EXPLANATION OF PLATE XXXIV. The nuclear wall has nearly all disappeared and the linin has now formed a network. The perikaryoplasm is being gradu- ally transformed into a network of threads which stain deep violet, in contrast to the orange of the surrounding peri- karyoplasm. This stage is about the same as fig. 7, only the network appears to be of a more definite nature. The nuclear wall has entirely disappeared and the area once occu- pied by the nucleus is now filled with a clear network of threads; where the threads cross one another granules are present. The entire contents of the cell is now beautifully differentiated into three parts: the outer cytoplasmic reticu- lum which stains light blue; perikaryoplasm which stains orange, and the central network which stains deep violet. The figure also shows that the central network has in three places grown out into projections. This shows the projections to have grown considerably and the meshes of the central network are more elongated, as if pulled out. In this figure the projections have grown out to such an extent and the meshes have become so elongated that they form dis- tinct fibres. The more definite the fibres, the more deeply vio- let they stain. The figure also shows the vacuolated nucleolus. This stage is probably earlier than fig. 11, as there are no fibres developed as yet. It shows the central network grown out in three distinct projections, and we have the outline of a multi- polar spindle. [Lawson] Prats XXXIV. coc CaLAcAD.StI.g> Ser. Bor. Vout. = cng ’ \ 186 = Fig. Fig. Fig. Fig. 14. 16. CALIFORNIA ACADEMY OF SCIENCES. [PrRoc. 3D SER. EXPLANATION OF PLATE XXXV. Here we see a characteristic multipolar spindle, the fibres straight and definite, and the cones terminating in sharp points. We also see fibres with free ends projecting from the apices of the cones. These eventually become the mantle-fibres in the mature spindle. It will also be noticed that the chromosomes have taken up a more central position. This figure shows that certain of the poles have approached each other and we have the indication of a bipolar spindle. The mantle-fibres have also grown to a considerable length. This figure shows the mature bipolar spindle. The chromosomes lie in the equatorial plate. The mantle-fibres have developed to an extraordinary length; those from one pole cross those from the opposite pole. “The mantle-fibres, although very long, do not extend beyond the perikaryoplasm. The contractile fibres, which are connected with the chromosomes, are plainly compound in nature. There are also continuous fibres to be seen which extend from pole to pole uninterrupted. This stage is a little later than that shown in fig. 15. The chro- mosomes have separated and have commenced to move towards the poles. The poles of the spindle terminate in sharp points. The mantle-fibres have reached their maximum development. By the contraction of the contractile fibres the chromosomes have been drawn toward the poles. The mantle-fibres now appear to be more divergent from their respective poles. The chromosomes have now reached the poles. The mantle- fibres have lost their straight appearance and hang loosely alongside of the remaining continuous fibres. 7 ¥ = i [Lawson] Phare 3000 rel Ser 37 Ser. Bor. Vo NEP @ Sorel r CALABAR = a] Dn ye 7 ’ : 7 *. . aa 188 lig. Fig. Fig. ig. 20. Os CALIFORNIA ACADEMY OF SCIENCES. [Proc. 3D SER. EXPLANATION OF PLATE XXXVI. Here we see each of the daughter-nuclei surrounded by a membrane, with the chromatin stained blue, and in the form of a necklace. There is also seen a nucleolus in each daughter- nucleus. The continuous fibres still connect the nuclei with one another. The granular substance appears to surround the nuclei except where the continuous fibres are connected with the nuclear membrane. The perikaryoplasm now entirely surrounds each daughter-nucleus. The connecting fibres have curved out toward the cell-wall but can no longer be traced to the walls of the nuclei. The old spindle-fibres have disappeared and each of the daughter- nuclei is surrounded by a definite zone of perikaryoplasm. The black bodies form a ring at the outer margin of each zone. This figure shows the two spindles of the second division lying at right angles to one another. The one is only represented by the cross-sections of its fibres, the other shows the entire spindle with the chromosomes on their way to the poles and with the characteristic mantle-fibres. This figure also shows that the perikaryoplasm has accommodated itself to the shape of the spindles. In one view the zone is round and in the other oblong, corresponding to the size and shape of the spindle. Here the spindles of the second division are more nearly in the same plane than in the preceding figure. The chromosomes are at the poles. The mantle-fibres of the two spindles have reached over and united, thus connecting the nuclei with one another. Here we see three of the daughter-nuclei in the same plane, each surrounded by a nuclear wall. The chromatin is in the spi- reme stage. All the daughter-nuclei are connected by fibres which almost fill the entire cell-cavity. These fibres are curved out toward the cell-wall, and we see swellings on some of them. What little of the perikaryoplasm is left is scattered