! f'JJiTf gUH: ; /u fva; i/A v$aOSANCElfj> >& * aqq. ^o q.dodaa -[enuu^ q^uaa^ J.T qq. The Yueeeae. BY WILLIAM TRELEASE. (FROM THE THIRTEENTH ANNUAL REPORT OF THE MISSOURI BOTANICAL GARDEN .) Issued July 30, 1902. THE YUCCEAE.* C? BY WILLIAM TKELKASE. INTRODUCTION. H The large family Liliaceae has been subjected to very different treatment by the writers who at various times have monographed it or attempted to indicate a natural > sequence for its genera. The tribes Aloineae and Yuccoi- deae, respectively African and American, were treated to- -4 gether by Mr. Baker f with the implied recognition of ^ close affinity, the principal synoptic differences between them consisting in the succulent leaves and gamophyllous perianth of the former, and the less succulent more fibrous leaves and distinct perianth segments of the latter, in which he includes Yucca, Hesperaloe, Herreria, Beau- carnea, and Dasylirion.\ Bentham and Hooker also place the aloids and yuccoids ^ close together, characterizing the tribe Dracaeneae, in which the latter are included, by its mostly distinct perianth V segments, U and including in it Hesperocallis, Hesperaloe, Yucca, Nolina (Beaucarnea), and Dasylirion, of the New * Presented in abstract, with lantern illustrations, before the Botan- ical Society of America, at its New York meeting, June 28, 1900, and before the Academy of Science of St. Louis, Feb. 3, 1902. t Journ. Linn. Soc. Bot. 18: 148. (1881). J 1. c. 152. Genera Plantarum. 3: 750, 777. (1883). t The generic descriptions show that the segments are connate into a tube in Hesperocallis , Dracaena, Cordyline, Milligania, and some species of Astelia, and barely united at the base in Yucca. 1. c. 778. (27) 271935 28 MISSOURI BOTANICAL GARDEN. World, and Dracaena, Cordyline, Astelia and Milligania, of the Old World, while the South American Heireria is removed to another tribe. Professor Engler,* who treats the Aloineae as pertaining to a group placed at a considerable distance from the Dracaeuoideae, includes in the latter the Old World Dra- caeneae, of the genera Cohnia, Cordyline, Astelia, and Milligania, with perianth segments connate at base, and the New World groups Nolineae, of the genera Nolina and Dasylirion, and Yucceae, of Yucca and Hesperaloe, with the segments distinct. f Hesperocallis is very properly removed to another group. The present paper deals only with this group Yucceae of Engler, and includes the principal conclusions reached in an intermittent herbarium, garden and field study extend- ing over the last sixteen years, in the course of which nearly all of the spontaneous species have been examined and photographed in their native homes and many of them in- troduced or reintroduced into cultivation in this country and Europe from definitely located sources. In its alliance, the group Yucceae is characterized by the possession of similar subequal withering-persistent petaloid perianth segments, a 3-celled ovary with more or less in- truded dorsal false septa, many ovules 2-ranked in each cell, a subterete elongated embryo obliquely placed across the seed, and germination with arched cotyledon.} * Natiirl. Pflanzenfamilien. II Teil. 5 Abteil. 19, 70. (1888). t It is to be observed that, with most writers, Engler speaks of the segments as free or somewhat united at base, in his generic description of Yucca. 1. c. 70. J In all of the genera of this group, in germination the cotyledon as- sumes an arched form, with the seed remnant on or in the soil (from which it is ultimately raised in some cases), instead of directly carrying this up on its end as it commonly does in Liliaceae. See The Garden. 8 : 300. /. Gard. Chron. n. s. 24: 216. Lubbock, Contr. Knowl. of Seed- lings. 2 : 578, 613. Copeland, Bot. Gaz. 31 ; 419. /. 3. THE YUCCEAE. 29 REVISION OF THE YUCCEAE. The genera constituting the group appear to admit of most natural limitation as follows : Flowers oblong or narrowly campanulate, scarcely 15 mm. wide, rosy- red or greenish: filaments shortly adnate to the petals below, slender, erect, inflexed at apex; anthers oblong: style filiform, minutely papillate about the scarcely enlarged stigma. Hesperaloe. Flowers globose or broadly campanulate, spreading to a width of 50 to 100 mm., white or creamy, often tinged with green, bronze or violet : filaments clavately enlarged ; anthers shortly sagittate. Style filiform, abrupt; stigma capitate, long-papillate : filaments ad- nate to the petals below, erect. Hesperoyucca. Style stout or wanting, gradually if at all narrowed; stigma openly perforate, not papillate, more or less deeply 6-notched : fila- ments mostly outcurved at apex. Perianth polyphyllous, or the segments barely connate at base, to which the filaments are slightly attached. Segments of perianth thick, mostly inflexed: style wanting: nectar glands in walls of ovary small. Clistoyucca. Segments thin and petaloid, spreading at night: style evi- dent : nectar glands large but mostly inactive. Yucca. Perianth gamophyllous and tubular below, the stamens inserted in its throat, otherwise as in Yucca. Samuela. HESPERALOE Engelmann. Perianth oblong or narrowly campanulate, of subequal closely applied distinct oblong succulent segments out- curved at tip. Filaments adnate to base of perianth, slender, erect, inflexed at apex; anthers oblong, introrse. Ovary ovoid, shorter than the long slender style ; stigma not enlarged, minutely papillate and perforate. Fruit capsular, globose-oblong, rugose-veiny, 3-celled, 6- valved at least above, the valves with short solid erect beak. Seeds thin, flat: albumen not ruminated. Subacaulescent plants with filiferous-margined long con- cave striate scarcely pungent smooth leaves, and loosely panicled few-branched inflorescence. SYNOPSIS OF SPECIES. Flowers rosy-red or salmon-colored. H. parviflora. Flowers green, tinged with purple. . H. funifera. 30 MISSOURI BOTANICAL GARDEN. H. TARVIFLOKA (Torrey) Coulter, Contr. U. S. Natl. Herb. 2: 436. (1894.) H. yuccaefolia Engelmann, Bot. King. 497. (1871). Trans. Acad. St. Louis. 3 : 55. Baker, Gard. Chron. 1871 : 1516. Journ. Linn. Soc. Bot. 18 : 231. Watson, Proc. Amer. Acad. 14 : 250. Yucca (?) parviflora Torrey, Bot. Bound. 221. (1859). Baker, Gard. Chron. 1870:923. Y. paviflora Hemsley, Garden. 8 : 132. Aloe yuccaefolia Gray, Proc. Amer. Acad. 7:390. (1867). Gard. Chron. 1870:1092. Usually cespitosely suckering. Leaves arcuately spreading, 1 to 1.25 m. long, something over 25 mm. wide, striate-ridged on the back. In- florescence 1 to 1.25 m. high, the few branches divaricate, glabrous and subglaucous. Flowers fascicled above the bracts, on soft articulated rosy pedicels, ephemeral, rosy, tubular, mostly about 35 mm. long; style long-exserted. Capsule something over 25 mm. long; seeds 5X 8 mm. Plate l,f. 1. Southwestern Texas; between the Rio Grande and the southern part of Valverde County, Kinney County, and the western part of Zavalla County. Plate 84, f. 1. One of the puzzling plants brought in by the naturalists of the United States and Mexican Boundary Survey, col- lected between the mouth of the Pecos and the Nueces, was described by Dr. Torrey * under the name Yucca? parvi flora, the description of the filifcrous Yucca-like leaves and of the inflorescence being good, but that of the flowers and fruit indifferent, the perianth noted as "white? ", and the unripe fruit as " doubtless fleshy." In his enumeration of the known forms of Yucca in 1870, Mr. Baker, referring to dried specimens in the Kew herba- rium, as well as to the original description, characterizes the plant in much the same way, but observes that the flower is more like that of an Ornithogalum of the Pyrenaicum group than that of its neighbors of the genus Yucca. Mention is also made of the peculiarity of the flowers in an article on Yucca by Mr. Hemsley, who, evidently through a typographical error, calls the species Y. paviflora. * Emory, Kept. U. S. & Mex. Bound. Surv. 2. Botany of the Boundary by John Torrey. 122. Referred to in this paper as " Bot. Bound." THE TUCCEAE. 31 Before these articles by Baker and Hemsley were pub- lished, living specimens had been sent to Dr. Gray, and an examination of flowers which these bore in the Harvard Bo- tanical Garden showed the generic distinctness of the plant from Yucca, and so strong a resemblance to the true Aloes of Africa that Dr. Gray did not hesitate to transfer it to the genus Aloe, under the new and descriptive specific name yuccaefolia. The redescription shows that the flow- ers are pale red and the fruit capsular. Recognizing sufficient differences between this American Ywcca-leaved and Twcca-fruited Aloe and the African plants properly representative of that genus, Dr. Engel- mann* created for it the genus Hesperaloe, in 1871, noting that the leaves, pollen and seeds are those of Yucca, the perigone and pistil are those of Aloe, and the filaments, adnate at base and geniculate upwards, resemble those of Agave. This description was repeated by Mr. Baker the same year, the specific name yuccaefolia, introduced by Dr. Gray, being employed in both instances. The original specific name proposed by Dr. Torrey was restored, in combination with the generic name Hesperaloe, by Professor Coulter in his account of the botany of west- ern Texas, in 1894. Notwithstanding its beauty and unusual characters, little is known of this plant in its typical form, aside from the original observations of Torrey, Gray, Baker and Engel- mann. The only herbarium specimens that I know of were collected by Wright: in June, 1849, be- tween the Nueces river and Elm creek and on the banks of the latter ; apparently in the autumn of the same year, on hills of Devil's river; and May 15, 1851, between the Leona and Nueces. f * King, Eept. U. S. Geol. Explor. Fortieth Parallel. 5. Botany, by Sereno Watson. 497. Referred to here as "Bot. King." f For the localities represented by specimens contained in the Gray herbarium, I am indebted to Miss Mary A. Day. 32 MISSOURI BOTANICAL GARDEN. In April, 1900, while passing a day in San Antonio, Texas, I observed a lltsperaloe planted in one of the plazas of that city, which in its long arching concave filiferous leaves, oblong Aloe-red flowers with white styles pro- truding for a distance equal to one-third or one-half the length of the perianth, and very short anthers, agreed with the description and scanty available herbarium material of //. yuccarfolia, and from this plant, offsets of which are now growing in the Missouri Botanical Garden, the following notes have been made. The flowers are ephemeral, and their original appearance would scarcely be guessed from the withered remains after they have fallen, or from such herbarium material as is usually seen. Though the buds are erect, the soft, rosy articulated pedicels ultimately arch over, so that the ex- panded flowers are horizontal or more frequently pendent. In texture they are suggestive of Lapageria, and this re- semblance, notwithstanding their smaller size and some- what different form, is increased by their beautiful outward shading with rose-color, on a creamy ground color which prevails on the inner surface. The firm succulent distinct but closely appressed segments of the perianth are about half a millimeter thick in the middle and outwardly recurved near the end, which, as in Yucca, is tipped with a minute tuft of white hair- like papillae. The inner segments are 8 or 9 mm. wide, and the outer segments a little narrower. The white or rosy slightly tapering filaments are adnate to the seg- ments for a short distance and then stand erect, with the very slender apex abruptly incurved so as to make the oblong versatile anthers suberect and introrse, close against the filaments, with their abundant bright yellow powdery pollen exposed toward the style. The conical- ovoid greenish ovary is very slightly 6-grooved, and the white style, somewhat tapering and triquetrous near the base, soon becomes filiform and terete except for three THE YUCCEAE. 33 faint grooves which persist to the very inconspicuously 3-lobed perforate somewhat fimbriate stigma. The ovary possesses three large plane septal nectar glands, passing outward at top into conducting grooves which open at the base of the pistil, and the abundant secretion of which, when not removed, drips to the mouth of the pendent flower so that toward the end of the day, when the flower closes, the anthers, style and perianth are gummed together into a nearly inseparable mass. The ovules resemble in shape and arrangement those of the capsular species of Yucca, and the erect capsule and thin flat black seeds are equally suggestive of this section of Yucca. H. parviflora Engelmanni (Krauskopf) Trelease. H. Engelmanni Krauskopf, Notice to Botanists, etc., Aug. 1878 [cir- cular]. Watson, Proc. Amer. Acad. 14 : 250. (1879). Baker, Journ. Linn. Soc. Bot. 18:231. Coulter, Contr. U. S. Natl. Herb. 2 : 436. H. yuccaefolia Garden. 18 : 188. 20 : 71, 361. 21 : 324. Gard. Chron. n. s. 18 :87, 109, 199. /. 34. Andrt, Rev. Hort. 58 : 64. Hooker, Bot. Mag. iii. 56. pi. 7223. Flowers oblong-campanulate, about 25 mm. long; styles scarcely ex- -ceeding the perianth. Plates l,f. 2. 2. Southwestern Texas, about the head of the west fork of the Nueces river. In 1878, Mr. E. Krauskopf, of Fredericksburg, Texas, issued an advertising circular mentioning H. yuccaefolia and offering for sale plants of a Hesperaloe which he had brought from the western dry branch of the Nueces river and for which he proposed the name H. Engelmanni. The flowers are described as bell-shaped, red, with short thick style and anthers as much as a quarter of an inch long, whereas in H. yuccaefolia the latter are said to be several times shorter than the filiform style. Specimens of this supposed second species were sent to Dr. Engelmann, through Lindheimer, and are noted in his herbarium as having been collected by Meusebach, though they are evi- dently of the collection referred to by Krauskopf. 34 MISSOURI BOTANICAL GARDEN. Some time after this, John Saul, of Washington, sent flowers of Hesperaloe, from the Nueces river, to the editor of The Garden, under the name of H. yuccaefolia* and at about this time the genus seems to have gone into one or more English gardens, probably from this source. f The same form apparently was again introduced into En- gland in 1888, % but I have not learned from what source. Dr. Watson, in his revision of the North American Liliaceae, shortly after the discovery of H. Enrjelmanni, mentions this proposed species as from the same region as //. yuccaefolia, but imperfectly known, though perhaps to be distinguished by the more slender and flexuous branches of its inflorescence, smaller bracts, twice longer anthers, and stouter included style scarcely longer than the ovary. A similar equivocal mention was made in 1880 by Mr. Baker, of H. Engelmanni, which is ignored by Professor Engler, but distinctly recognized by Pro- fessor Coulter in his Botany of western Texas, in connec- tion with the earlier species. So far as the evidence goes, all of the Hesperaloe culti- vated in Europe, and to which reference has been made above, belongs to this second form, and may perhaps have been derived from Krauskopf 's original collection. In May, 1900, a plant procured some three years before fromMr.P.J.Berckmans,1f and itself possibly derivedfrom Krauskopf, originally, came into bloom at the Missouri * Garden. 18: 188. From the phraseology of a quotation from Mr. Saul, it may be inferred, perhaps, that the plant bearing these flowers was derived originally from Krauskopf. t See The Garden. 20: 71, 361. 21: 324, where a plant is said to hare been in continuous bloom from July 1881 until May 1882, with promise of continuing to flower for another month or two. Gard. Chroa. n. s. 18 : 87; 109, 199. /. 34. I Curtis's Bot. Mag. iii. 56. pi. 7223. Engler & Prantl. 1. c. 71. 1 See Berckmans, Gard. Monthly. 1883: 323. Wiener 111. Gart.- Zeit. 11 : 268. THE YUCCEAE. 35 Botanical Garden, and continued to flower until well into the fall. The first flowers which opened, though shorter than those of the San Antonio plant referred to H. parvi- flora, and consequently broader relatively to their length, possessed the conspicuously exserted white style and short anthers (scarcely over 2 mm. long) of that species. After the first few flowers, those which opened were relatively much broader, because of a considerable actual shortening, so that the expression bell-shaped, which has been used for H. Engelmanni, might be applied to them, and the style was not exserted, merely reaching to the mouth of the perianth, and, in fact, was slightly shorter than the stamens. Except for having their anthers a very little shorter, these flowers are the counterpart of a well-pre- served specimen of the original of H. Engelmanni sent to Dr. Engelmannby Lindheimerin 1878, though the included style of the latter is a little longer than the stamens. Still later flowers of the same plant, while preserving the short broad form, again had the style a little exserted (Plate 2). As in typical H. parviflora, the leaves, which are deeply concave and with free marginal fibers, differ in width, as indeed, is usual in the genus Yucca, and the inflorescence, which in vigorous plants has a few spreading branches, may sometimes be simple, in either case the fascicled flowers continuing to develop in succession for many months, and varying from deep rosy-red, when well lighted, to a salmon-color, when shaded from strong light. For the present, this short-flowered plant, with the style included or very slightly exserted, and which seems to come from a point a little north of but very close to the known range of H. parvijlora, appears to be varietally separable from the latter in these characters, and should bear the name Engelmanni given to it as a specific name by Kraus- kopf. 36 MISSOURI BOTANICAL GARDEN. H. funifera (Koch) Trelease. H. Davyi Baker, Kew Bull. 1898 : 226. //. Engelmanni Baillou, Hist, des PI. 12 : 511. Urbina, Cat. PI. Hex. 352. Yucca funifera Koch, Belg. Hort. 12:132. (1862). Lemaire, 111. Hort. 13:99. (1866). Baker, Journ. Linn. Soc. Bot. 18:228. Agave funifera Lemaire, 111. Hort. 11 : Misc. 65. [66a]. (1864). Often cespitose. Leaves larger, at length less concave, often with much coarser marginal fibers. Inflorescence 2 to 2.5 m. high, few branched near the top. Pedicels and flowers purplish green, glaucous, the latter about 25 mm. long; style scarcely exserted. Capsule 25 to 50 mm. long, with strong beak, the false septum evanescent or protruding into the cell only toward the base, where it forms a large thin tooth ; seeds 6X9 mm. Plates 3. 4,f. 1. 81, f. 8. Northern Mexico, between the Rio Grande and the Sabinas, and, apparently, in the state of San Luis Potosi (Pringle, 3911). Plate 92, f. 1. The Engelmann herbarium contains a fruiting fragment, at first referred to Yucca but afterward to ffesperaloe, col- lected in 1847 by Dr. Wislizenus at Cerralvo, northeast of Monterey. Similar capsules were brought by Dr. Parry, in 1878, from the plains between Monterey and the Rio Grande." The herbarium of the Field Columbian Museum contains excellent specimens of the same plant from Buste- mente, in the State of Nuevo Leon, collected by Henry W. Wood in July, 1900. In 1891, Mr. Pringle made good leaf and fruit specimens, representing the same genus, at the Hacienda de Angostura, east of San Luis Potosi, which were distributed as IT. Engelmanni, under the number 3911, and so referred to by Bail! on. In March, 1900, when going over the Mexican Interna- tional railroad, north of the Sabinas river, I observed a considerable quantity of what was evidently a Hesperaloe, with persisting capsules of the preceding year, which came down to the railroad only on the higher ridges through which cuts had been made. Toward the end of April, when the plants had begun to bloom, I visited this region again, and some six kilometers south of Peyotes collected THE YDCCEAE. 