irregu- larly through the cell. THE ORIGIN OF THE KARYOKINETIC SPINDLE IN PASSIFLORA CC@RULEA LINN.! BY CLARA L. WILLIAMS, M. S. PLaTtEs XXXVII-XL. RecEnT work on karyokinetic division has made it appear very probable that no centrosome is present in the higher plants. In view of this the origin of the spindle becomes a subject of especial interest. The study of karyokinesis in the pollen-mother-cells of Pass¢flora cwrulea was taken up in the hope of shedding some light on this question. The origin of the spindle is the sole question considered in this investigation other matters receiving only casual attention. Among the fixing fluids experimented with in the prepar- ation of material were Wilson’s sublimate-acetic, Boveri’s picro-acetic, Flemming’s strong mixture, Flemming’s strong mixture diluted with an equal part of water, and 2 per cent. iridium chloride. When anthers fixed in Wilson’s, Bo- veri’s, and dilute Flemming’s solutions were crushed and examined in their respective fluids they showed such shrink- age that these fluids were not used in making preparations. Good results were obtained with iridium chloride, but no better than with Flemming’s strong mixture (undiluted), which was used for fixing the greater part of the prepara- tions studied. The anthers were left in the fixing fluid for twenty-four hours, and afterwards washed in running water for six hours. 1 Contributions from the Botanical Laboratories of the University of California, No 4. Presented for the degree of Master of Science. Prepared under the direction of Mr. W.J. V. Osterhout. [ 189 | April 13, 1899. IgO CALIFORNIA ACADEMY OF SCIENCES. [PRoc. 3D SER. They were then placed in the inner part of a dehydrator! and barely covered with water. The outer part of the dehydrator was filled with 95 per cent. alcohol. After remaining in the dehydrator for twenty-four hours, the material was removed to 95 per cent. alcohol and left for six hours. It was then run up successively through a mixture of equal parts of 95 per cent. of alcohol and absolute alcohol, absolute alcohol, a mixture of equal parts of absolute alcohol and bergamot oil, bergamot oil, a mixture of equal parts of bergamot oil and paraffin (43°), and a mixture of parattin (43°) and paraffin (52°). The anthers remained six hours in each. When in bergamot oil they were placed on the paraffin oven. Paraffin (52°) was used for imbedding and the microtome sections were cut from 3 to 5 microns in thick- ness. The sections were fixed to the slide either with albu- men (the sections being first spread out on a layer of water over the albumen, after which the slides were placed upon the paraffin oven to dry) or with 80 per cent. alcohol according to Eisen’s method (Eisen 1897). The latter method is preferable as the sections may be stained almost immediately. After mounting, the sections were stained with the Flemming triple stain. They were first placed in safranin, where they remained twenty-four hours. On removal they were decolorized with 95 per cent. alco- hol until the stain was removed from everything except the nucleolus. They were then washed with water and placed in gentian violet, where they were allowed to remain from five to fifteen minutes. There was then poured over the slide in rapid succession, a concentrated solution of orange G, 95 per cent. alcohol, and absolute alcohol. Clove oil was then poured upon the slide and the progress of decolorization watched under the microscope. ' The dehydrator was made in accordance with the suggestion of Prof. W. A. Setchell, as follows: A funnel of appropriate size is deprived of its neck and supported in the mouth of a tumbler or beaker. A piece of parchment paper folded in the same manner as for filtering is placed in the funnel. A cover fits over the top of the tumbler to pre- vent evaporation. Alcohol is placed in the tumbler; the material is placed in the funnel and covered with water. The rapidity of dehydration depends upon the quantity of water, the strength of the alcohol, and the thickness of the parchment paper, and can be kept under perfect control. Bor.—VoOL, I.] WILLIAMS—PASSTFLORA CQERULEA. IOI When the sections were sufliciently decolorized the clove oil was removed by placing the slide in xylene for a few moments. The sections were finally mounted in Can- ada balsam. In the young pollen-mother-cell, when the chromatin is still in the spireme stage (fig. 1), the cytoplasm is composed of two distinct elements. One is fibrous and forms a net- work throughout the cell from the nuclear wall to the cell- wall. The other is granular and is uniformly distributed within and upon the meshes of the network. The threads of the reticulum are knotted and sinuous and take the violet stain. The granular part of the cytoplasm usually stains yellow or yellowish brown. About the time when the chromatin thread breaks up a change takes place in the cytoplasm. The meshes of the reticulum immediately sur- rounding the nucleus are drawn out parallel to the nuclear wall and form a sort of weft about it (fig. 2). This con- dition of the cytoplasm is of short duration. Soon the meshes of the cytoplasmic reticulum assume a new arrange- ment. They become drawn out at right angles to the nuclear wall and appear as if radiating from the nucleus (fig. 3).