37 herbarium specimens and viable seeds of the plant. This Hesperaloe appears to be the same as the herbarium material referred to, though neither foliage nor flowers accompany the capsules first collected, and the few flowers distrib- uted by Mr. Pringle from further south are not in very satisfactory condition while the marginal threads, which are slender in the many plants seen by me, are very thick, triquetrous, wavy and rigid on his leaves. This species, the at first very concave leaves of which may be as much as 40 mm. wide and nearly 2 m. long, finely striate-grooved on the back and with long con- spicuous marginal fibers, as in the other representatives of the genus, produces a divaricately few-branched, tall panicle, on which, fascicled in the axils of the bracts, are borne the oblong ephemeral flowers. Unlike those of H. parvi flora and its variety Engelmanni, both of which have pedicels and flowers ranging from a creamy tint through salmon-color to typically a beautiful shade of red sugges- tive of Aloe and Gasteria, the flowers and short pedicels of this species are noted by Mr. Pringle as being " purplish, shading to whitish," and in the plants observed about Pe- yotes were of a dingy purplish green and decidedly glau- cous, the spreading flowers being about 25 mm. long, with stamens and style included and of about equal length, and the anthers 5 to 7 mm. long. The globose to broadly oblong solid-beaked capsules are strongly transversely reticulate- veined, and the thin black seeds are like those of the other species. In 1898 Mr. Baker described, under the specific name Davyiy a green-flowered Hesperaloe from " California? " which had been sent him by Mr. J. Burt Davy from the garden of the University of California at Berkeley. Mr. Davy tells me that no record is found of the source of the seeds from which this was grown. Dr. F. Franceschi, of Santa Barbara, California, states that two original plants were raised, one of which flowered in 1898, yielding the ti 38 MISSOURI BOTANICAL GARDEN. material on which Mr. Baker's description was based, while the other was secured by Dr. Franceschi, who has since sent vigorous suckers from it to Kew and to the Missouri Botanical Garden, these suckers having formed after the plant bloomed. It is not improbable that the seeds from which these plants were raised were derived from Mr. Pr in- gle's collection of 1891, and the living plant which I have examined shows, as would hardly have been expected from Mr. Baker's description, leaves at first as concave as those of the other species of Hesperaloe, and quite indistinguish- able from those of the plants seen below Peyotes, so that it seems safe to refer all of these specimens of the Mexican table land to //. Davyi, which appears therefore to be rather widely distributed and which differs markedly from the Texan forms in the color of its flowers. Many years ago the Tonels introduced into European gardens a plant which seems never to have flowered there, and which was mentioned a number of times under the gar- den name Yucca funif era. No Yucca is yet known which possesses channeled filif erous dorsally striate leaves com- parable to those of Y. funif era as described, and though its apparent complete disappearance from cultivation makes its identity a matter of conjecture only, the foliage description so well fits this Mexican species of Hesperaloe as to leave little doubt in my mind that the latter should bear the name H. funif era . HESPERO YUCCA (Engelmann) Baker. Perianth broadly campanulate, of subequal distinct thin broadly lanceolate concave segments. Filaments evidently adnate to perianth below, clavate, suberect; anthers didy- mously cordate. Ovary oblong-ovoid or obovoid, mostly longer than the short slender style; stigma capitate, long- papillate, minutely perforate. Fruit capsular, incompletely 6-celled, 3-valved through the laciniate false septa. Seeds THE YTJCCEAE. 39 thin, flat ; albumen not ruminated. Subacaulescent plants with straight needle-pointed rough-margined flat leaves, and ample panicle. H. WHIPPLEI (Torrey) Baker, Kew Bull. 1892: 8. Tre- lease, Eept. Mo. Bot. Gard. 4: 208. pi. 16, 23. Yucca Whipplei Torrey, Bot. Bound. 222. (1859). Baker, Gard. Chron. 1870:828. 1871 1 1516. n. s. 6: 196. /. 42. n. s. 23: 796. Journ. Linn. Soc. Bot. 18 : 230. Palmer, Amer. Joum. Pharm. 50 : 687. Garden. 27 : 266. 35 : 561. /. Engelmann, Bot. King. 497. Trans. Acad. St. Louis. 3:54, 214, 372. Watson, Proc. Amer. Acad. 14 : 254. Bot. Calif. 2 : 164. AndrS, Rev. Hort. 58 : 67. /. 13, Smith, Gard. Chron. iii. 13 : 749. Coville, Contr. U. S. Natl. Herb. 4: 203. Merriam, N. Amer. Fauna. 7:359. Tre- lease, Eept. Mo. Bot. Gard. 3: 164. pi. 11,12, 54. Gard. & For. 8 : 414-5. /. Hooker, Bot. Mag. iii. 65. pi. 7662. Land of Sun- shine. 11 : 251. /. Orcutt, West Amer. Scientist. 6 : 134. Y. Whipplei glauca Wiener 111. Gart.-Zeit. 14 : 197. Y. Whipplei graminifolia Baker, Journ. Linn. Soc. Bot. 18 : 230. Y. aloifolla Torrey, Pac. R. R. Rept. 4 : 147. Y. fllamentosa Home and Flowers. II 2 : 12. /. Y. graminifolia Wood, Proc. Phil. Acad. 1868 : 167. Y. Ortgiesiana Roezl, Belg. Hort. 1880 : 61. Y. Engelmanni Gard. Chron. n. s. 14 : 43. (1880). ? Y. Californica Greenland, Rev. Hort. 1858 : 434. Lemaire, 111. Hort. 10 : after pi. 372. (1863). 13 ; 96. Gard. Chron. n. s. 5 : 794, 829. Simple or, in the mountains, frequently cespitose. Leaves ascending, rigid, .3 to 1 m. long, about 15 mm. wide, plano-convex, subtriquetrous, or keeled on both faces, sometimes falcate, striate, glaucous, keenly but finely denticulate, with very slender pungent end spine. Inflorescence 2 to 5 m. high, oblong, long peduncled, glabrous. Flowers Yucca-like, pendent, fragrant. Capsule about 5 cm. long: seeds 6 to 7X8 mm. Plates 4, f. 2. 5. 81, f. 9. California, from the mountains above Monterey to the vicinity of Alamo, lower California; eastward to the vicin- ity of San Bernardino Plate 84, f. 1. Yucca WJiipplei is the name proposed by Dr. Torrey, and still commonly employed, for a plant which, when in bloom, forms one of the most striking and beautiful fea- tures of the Coast-range vegetation of southern California. 40 MISSOURI BOTANICAL GARDEN. From all other Yuccas it differs in the slender style rising- abruptly from the top of the ovary and capitately enlarged into a papillate stigma, and in possessing somewhat gluti- nous pollen, as well as in certain capsular characters, which led Dr. Engelmann * to give it the sectional name Hespero- yucca, which both Mr. Baker and the writer have proposed to employ as a generic name. Though the mountain and valley forms vary greatly in amplitudeof panicle, etc., only one species of Hesperoyucca appears capable of characterization, and this has long been in cultivation in European gardens, partly under the name Yucca Whipplei and partly under the name Y. Calif arnica* which has further been applied to very diverse things. If it were certain that the brief foliage description given by Greenland in 1858 really refers to this plant, the specific name Californica has a slight priority over the name Whipplei, which though written in 1858 was not published until the following year, but the propriety of this substitu- tion of name is open to considerable question. Y. graminifolia f Wood, from the vicinity of Los Angeles, though the leaves are described as more flaccid, can hardly refer to other than the typical form, which to the north of Los Angeles becomes very large, and the name is not there- fore applicable to the plant that is abundant about San Bernardino, e. g. at Arrowhead Springs and in the Cajon pass, as I at one time thought might be the case. J This latter plant very frequently has the flowers shaded with purple or violet, and it was to one of the most pronounced of these tinted forms that M. Andre in 1884 applied the name Y. Whipplei violacea, though the name stands for too inconstant a character to have more than horticultural value. * Bot. King. 497. (1871). t This name had been applied, in 1837, to the plant subsequently named Dasylirion graminifolium. J Kept. Mo. Bot. Gard. 4: 2U.pl. 17, 23. Rev. Hort. 56 : 324. pi. THE YUCCEAE. 41 No other species of this type could have been collected about San Diego, where H. Whipplei occurs in abundance, by Roezl, who in 1869 reintroduced it into European gardens through De Smet, under the name Y. Ortgiesiana, so that there appears no doubt as to the proper reference of this synonym. On April 3d, 1858, Professor Newberry collected leaves of a plant growing in tufts on rocks ' ' at the mouth of Diamond river, at the eastern end of the grand canon of the Colorado, in northern Arizona, which neither Professor Torrey* nor Dr. Engelinann could distinguish from those of this species as collected by Bigelow at the Cajon pass in California. The single leaf of Newberry's collection in the Engelmann herbarium is glaucous, falcate, elongated and scarcely to be referred elsewhere, but the locality is so far from the known range of this species on the other side of the desert as to warrant doubt as to the correctness of the record, and I know of no confirmation of this isolated locality. CLISTOYUCCA (Engelmann) Trelease. Perianth oblong to globose, of nearly distinct thick ob- long or lanceolate segments often incurved at end. Fila- ments nearly free, thickened, mostly outcurved above; anthers sagittate, horizontal. Ovary ovoid, tapering to the transiently stellate 6-lobed openly perforate stigma. Fruit dry, spongy about a papery core, 6-celled, indehiscent. Seeds rather thin, flat, nearly round ; albumen not rumi- nated. Large tree, with short thick and pungent rough- margined leaves and compact sessile panicle from an ovoid large-bracted bud. C. arborescens (Torrey) Trelease. Yucca Draconis (?) arborescens Torrey, Bot. Whipple. 147. (1857). F. brevifolia Engelmann, Bot. King. 496. (1871). Trans. Acad. St. Louis. 3 : 47, 213, 371. Palmer, Amer. Journ. Pharm. 50 : 587. * Ives, Kept, upon the Colorado river of the West. Part IV. Botany. 42 MISSOURI BOTANICAL GARDEN. Parry, Amer. Nat. 9: 141. Watson, Proc. Amer. Acad. 14: 252. Bot. Calif. 2: 164. Baker, Journ. Linn. Soc. Bot. 18: 221. Gard. Chron. n. s. 3 : 492. n. s. 26: 18. iii. 1: 772. /. 145. Land of Sunshine. 10: 1. Trelease, Kept. Mo. Bot. Gard. 4 : 193. pi. 6-9, 21. Schiraper, Pflanzengeographie. 669. /. 369. Y. arborescens Trelease, Kept. Mo. Bot. Gard. 3: 163. pi. 5, 49. (1892). Coville, Contr. U. S. Natl. Herb. 4: 201. frontispiece. Merriam, N. Amer. Fauna. 7: 353-8. frontispiece and pi. 13. Sargent, Silva. 10: 19- pi- 502. Large at length much branched rough-barked tree. Leaves spread- ing, less than .3 m. long, 15 mm. wide, plano-convex or triquetrous, striate, minutely denticulate, very rigid, pungently pointed. Inflorescence sessile, dense, often scabrous-hispid. Flowers sometimes puberulent, greenish-white, 25 to 50 mm. in diameter. Fruit ovoid, erect or var- iously directed, 50 to 100mm. long; seeds 10X12 mm. across, 1 to 1.5 mm. thick. Plates 6. 7. 85, f. 10. 87, f. 1. Mohave desert, California, to Detrital valley, Arizona, and the Beaverdam mountains, Utah. Plate 84, f. 2. The Joshua tree of the Mohave desert region, the largest and most imposing of the Yucceae of the United States, which was first called Yucca Draconis (?) arborescens by Torrey, subsequently Y. brevi folia by Engelmann, and which is now commonly known as Y. arborescens, differs in its collective flower and fruit character about as much from typical Yuccas as does Hesperoyucca. In separating it from Yucca, I have thought best to apply to it as a generic name the sectional name Clistoyucca under which Dr. Engel- mann* separates it from the other species of Yucca, since there can be no question as to the applicability of that name to this particular tree, though Dr. Engelmann f sub- sequently found it desirable to add Y. gloriosa to this sec- tion, to which the writer J afterwards added Y. gigantea. Only the one species is known. YUCCA Linnaeus. Perianth open-campanulate, of nearly distinct thin lanceo- late or ovate-lanceolate segments. Filaments nearly free, * Bot. King. 496. Trans. Acad. Sci. St. Louis. 3 : 47. t Trans. Acad. Sci. St. Louis. 3:213. J Kept. Mo. Bot. Gard. 9 : 142. THE YTJCCEAE. 43 thickened and outcurved above ; anthers short, sagittate, soon horizontal. Ovary oblong, mostly longer than the stout oblong or swollen style ; stigma unequally 6-lobed, openly perforate. Fruit nearly or quite 6-celled: erect, capsular, 6-valved above, and with thin seeds with the albumen not ruminated ( Chaenoyucca) ; variously pendent or erect, soon drying about a papery core, indehiscent, with thin seeds without rumination ( Heteroyucca} ; or pendent, bac- cate mostly about a papery core, indehiscent, with very thick seeds having the albumen ruminated ( Sarcoyucca ) . Acaulescent or arboreous plants occasionally of large size, with flaccid and pointless or usually rigid and very pungent entire, minutely denticulate, orfiliferous leaves, and mostly ample panicle. The true Yuccas, which (including Clistoyucca ) , in con- trast with his section ffesperoyucca, Dr. Engelmann* treated under the sectional name Euyucca, have for many years been in cultivation in considerable numbers, and hence under the eyes of both gardeners and botanists, but no ad- ditions have been made to the number of known spontane- ous species within recent years f except by the separation or rehabilitation of what had passed for varieties, forms or synonyms of described species, though some twenty years ago a number of hybrids, referred to below under Y. gloriosa, were introduced into cultivation, and it is certain that within the next few years our gardens will be still further enriched by many artificial hybrids between the known species. This genus is not only larger than any of the others of the group Yucceae, but has a much greater geographical dis- tribution, extending southwards from the great bend of the Missouri river to the table land north of the City of Mex- * Hot. King. 496. (1871). Trans. Acad. St. Louis. 3 : 34. t T. Pringlei Greenman, distributed from Mt. Ajusco, Mexico, in 1897 (Pringle, No. 6669), was subsequently shown by Mr. Greeuman to be Furcraea Bedinghausii.Proc. Amer. Acad. 83 : 474. (1898). 44 MISSOURI BOTANICAL GARDEN. ico, and, after a break of unknown extent, into the center of Central America, and eastwards to the Atlantic coast and the Bermudas and eastern Antilles. The capsular species are the prevalent northern form, and reach from South Dakota to the Mexican state of Durango, and from the Atlantic coast to Nevada, with the exception of the Great Lake region and the upper Mississippi river and its tribu- taries from the east. The baccate species with papery core are of the southern Rocky Mountain and western region, reaching the Pacific coast in the southern part of California and at the extremity of Lower California, and are the preva- lent form of the high table land of Mexico. A single spe- cies with coreless fleshy fruit appears to be restricted to the southern Atlantic coast of the United States, a small part of the Gulf coast, and some of the islands to the east, though it has given rise to a marked variety in the isolated peninsula of Yucatan ; and a single species with the foliage of this outlying species but forming a core in the fruit occurs in Central America, where, though abundantly culti- vated, its distribution is unknown. Several species and many varieties are known only in gardens, and two species with very aberrant fruit are of local distribution on the southeastern seacoast of the United States. Plate 99. KEY TO 8PKCIB8. Fruit erect, capsular, dehiscent. Seeds thin, flat, margined: albumen not ruminated. CHAKNOYTJCCA. Leaves finely filiferous (entire in forms of the second). Style oblong, white. Inflorescence a long-peduncled panicle (subracemose in some garden forms of Y. flaccida). Leaves lanceolate or spatulate, often plicate, at most very narrowly lined with gray or brown next the mar- ginal threads. Leaves rigid for the group, rather coarsely curly- flliferous, subspatulate. Segments of young fruit regularly convex. y. filamentosa. Leaves more flexible and attenuate, with finer etraighter threads. Segments of young fruit with angular facets. y. flaccida. THE YUCCEAE. 45 Leaves linear or linear-spatulate, white-margined. Leaves grass-like. Eastern Texas. Y. tenuistyla. Leaves more rigid and spreading. Western. Low. Seeds small. Y. constricta. Arborescent. Seeds very large. Y. radiosa. Inflorescence racemose or branched close to the leaves. Not arborescent. Leaves as in the last. Y. angustissima. Leaves lanceolate, often short. Y. Sarrimaniae. Style swollen, green. Inflorescence racemose or branched close to the leaves. Leaves linear, rather stiff. Seeds large. Y. glauca. Leaves grass-like, flexible. Y. Arkansana. Inflorescence panicled on a long scape. Leaves as in the last or wider. Y. Louisianensis. Leaves with a distinct thin yellow or brown horny finely denticulate border. Capsule mucronate, with flat-backed valves. Arborescent. Leaves linear to lanceolate. Y. rigida. Acaulescent. Leaves lanceolate. Y. rupicola. Capsule attenuate-beaked, with round-backed valves. Arborescent. Leaves linear. Y. rostrata. Fruit (so far as known) indehiscent. Fruit erect or pendent, soon drying. Seeds thin, flat, slightly mar- gined : albumen not ruminated. HBTKROYUCCA. Leaves finely denticulate, softly green-pointed. Y. gigantea. Leaves at most sparingly denticulate or filiferous, pungent. Leaves broad, rigidly ascending or spreading. Fruit mostly pendent. Y. gloriosa. Leaves more elongated, recurved. Fruit erect so far as known. Inflorescence close to the leaves, the latter relatively broad. Y. recurvifolia. Panicle long-stalked. Leaves narrower. Y. flexilis. Leaves crowded, regularly arcuate. Y. DeSmetiana, Fruit pendent, fleshy and edible. Seeds thick, often convex, nearly or quite marginless : albumen ruminated. SARCOYTTCCA. Fruit coreless, with purple pulp. Ovary stalked. Leaves with sharply denticulate horny border. Y. aloifolia. Fruit with a papery core and greenish or yellowish-white pulp. Ovary sessile. Leaves very minutely denticulate, not filiferous, flat or plicate. T. elephantipes. Leaves soon more or less flliferous, concave. Margin at first slightly denticulate. Leaves thick and firm, scabrid Y. Treculeana. Not denticulate. 4(5 MISSOURI BOTANICAL GARDEN. Thin, flexible: threads sparing, fine. Y. Schottii. Thick, rigid, with usually coarse threads. Leaves narrow, smooth. Small tree. Y. brevifolia. Leaves relatively broader, usually smooth. Large trees. Panicle narrow, pendent. Y. australis. Panicle broad, erect, to recurved Y. valida. Leaves large, very coarsely filiferous, the back very scabrous except in the last. Acaulescent. Flowers very large for the genus: style elongated. 1". baccata. Arborescent. Flowerg of average size. Style elongated. Y. macrocarpa. Style short. Y. Mohavensis. SYNOPSIS OF SPECIES. A. Fruit erect, capsular, dehiscent. Seeds thin, flat, margined: albu- men not ruminated. Chaenoyucca. 1. Leaves finely flliferous at the margin (entire in aberrant garden forms of the second). 2. Style oblong, white, 3. Inflorescence a long-peduncled panicle (reduced to a simple ra- ceme in aberrant forms or secondary inflorescences of the second). 4. Leaves lanceolate or spatulate, often plicate, not conspicuously white-margined. Y. FILAMENTOSA Linnaeus, Sp. PL 319. (1753). Walter, Fl. Carol. 124. Michaux, Fl. 1:196. Pursh, Fl. 1 : 227. Gawler, Bot. Mag. 23. pL 900. Redout^, Liliacees. 5. pi. 277-8. Haworth, Syn. PI. Succ. 70. Gambold, Amer. Journ. Sci. 1819 : 251. Mordaunt, Herb. Gen. 4. pi. 258. Elliott, Bot. S. C. and Ga. 1 : 400. Frost, Plants Abbeville Distr. 317. Lemaire, 111. Hort. 13 : 98. Porcher, Resources So. Fields and For. 530. Curtis, Bot. N. C. 56. Engelmann, Trans. Acad. St. Louis. 3 : 52, 214. Baker, Gard. Chron. 187O: 923. Journ. Linn. Soc. Bot. 18: 227. Britton & Brown, 111. Fl. 1 : 427. /. 1027. Holler's Deutsche THE YUCCEAE. 47 Gartner-Zeit. 11: 361. /. Mohr, Contr. U. S. Natl. Herb. 6: 441, as to southern localities. Tuca foliis fllamentosis. Morison, Plant. Hist. 2: 419. Sect. 4. pi. 23. (1680). Juca Americana fllamentosa. Hunting, Waare Oeftening der Planten. 471. /. (1682). Naauwkeurige Beschryv. der Aardgew. 663. (1696). Yucca Virginiana, foliis per ambitum apprime fllatis. Plukenet, Almag. Bot. 396. (1696). Raius, Hist. Plant. 8: 573. (1704). Yucca foliis lanceolatis etc. Trew. PI. Sel. 9. pi. 37. (1754). Yucca foliis lanceolatis acuminatis integerrimis margine fllamentosis. Gronovius, Fl. Virgin. 152. (1739). 53. (1762). Acaulescent, cespitosely suckering. Leaves rather firm, generally stiffly erect or spreading, about half a meter long, usually something over 25 mm. wide, narrowed above the base, attenuate or typically abruptly acute, occasionally somewhat pungent, green or a little glaucous, the back frequently roughened in lines; marginal threads rather thick and curly for the group. Inflorescence 1.5 to 3 or 4 m. high, long- pedunculate, glabrous or very exceptionally puberulent. Flowers white, usually tinged with cream color or green or rarely browned, expanding 50 to 75 mm. ; style white, elongated, at most slightly swollen, 3-grooved. Capsule apple-green and with regularly convex carpels when maturing, 50 or 60 mm. long and brown when ripe: seeds glossy, 4 to 5X 7 mm. Plates 8-12. 79. 87. In a generalized sense, a species usually of the coastal plain of the southeastern Atlantic region, from Tampa, Fla., to above Charleston, S. C., and extending back to northwestern Georgia, west-central North Carolina, southwestern Alabama, and the gulf coast of Missis- sippi. Plate 87, f. 1. The principal forms appear separable as follows : Leaves of medium size, little recurved. Y. filamentosa. Variegated with white or yellow. f. variegata. Outer leaves attenuate, recurved, the inner very broadly lanceolate, erect. var. media. Leaves narrow, very spreading. var. patens. Leaves very long, attenuate, recurving. var. bracteata. Leaves very broadly spatulate, not recurved. var. concava. Y. FILAMENTOSA Linnaeus. Synonymy as above. 48 MISSOURI BOTANICAL GARDEN. Leaves 25 to 40 mm. wide, gradually acute, rather rigid, striate, the outer rarely recurving. Petals broad, acute. Capsules rather narrowly cylindrical. Plates 8.12,f.l. West-central North Carolina to southeastern South Caro- lina, Florida from Jacksonville to Tampa, and doubtless in the intervening country. Plate 85, f. 1. Y. FILAMENTOSA VARIEGATA Carrierc, Rev. Hort. I860: 215. Naudin, Plantes Feuill. Colors'. 1. pi. 51. Lowe, Beautiful Lvd. Plants. 105. pi. 51. Garden. 1:152. /. 27 t :266,309. 32:600. Gardeners' Chron. n. s. 7 : 341. * n. s. 13 : 594. n. s. 23 : 803. ? T. ftlamentosa aurea elegantissima Wiener 111. Gart.-Zeit. 5 : 389. (1880). T. filamentosa bicolor Hort. T. recuroifolia Park & Cemetery. 11 : 184. /. Leaves margined and striped with various shades of white and yellow. A garden sport, or series of sports, the color extremes of which should doubtless bear distinctive horticultural names. Y. FILAMENTOSA PATENS Carriere, Rev. Hort. I860 : 216. Y.filamentosa Mohr, Contr. U. S. Natl. Herb. 6: 441, in part. Leaves rather rigidly spreading, 15 to 20 mm. wide, gradually attenu- ate to a sharp point. From northwestern to southeastern Georgia. Plate 85, f- 2. Y. FILAMENTOSA BRACTEATA Engelmann, Trans. Acad. St. Louis. 3:52-3. (1873). Watson, Proc. Amer. Acad. 14:254. Baker, Journ. Linn. Soc. Bot. 18 : 228. ? Y. filamentosa maxima Carriere, Kev. Hort. 1860:213. Very large, with elongated leaves, the outer recurved, mostly large foliaceous scape bracts, more frequently puberulent panicle sometimes nearly 5 m. high, and more attenuate petals. Capsule narrowly oblong, mucronate-beaked. Plate 9. About Charleston, S. C., and doubtless along the adja- cent Georgia coast, where it is sometimes seen in cultivation. THE YUCCEAE. 49 Simulating in aspect or bract characters cultivated forms of Y. flaccida. Plate 86, f. 1. Y. FILAMENTOSA coNCAVA (Haworth) Baker, Journ. Linn. Soc. Bot. 18:228. (1880). Y. concava Haworth, Suppl. PI. Succ. 34. (1819). Lemaire, 111. Hort. 13:98. Y. fllamentosa latifolia Engelmann, Trans. Acad. St. Louis. 8 : 52. ( 1873). General characters of the type, into which it appears to pass, but the usually very plicate abruptly acute or obtuse leaves deeply concave and spatulately enlarged to a width of as much as 100 mm. Plates 10. 79, f.l. About Charleston, S. C., to below Savannah, Ga., at Salisbury, Md., and doubtless in much of the intervening coast region. Plate 86, f. 2. Y. FILAMENTOSA MEDIA Carriere, Rev. Hort. 1860:213. /. 47-8. Y. fllamentosa laevigata Engelmann, Trans. Acad. St. Louis. 8 : 52, 54, 214. (1873). Watson, Proc. Amer. Acad. 14 1 254. Baker, Journ. Linn. Soc. Bot. 18 : 228. Y. fllamentosa Journ. of Hort. 52 : 271. /. ? Y. flaccida Lindley, Bot. Reg. 22. pi. 1895. Baker, Kef. Bot. 5 . text to pi. 323. ? Y. puberula Baker, Ref. Bot. 5 .pi. 322, not text. ? Y. glauca Baker, Ref. Bot. 5 .pi. 315. Leaves rather thinner, the outer gradually more attenuate and re- curved, the inner broadly lanceolate ; the marginal threads straighter. Inflorescence mostly puberulent and sometimes tomentose. Plate 11. A garden form, passing towards Y. flaccida glaucescens and Y. Louisianensis. Y. FLACCIDA Haworth, Suppl. PL Succ. 34. (1819). Lemaire, 111. Hort. 13: 99. Baker, Gard. Chron. 1870: 923. Ref. Bot. 5. pi. 323. Y. puberula Haworth, Phil. Mag. 1828 : 126. Sweet, Brit. Fl. Gard. pi. 21. Lemaire, 111. Hort. 18 1 99. Baker, Gard. Chron. 1870: 923. Y. filamentosa flaccida Engelmann, Trans. Acad. St. Louis. 3 ; 52, 214. 5() MISSOURI BOTANICAL, GARDEN. (1873). _ Watson, Proc. Amer. Acad. 14 : 254. Baker, Journ. Linn.. Soc. Bot. 18 : 228. Garden. 58 : 447. /. 1'. filamentosa puberula Baker, Journ. Linn. Soc. Bot. 18:228. (1880). Y.fllamentosa Gattinger, Tenn. Flora. (1887). 58. (1901). 86. Mohr, Contr. U. S. Natl. Herb. 6 : 441, as to northern localities. Garden. 68 : 445. /. Park and Cemetery. 11 : 184. /. r. Meldensis Garden. 8 : 147. (1875). Acaulescent, cespitose. Leaves thin, flexible, the outer almost always recurved, 10 to 40 mm. wide, elongated lanceolate, very gradually long attenuate, mostly plicate, with fine long and rather straight thin marginal fibers except in two threadless garden forms. Panicle mostly pubescent. Maturing capsule dull grayish-green, the carpels variously and irregularly flattened in places, as if shaved off with a knife ; when ripe, broad, usually constricted, and mostly flaring above : seeds rather dull, larger, 7 to 8 X 8 to 10 mm. Plates 12-1 7. 76. 79. Asheville, N. C., to Gadsden and Anniston, Ala., in and near the mountains. Plate 87, f. 2. Occasional simple racemes are produced from small lateral crowns, when the main crown is in bloom (Plate 13), as has been observed on some species of Agave, and one depauperate garden form produces an unbranched main inflorescence. An interesting winter adaptation of the foliage of this spe- cies is readily observed in the North whenever the tempera- ture remains for any time below the freezing point, for at and below this temperature the spreading unflexed middle leaves, which are ordinarily somewhat concave, have their margins rolled inwards so as nearly or quite to meet at the center, though they scarcely become involute in the proper meaning of that word. (Plate 14). The numerous intergrading garden forms of Y. flaccida seem capable of most natural arrangement as follows : Petals broad, acute or acuminate. Panicle mostly pubescent. T. flaccida. Inflorescence a raceme. f. orchioides. Petals usually more lanceolate, attenuate. Leaves flliferous. Panicle very pubescent. var. glaucescens. Leaves transiently variegated. f . lineata. THE YUCCEAE. 51 Panicle mostly glabrous. var. grandiflora. Leaves without marginal threads. Panicle pubescent. f. exigua. Panicle glabrous; petals blunter. f. Integra. Y. FLACCIDA Haworth. Synonymy as above. Leaves rather green, scarcely 25 mm. wide, very flexible. Panicle moderately pubescent to glabrous. Petals usually broad and rather short. Plate 16. The commoner wild form. Y. Meldensis of gardens appears to differ only in having more spreading panicle branches, in which it agrees with some garden forms of Y. filamentosa. Y. flaccida orchioides (Carriere) Trelease. T. orchioides Carriere, Eev. Hort. 1861 : 370. /. 89, 90. Lemaire, 111. Hort. 13 : 99. Baker, Gard. Chron. 1870 : 1122. Engelmann, Trans. Acad. St. Louis. 3 : 43. A depauperate garden form with stiffer more erect nearly threadless leaves, and racemose inflorescence. Y. flaccida glaucescens (Haworth) Trelease. Y. glaucescens Haworth, Suppl. PL Succ. 34. (1819). Sweet, Brit. Fl. Gard. pi. 53. Bommer, Journ. d'Hort. Prat. 1859:41. Lemaire, 111. Hort. 13:98. Baker, Gard. Chron. 1870:923. Hemsley, Garden. 8 : 132. Y. filamentosa glaucescens Baker, Journ. Linn. Soc. Bot. 18 : 228. (1880). Y. filamentosa Antwerpensis Baker. Z. c. Y. orchioides major Baker, Bot. Mag. iii. 33. pi. 6316. (1877). Y. flaccida Carriere, Rev. Hort. 1859 : 555. /. 11 9, 120. Y. filamentosa Baker, Ref . Bot. 6. pi. 324. Rept. Mo. Bot. Gard. 3. pi. 10. Amer. Florist. 8:55. /. A more glaucous form, with the leaves mostly broader and erect until a later period, almost tomentose panicle, and more attenuate petals. Plates 12, f. 2. 13-15. 17, f. 1. 76, f. 2. 79, f. 2. The common form of American gardens. 52 MISSOURI BOTANICAL, GARDEN. Y. flaccida lineata Trelease. A garden sport, apparently of var. glaucescens, but in habit more resem- bling T. filamentosa media, having the young leaves striped with dingy or yellowish white, the variegation soon fading for the most part. Cultivated at the Missouri Botanical Garden and said to have come from Haage & Schmidt in 1881. Doubtless it is this by which the variegated form of Y. filamentosa proper is represented in many gardens. Y. flaccida exigua (Baker) Trelease. Y. exigua Baker, Ref. Bot. 5. pi- 314. (1872). Journ. Linn. Soc. Bot. 18 : 223. Engelmann, Trans. Acad. St. Louis. 3 : 43. A garden form of var. glaucescens with the leaves without marginal threads. Y. flaccida grandiflora (Baker) Trelease. Y. filamentosa grandiflora Baker, Ref . Bot. 5. pi. 325. (1872) Y. filamentosa maxima Baker, Journ. Linn. Soc. Bot. 18:227. (1880). Y. filamentosa Garden. 1:152./. 12 1 72. /. Gartenflora. 24:372. f. Wiener 111. Gart.-Zeit. 13: 119. /. Step, Favourite Flowers. 4. pi. 272. Scarcely more than a large sometimes glabrous form of var. glauces- cens, in aspect resembling Y. filamentosa bracteata. Y. flaccida Integra Trelease. Y. glauca Sims, Bot. Mag. 53. pi. 2662. (1826). Regel, Garten- flora. 8 : 36. Bommer, Journ. d'Hort. Prat. 1859 : 43. Lemaire, 111. Hort. 13 : 97. Engelmann, Trans. Acad. St. Louis. 3 : 43, 53. Baker, Gard. Chron. 1870 : 1122. Journ. Linn. Soc. Bot. 18 : 223. Scarcely more than a narrow-leaved glabrous form of f . exigua. The name employed by Sims is antedated thirteen years by Y. glauca Nutt. The filiferous-leaved " bear grasses " of the southeastern Atlantic States are not easily disposed of in an attempt to monograph the genus to which they belong, partly because they are more commonly seen in cultivation than in a state of nature, partly because of their interblending characters, THE YUCCEAE. 53 and partly because of generalized earlier descriptions. One of the representatives of this group (probably true Y. fila- mentosa) was introduced into Europe about 1675, and Y. filamentosa was one of the four Yuccas known to Lin- naeus a century later, his description of it reading merely "foliis serrato-filamentosis," and the only figure cited by him * being very unsatisfactory. That two species, Y. filamentosa and Y. flaccida, are separable, appears certain, as is also true of Engelmann's conclusion f that the filamentosa of Linnaeus was th form to which that name is here applied ; but I have found it possible to fix only an approximate geographical range for either, and the garden forms are not separated as sharply as is desirable, nor so as to prevent some of them from obscuring the demarcation line between the species. It is not improbable that some of them represent hybrids between the latter. 44. Leaves linear or linear-spatulate, white-margined. Y. tciiuistyla Trelease. Acaulescent. Leaves rather soft and mostly recurving, often a little scabrid on the back, about .5 m. long and 10 to 15 mm. wide, dark green, lanceolate, long-attenuate, scarcely pungent, white-margined, finely flliferous. Inflorescence about 1 m. high, panicled at some distance above the leaves, glabrous or slightly puberulent. Flowers with narrower, more pointed segments : style oblong, white, often deeply parted. Capsule stout, even : seeds glossy, 7 to 8 X 8 to 10 mm. Plates 17,f. 2. 18. 19. 83>f.S. Southeastern Texas, from about Galveston (Lindheimer, May, 1843), to Sealy (Trelease, Harvey), and New Braunfels (Lindheimer, June, 1845), at the latter place associated with Y. Arkansana, which it closely resembles in foliage. Plate 92, f. 1. Some of the Lindheimer material in the Engelmann her- * Morison, Plant. Hist. 2 : 419. t Trans. Acad. St. Louis. 3 : 52. 54 MISSOURI BOTANICAL GARDEN. barium consists of loose flowers, some of which have a short thick green style, while others have the style longer, slenderer, and white; while the fragments of in- florescence are equally suggestive of mixed material, some of which was from racemes while the rest represent pan- icle branches. Field observation the present season, and material received from Mr. J. Eeverchon, of Dallas, and Mr. J. A. Harvey, of Sealy, confirm the conclusion reached, that the grass-leaved Yuccas of eastern Texas comprise three species, Y. ArJccmsana, Y. Louisianensis, and the one here characterized. Y. CONSTRICTA Buckley, Proc. Acad. Nat. Sci. Philadelphia. 1862:8. Gray, Proc. Acad. Phila. 1862:167.- Engelmann, Trans. Acad. St. Louis. 3:213. Bot. Gazette. 7 : 17. ? Y. alba-spica Koch, Belg. Hort. 12: 111. (1862). Rev. Hort. 1865 : 151. 48 : 432. ? Flore des Series. 17 : 110. f. 1612. Engelmaun, Trans. Acad. St. Louis. 3 : 213. Garden. 8 : 147. Y. angustifolia Carriere, Rev. Hort. 1860 : 20. /. 3, 4. 1864 : 151. Garden. 8: 134. /.Bray, Bot. Gaz. 32:280, in part*. F. glauca Bray, 1. c. 271. /. 18, in part*. Low or acaulescent. Leaves rather rigidly divergent, about 10 mm. wide, whitish green, the white margin soon shredding into fine threads. Inflorescence about 1.5 m. high, rather amply branched at top. Flowers white, globose-campanulate, with broad segments : style white, more or less tumid. Capsule constricted, flaring above, dark, with a ridge over each false septum: seeds 5 to 6X 7 to 9 mm. Plates 20. 21, f. 1. 83, f.4. Seward County, Kansas, to the Pecos river region of Texas. Plate 92, f. 2. Among other plants from western Texas which Mr. S. B. Buckley characterized about forty years ago was a Yucca * As is more clearly shown in a print from his negative, furnished me by Professor Bray, than in his published figure, the latter represents two species, Y. glauca, with simple racemes in full bloom, and Y. constricta, with branched pedunculate inflorescence still in bud. THE YUCCEAE. 55 which he called Y. constricta, and described as being shortly caulescent with leaves similar to but shorter than those of the Rocky Mountain species now called Y. glauca, long- stalked panicle, and capsules constricted in the middle. When Dr. Engelmann raised to specific rank the arborescent species that replaces this to the west, under the name Y. elata,* he was particular to exclude from it Y. constricta, which he regarded as a caulescent form of Y. glauca; but this conclusion, which did not accord with the description of fruit and inflorescence given by Buckley, was subsequently changed by himf and has not been followed by other writers, who have considered F. data and Y. constricta to be syn- onymous, t From observations made about Putnam, Texas, in 1892, and at various points west of San Antonio in 1900, 1 should say that Y. constricta is quite distinct from both the pre- ceding and the next species, differing from the former in its narrower and firmer leaves and more ample inflorescence, and from the latter in its usually very short stem, smaller constricted dark capsules, and much smaller seeds. Among a number of plants selected by Mr. James Gur- ney a few years since in Seward County, Kansas, for the demonstration of the great variability in the leaves of Y. glauca, is one which in foliage could hardly be dis- tinguished from the usual form of that species, or the somewhat broader-leaved variety by which the latter is represented in that part of Kansas, but which, on blooming in the Missouri Botanical Garden in 1900 produced a rather ample long-pedunculate panicle of pure white flow- ers, with white styles, which began to expand with the * Bot. Gazette. 7 : 17. (1882). f Trans. Acad. St. Louis. 3:213. % Baker, Journ. Linn. Soc. Bot. 18 : 229. Sargent, Silva. 10 : 27. Kept. Mo. Bot. Gard. 4 : 207, under T. glauca stricta (= Y. Ar- kansana). 56 MISSOURI BOTANICAL GARDEN. first flowers of Y. jlaccida, which they closely resemble, and at the end of the flowering period of Y. glauca and its va- riety stricta. It is hard to see how this plant can be separated from Y. constricta. What appears to be the same has been collected by Dr. Kleinschmidt at Mt. Kiowa, Okl., and the character of the intervening country is such as to make its extension probable from southwestern Kansas ta the Pecos river of Texas, while Professor Bray's photo- graph referred to above shows it to be a characteristic plant of the staked plains. Y. RADIOSA (Engelmann) Trelease, Rept. Mo. Bot. Gard. 3: 103. (1892). Y. angustifolia radiosa Engelmann, Bot. King. 496. (1871). Y. angustifolia data Engelmann, Trans. Acad. St. Louis. 3 : 50, 51. (1873). Watson, Proc. Amer. Acad. 14 : 253. F. elata Engelmann, Bot. Gaz. 7 : 17. (1882). Coulter, Contr. U. S. Natl. Herb. 2 : 437. Garden. 86: 573. Gard. & Forest. 2: 568. /. 146. 9 : 313. Rept. Mo. Bot. Gard. 3 : 164. pi. 9. 4 : 201. pi- 10,15, 22. Bot. Mag. iii. 55. pi. 7650. Y. constricta Baker, Journ. Linn. Soc. Bot. 18 : 229. Sargent, Silva. 10:27. pi. 504. In part. Y. angustifolia Havard, Proc. U. S. Natl. Mus. 8 : 470. Caulescent, the larger trees reaching a height of 5 to 7 m., simple or with a few short branches at top. Leaves pallid, rather rigidly diver- gent, long, 3 to 10 or rarely 13 mm. wide, white-margined and soon finely and copiously filiferous. Inflorescence large, panicled on a long ex- serted peduncle, glabrous. Flowers white, bell-shaped, with lanceolate attenuate segments : style white, oblong. Capsule oblong, smooth, not or rarely constricted, with ribless convex valves, straw-colored: seeds rather dull, 8 to 10X 12 to 15 mm. Plates 21, f. 2. 22. 83, f. 5. S6,f.l. Southern Arizona to the Rio Grande, as far as the big bend, and south to about the city of Chihuahua. Plate 93, f. 1. In describing the Yuccas for Watson's Botany of the Fortieth Parallel, Dr. Engelmann characterized an arbores- cent plant with large panicles and lanceolate petals under THE YUCCEAE. 57 the name Y. angustifolia ft. radiosa, which varietal name, two years later, he replaced by the varietal name elata which was still later applied specifically by him. With Mr. Baker, and against the opinion of Engelmann, Professor Sargent identifies this plant with the earlier Y. constricta of Buckley and applies the latter name to it. As has been stated above, however, there is reason to believe that Y. constricta is really a distinct species of more eastern and northern range, and to the present one the name radiosa, first used varietally by Engelmann, is applicable as a specific name. As in Y. glauca, the fruit of this species is stout, oblong, and unusually symmetrical among the capsular species, and it is here very smooth and of a clear straw-color at matur- ity, and the seeds are exceptionally large. The leaves, which are usually about 6 mm. wide, occasionally reach a minimum of 3 mm. and a maximum of about 12 mm., but both the broad- and narrow-leaved trees occur associated with the usual form, from which they do not appear other- wise distinguishable. So far as can be told from young leaves from Mr. Baker, in the Engelmann herbarium, Y. polyphylla Baker,* which its author subsequently f treated as a synonym of Y. radiosa, under the name Y. constricta, is more likely to have been based on an immature and aberrant garden seedling of Y. fiUfera than one of the representa- tives of this group, since the leaf possesses a distinct brown margin, very different from the white margin of Y. radiosa and its allies, which at most very exceptionally has a narrow brown line between the white border and the green body of the leaf. Though Y. alba-spica (or albospica as it is commonly written) seems to refer to the * Gard. Chron. 1870 : 1088. t Journ. Linn. Soc. Bot. 18 : 229. 58 MISSOURI BOTANICAL GARDEN. preceding rather than the present species, the latter is doubtless now in cultivation under that name. * For some reason this very striking Yucca does not ap- pear to have been collected or commented on by the bota- nists of the original boundary survey, though it is abundant in the Rio Grande valley about Presidio. The botanists of the later survey seem to have passed in by for Y. glauca, which I have not seen from so far south. 33. Inflorescence racemose or branched close to the leaves. Sub- acaulescent plants. Y. angustissima Engelmann, in herb. Y. glauca Coville, Contr. U. S. Natl. Herb. 4 : 202. Y. radiosa Coville, 1. c. 203, 277. Y. elata ? Merriam, N. A. Fauna. 7: 358. Acaulescent, from thick horizontal root-stocks. Leaves as in the narrowest forms of Y. radiosa and Y. glauca, 2 to 5 mm. wide, .2 to .4 m. long, pungent, white-bordered, very freely and often curly-filiferous below. Inflorescence glabrous, 1 to 1.5 m. high, racemose, or short- branched below. Perianth segments rather short, mostly acutely lan- ceolate : style as in the preceding. Capsule scarcely exceeding 50 mm. in length, rough, brown, constricted, with a median rib on each valve : seeds glossy, 5 to 7 X 7 to 8 mm. Plates 23, f.l.24,f.l. 83, f. 6. Southwestern Utah, southeastern Nevada, and north- western Arizona, in the region of the Colorado river. Plate 93, f. 1. In habit, this species, which is briefly referred to without name by Professor Sargent, f recalls the narrow-leaved form of Y. glauca as found, for example, about Albuquerque, N. M., or the narrowest-leaved forms of Y. radiosa, when the latter is acaulescent. From the former it differs in its more frequently branched inflorescence, oblong (white ?) style, and smaller capsule and seed ; and from the latter in never becoming a tree and in its subsimple inflorescence, smaller, rougher and darker, constricted capsules, and muck * See Baker, Kew. Bull. 1892 : 8. t Sargent, Silva. 10 : 28. Note. THE YUCCEAE. 59 smaller seeds. Specimens examined: "Deserts of the Colorado river" (Bigelow in 1853 and 1854); Grand canon region, Ariz. (Tourney in 1892, Trelease in 1901); '* Arizona" (Palmer, 799); "Southern Utah, northern Arizona, &c." (Palmer in 1877); St. George, Utah (Palmer in 1870) ; and La Verken, Utah (Jones, 5180). Y. Harrimaniae Trelease. Acaulescent, often cespitose. Leaves linear to spatulate-lanceolate, usually 6 to 15, or even 40 mm. wide, thin but firm, rigidly spreading, glaucous, or green with age, concave, pungent, narrowly brown-bor- dered, with relatively coarse, at length circinate, white marginal fibers. Inflorescence .25 to .5 m. high, simple, flowering from close to the base, glabrous. Flowers greenish, large, with broad often obtuse segments : style slender. Capsule brown, broadly oblong, about 40 mm. long, constricted, flaring above, the valves sometimes attenuate-mucronate : seeds 4 to 5X 5 to 6 mm. Plates 28. 29. 83, f. 10. Utah: Cedar City (Parry, July 6, 1874), Near King- ston (Jones, 5322), Helper (Trelease in 1899 and 1901), to western Colorado: Cimmaron (Baker, 281), on gravelly hillsides. Plate 93, f. 1. A very distinct species, often flowering when the leaf- rosette is not over a span wide, the broadly spatulate foliage of these small plants being strikingly unlike that of any other mature Yucca. My first acquaintance in the field with this plant resting upon the detention of our train at Helper, Utah, because of a washout, on the return of the Harriman Alaska Expedition, I take pleas- ure in dedicating it to our hostess on that occasion, Mrs. Edward H. Harriman. 22. Style stout, green. 3. Inflorescence racemose or branched close to the leaves. Y. GLAUCA Nuttall, Eraser's Cat. no. 89. (1813). Pit- tonia. 2: 115. Coulter, Contr. U. S. Natl. Herb. 2:437. Trelease, Kept. Mo. Bot. Gard. 4:205. 6. pi. facing p. 7. Schimper, Pflanzengeographie. (JO MISSOURI BOTANICAL GARDEN. 677. /. 384. Bush, Kept. Mo. Bot. Gard. 6: 122, 133. _ Britton & Brown, 111. Fl. 1 : 427. /. 1026. - Bray (in part*), Bot. Gaz. 32 : 271. /. 18. Y. angustifoliaPursh, Flora. 1 : 227. (1814). Nuttall, Gen. 1 S 218. Sims, Bot. Mag. 48. pi- 2236. Bommer, Journ. d'Hort. Prat. 3. 41. _ Lemaire, 111. Hort. 13 : 99. Baker, Gard. Chron. 1870: 923. Journ. Linn. Soc. Bot. 18 : 226. Engelmann, Bot. King. 496. Trans. Acad. St. Louis. 3 : 50. Palmer, Amer. Journ. Pharm. 50 : 587. Watson, Proc. Amer. Acad. 14 : 253. Gard. & Forest. 2 : 244, 247. /. Garden. 58 : 446. Kept. Mo. Bot. Gard. 3 : 163. pi. 8) 51. Wiener 111. Gart.-Zeit. 12 : 35. Bray (in part*), Bot. Gaz. 32 : 280. ? Y. Hanburii Baker, Kew. Bull. 1892 : 8, 217. Gard. Chron. iii. 11 : 749. _ Wiener 111. Gart.-Zeit. 17 : 433. Subacaulescent or with branching prostrate stem. Leaves rather rigidly divergent, 6 to 12 mm. wide, pallid, white-margined, soon finely but usually sparingly flliferous. Inflorescence 1 to 2 m. high, simple or with an occasional short included branch, floriferous from near the base, glabrous. Flowers greenish-white, globose or oblong, campanulate 3 the segments varying from broad and acute to longer and more attenuate ; style green, tumid. Capsule large, oblong, usually not constricted, somewhat roughened, brown: seeds very glossy, 7 to 9 X H to 13 mm. Plates 23, f. 2. 24, f. 2. 25. 83, f. 9. Central South Dakota and southern Wyoming, to north- west Missouri, Central Kansas and the vicinity of Santa Fe, New Mexico. Plate 93, f. 1. The usual form from Trinidad southward is prevailingly narrower-leaved than that of the north and east. This low capsular bear-grass or soap-weed of the central Rocky Mountain region and northern plains, is almost in- variably marked by a simple inflorescence, not carried on a scape above the cluster of leaves. Only exceptionally are any branches formed on the panicle, and then these, which are toward its base, are very small and few in num- ber, though when the developing inflorescence has been injured a greater development of these potential rudiment- ary basal branches is observed. * See note under Y. constricta above. THE YUCCEAE. 61 European gardens contain, under the name Y. angusti- folia, plants which are very different from the Yucca so- called by Pursh. In 1860, Carriere,* giving Y. albo-spica as a synonym, described and figured one such plant, with long-exserted glabrous panicle and rather broad filiferous leaves, which, with Mr. Baker, f I should more readily refer to Y. constricta than elsewhere, and Mr. Baker t states that Y. flexilis also occurs in gardens under this name. From the original description, Y. Hanburii possesses quite the inflorescence of Y. glauca; but has the leaves a little rough on the back and with a line of brown between the green tissue and the marginal line of white. I should have thought of connecting with it the narrower leaves of the preceding species, because of these characters, had not the Kew authorities given me positive assurance that the two are very distinct. Y. glauca stricta (Sims) Trelease. r. stricta Sims, Bot. Mag. 48. pi. 2222. (1821). Bommer, Journ. d'Hort. Prat. 3:41. Lemaire, 111. Hort. 18 : 95. Baker, Gard. Chron. 1870 : 923. Hemsley, Garden. 8 : 130, 132. /. As to Sims citation only. r. angustifolia stricta Baker, Journ. Linn. Soc. Bot. 18 : 227. (1880). As to Sims citation only. Of the habit of the northern form of Y. glauca, but of more vigorous growth, and with longer, more erect stem. Leaves very long, 12 mm. or less wide, at first somewhat glaucous, the entire white margin quickly shredding into slender fibers. Inflorescence usually tall, occasionally simple but typically paniculately branched within or close to the cluster of leaves. Flowers greenish white, often purple-tinted, varying from glo- bose to oblong-campanulate, and with correspondingly short and blunt or acutely attenuate perianth segments: style greatly swollen at base, green. Capsule and seeds unknown. Plates 26. 27. Seward County, Kansas, and doubtless elsewhere on the plains. In 1821, Dr. Sims applied the name Yucca stricta to a Rev. Horticole. 1860 1 20-22. /. 3-4. Journ. Linn. Soc. Bot. 18 : 229. 1. c. 224. 62 MISSOURI BOTANICAL GARDEN. filiferous-leaved plant, said to have been introduced a few years before from the Carolinas, by Mr. Lyon, and to have been confused, up to the time of its description, with Y. angustifolia* (for which the prior name Y. glauca is now commonly employed). The good illustration that he gives, and which is copied by Hemsley, shows, as the description indicates, that the plant is quite of the habit of Y. glauca , with similar narrow leaves and violet-tinged greenish flowers having the swollen green stigmas of Y. glauca; but the panicle is much branched below, the rather long branches reaching about to the top of the uppermost leaves, and the flowers are subglobose, with broad blunt perianth segments, in neither of the latter respects, however, differing from some specimens of Y. glauca. Yucca stricta, ever since its establishment, has been a puzzle to botanists, partly because no plant exactly cor- responding with Sims' figure seems to have been reported since then, and partly because M. Carriere,f and following him, Mr. Baker, % confused with it a garden plant, which, in fact, appears to be Y. Louisianensis. In his article in The Garden, Mr. Hemsley copies the original illustra- tions of both forms, though treating them as pertaining to one species. Both Baker and Hemsley mention her- barium specimens collected by Drummond in Texas and near New Orleans, as representing their Yucca stricta , which Mr. Baker subsequently called Y. angustifolia var. Y. stricta || and which cannot well be the stricta of Sims or of Carriere, but is what is here called Y. Arkansana or Y. tenuistyla, or both. It is interesting to note that although much collecting has been done in the South Atlantic region since the time of Sims' publication of Yucca stricta, no green-styled species of the alliance of * On this see Nuttall, Genera 1 : 218. (1818). t Rev. Horticole. 1859 : 466-470. /. 101-2. J Gard. Chron. 1870 : 923. Garden. 8 : 130, 132, 140. (1875). || Journ. Linn. Soc. Bot. 18:227. (1880). THE YUCCEAE. 63- the Kocky Mountain Y. glauca has been found in that region, the nearest approach being the Gulf plant here called Y. Louisianensis. A few years since, Mr. James Gurney, Head Gardener of the Missouri Botanical Garden, was struck with the variety of foliage and difference in vigor of growth shown by the soap plants of Seward County, in extreme southwestern Kansas, and he selected for the Garden and for Tower Grove Park a considerable number of plants to show the differences. Some of these plants, which have made a remarkably rapid growth, have now come into bloom. They differ considerably both as to their tendency to form a short trunk and in breadth and flexibility of foliage, though in this latter respect coming within the known range of variation of Y. glauca , and to an equal extent in inflorescence, the variation in the two characters, however, not appearing capable of connection. While some of the plants produce a simple inflorescence, indistin- guishable from that of Y. glauca , others almost exactly match the original figure of Y. stricta, and still others, with the same compound inflorescence, have the branches originating at about the top of the leaves instead of in the leaf -cluster. There seems to be little doubt that these plants represent the true stricta of Sims, and that the At- lantic States locality assigned to this when it was published rests upon some sort of error. Although, as has beer* said, the cultivated plants produce either simple or branched inflorescence, the prevalence of the latter in those which- are strongly developed, and the rareness of branching in the usual form of Y. glauca, make it desirable to recog- nize this form varietally. Y. Arkansana Trelease. f. angustifolia mollis Engelmann, Trans. Acad. St. Louis. 3 : 50, 51. (1873). Watson, Proc. Arner. Acad. 14:253. T. glauca mollis Branner & Coville, Ann. Kept. Geol. Surv. Arkansas- for 1888. 4 : 224. 64 MISSOURI BOTANICAL GARDEN. Y. stricta Baker, Gard. Chron. 1870 : 923. Hemsley, Garden, g . 132. As to herbarium citations, in part. Y. angustifolia stricta Baker, Journ. Linn. Soc. Bot. 18:227. As to herbarium citations, in part. T. glauca stricta Trelease, Kept. Mo. Bot. Gard. 4 : 206. pi. 22. Coulter, Contr. U. S. Natl. Herb. 2 : 437. Y. recuroifolia ? Nutt. Trans. Amer. Philos. Soc. 5: 156. Aspect and foliage of Y. tenuistyla. Inflorescence about 1 m. high, racemose or very rarely with a few branches, glabrous. Flowers with mostly greenish-white broad and obtuse segments : style green, usually very tumid below. Capsule little flaring, smooth: seeds dull, 7 to 8 X 10 mm - Plates 30.31. 83, f. 7. From about Catoosa, I. T. (Bush, 1278) and Little Rock, Ark. (Engelmann, May 1837) to the vicinity of San Antonio, Tex. Plate 88, f. 2. The specific name Arkansana, here used, is applied in deference to the prevalent American practice in nomencla- ture, Engelmann's varietal name mollis (1873) having been similarly used under Y. gloriosa by Carriere, in 1860. 33. Inflorescence amply panicled on a long scape. Foliage of the preceding or wider. Y. Liouisianensis Trelease. Y. filamentosa Riddell, N. O. Med. & Surg. Journ. 8 :763. Baflnes- que,Fl. Ludovic. 18. Gray, Manual. [6 ed.]. 524. Britton, Man- ual. 269. As to the Louisiana citation. Y. stricta, Y. stricta elatior, and Y. stricta intermedia Carrifcre, Eev. Hort. 1859 : 390, 466. /. 101-2. Of the aspect of the preceding, or, when the inner leaves are dilated, of Y. filamentosa media. The flaccid green leaves 10 to exceptionally 40 mm. wide, white bordered sparingly flliferous. Inflorescence an ex- serted glabrous or mostly pubescent panicle. Petals broad to attenuate. Style variously tumid aiid deep green, to pale and oblong. Capsule stout and short, angular in developing, as in Y. flaccida: seeds 6 to 7 X 6 to 10 mm. Plates 32-34. 83, f. 8. Louisiana (Alexandria, Ball 558; Minden and Alden Bridge, Trelease) to northern Texas (Jefferson, Tretease; Dallas, Reverchon; Texarkana, Trelease) and southeastern Indian Territory (Atoka, Butler; Standley, Ferriss; Poteau, Trelease). Plate 92, f. 1. Apparently a western derivation of the same stock as the THE YUCCEAE. 65 ^eastern Y. filamentosa and Y. flaccida , to both of which it bears some relationship, while apparently distinct from either. At Dallas, where Mr. Reverchon has long culti- vated this and Y. rupicola , spontaneous hybrids occur, with the leaf -margin neither denticulate nor filiferous. 11. Leaves not flliferous, with a distinct thin horny, finely denticulate border. 2. Capsule mucronate, with flat-backed valves. Y. rigida (Engelmann) Trelease. r. rupicola rigida Engelmann, Trans. Acad. St. Louis. 8 : 49. (1873). Watson, Proc. Amer. Acad. 14 : 253. Baker, Journ. Linn. Soc. Bot. 18 : 223. Caulescent, reaching a height of 3 to 5 m., simple or elongately few- branched above. Leaves glaucous, thin but rather rigidly spreading, about 25 mm. wide, mostly concave, often with scabrid ridges, slender- tipped but very pungent, the yellow margin minutely denticulale. Inflo- rescence rather large, panicled close to the branches, glabrous. Flowers not very large. Capsule oblong, thick-walled, rough, not constricted, the flat valves tipped with short outcurved points : seeds very dull, 4 to 5 X 6 to 6 mm. Plates 35. 36, f. 1. 84,f.l. Mexico, from central Chihuahua to eastern Durango. Plate 93, f. 2. The Engelmann herbarium contains two specimens (nos. A. and 477) of a Yucca collected in 1847 by Dr. Gregg, in a dry valley between Mapimi and Guajuquilla, in northern Mexico, which he noted as from 5 to 10 feet high, and which possesses glaucous denticulate-margined rather narrow leaves which in the herbarium appear quite rigid. In revising the Yuccas, Dr. Engelmann, recognizing a certain comparability of these specimens with Y. rupicola, desig- nated them by the varietal name rigida, under that species, evidently mistaking Gregg's note on the height of the plants for that of the scape, instead of the trunk, which it really appears to have referred to. Within recent years, the same plant has been collected (and sometimes referred to this variety) by Wilkinson (134715, 224209), Rae and Hough (4220), and Pringle (165) in the Santa Eulalia mountains, near the city of Chihuahua. 5 (J6 MISSOURI BOTANICAL GARDEN. South of Torreon, along the Mexican Central railroad, particularly from about Picardias to about Jalisco, this small tree is abundant, on or near the rocky hillsides, and conspicuously contrasted with accompanying Y. Treculeana bv its very glaucous narrower foliage. It may be that small trees between Monterey and Saltillo, visible from the Mexi- can National railroad, extend its range to the east. Yucca rigida, the specific name of which is descriptive only when its dried leaves are compared with those of Y. i-upicola, is one of the handsomest tree Yuccas, in its foli- age. The slender trunks are commonly simple, but occas- ionally once or more forked, with elongate branches. When well developed the leaves are from .3 to .6 m. long, 20 to 30 mm. wide, and, as would scarcely be inferred from herba- rium material, decidedly concave up to the very slender pungent terete point ; both surfaces are closely ridged and often minutely roughened, and the bright yellow margin, though occasionally nearly smooth, is usually finely den- ticulate, so as to possess a keen cutting power. Though, as has been said, the plant forms a low tree when developed, a few specimens have been seen bearing panicles when still practically acaulescent, as is also true of Y. radiosa about El Paso. The panicles are loosely branched shortly above the crown of leaves, and the very hard oblong capsules, about 50 mm. long and 25 mm. in diameter, are parted about to the middle into 3 valves which are conspicuously flattened or even concave on the back, and with short out- curved apical points, and the inner or placental dehiscence is very narrow, so that the small thin black seeds escape only when jarred out edgewise. Dr. Engeimann would doubtless have given specific rank to this tree, had he not misapprehended its relation in size and field appearance to the typical acaulescent often twisted- leaved Y. rupicola, which, in contrast with it, he called variety tortifolia. The foliage and capsular characters added above leave no room for question as to its specific distinctness from the latter. THE YUCCEAE. 67 Y. X rigida Deleuil, described by M. Andre*,* is a garden hybrid obtained from Y. gloriosa fertilized by Y. cornuta (which is considered to be a synonym of Y. Treculeana), and, as the name rigida, being preoccupied, cannot be re- tained for it, it may be named, after its originator, Y. X Deleuili, in case, as seems desirable for convenience of reference, it and other hybrids are to be designated by binomials. Y. RUPICOLA Scheele, Linnaea. 23:143. (1850). Le- maire, 111. Hort. 13: 96. Baker, Gard. Chron. 1870: 828. Engelmann, Trans. Acad. St. Louis. 3: 48. Garden. 1: 161. Watson, Proc. Amer. Acad. 14 : 253. Baker, Journ. Linn. Soc. Bot. 18 : 222. Coulter, Contr. U. S. Natl. Herb. 2 : 436. Bot. Mag. iii. 47.pl. 7172. Reverchon, Gard. & Forest. 6: 64. Kept. Mo. Bot. Gard. 3: 163. pi. 51. Y. rupicola tortifolia Eugelmann, 1. c. Y. lutescens Carriere, Rev. Hort. 1858 : 579. Y. tortilis Hort. Y. contorta Hort. Acaulescent. Leaves glaucous, pungent, firm or flaccidly spreading, often twisted, .3 to .5 m. long, 25 to 30 mm. wide, the yellowish finely denticulate margin soon turning brown. Inflorescence glabrous, panicled mostly above the leaves. Flowers white or greenish: style white or greenish, oblong, often 3-sided. Capsule thin-walled, with flat or con- cave mucronate valves : seeds rather dull, 5 to 6 X 7 to 9 mm. Plates 37-39. 84, f. 2. South-central Texas, from Tarrant County southwest- ward to and probably across the boundary. Plate 93, f. 2. One of the early discoveries of Lindheimer (1845), and Tre'cul (1848-9), sufficiently distinct from all of its con- geners. Dr. Engelmann designated it as a. tortifolia, to distinguish it from his /3. rigida, spoken of above, with the statement that it is cultivated under the two garden names given in the synonymy. * Revue Horticole. 55: 110. (1883). 67: 81. (1895). (J8 MISSOURI BOTANICAL GARDEN. In speaking of Y. rupicola and what he called its variety rigida, Dr. Engelmann* refers to intermediate specimens collected by Wright in " Eastern New Mexico " (no. 1909). The leaves of this number in the Torrey herbarium (Plate 37), it is true, are very hard to distinguish from narrower herbarium leaves of Y. rigida, but the cor- responding sheet in the Gray herbarium (Plate 38} clearly represents a crown of the acaulescent Y. rupicola with inner leaves, narrower and less twisted than the outer leaves probably were. A similar intermediate specimen in the Engelmann herbarium, collected by Wright in April or May 1850, on " Hills of the Blanco " is from the region of and accompanied by unmistakable, though detached, leaves of Y. rupicola j to which I should refer all of these speci- mens. 22. Capsule attenuate-beaked, with round-backed valves. Y. rostrata Engelmann, in herb. Of the aspect of T. radiosa. Caulescent, at length 3 m. high, simple or short-branched at the crown. Leaves very numerous, rigidly diver- gent, scarcely 10 mm. wide, a little glaucous, flat or biconvex, striate, thin, very pungent, the yellow margin minutely denticulate. Inflores- cence ample, with subincluded base or mostly exserted, glabrous. Flowers white, umbonate at base : style white, attenuate. Capsule oblong-ovoid, thick-walled, with convex valves long-attenuate and spreading above: seeds rather dull, 4 to 5X 6 to 7 mm. Plates 36, f. 2. 40-42. 84, f. 3. Northern Mexico, from northern Chihuahua to the Sabinas valley in eastern Coahuila. Plate 93, f. 2. In 1852, Dr. Bigelow, of the boundary survey, collected a Yucca with narrow denticulate leaves, somewhat resem- bling ]T. rigida, at Bufatillo, said to be in a volcanic moun- tainous region near Presidio del Norte, and what may pos- sibly have been the same thing on sand hills thirty miles below San Elizario, both along the Rio Grande, and on gravelly hills at Los Moros. In August, 1880, Dr. Edward Palmer collected leaves, capsules, and seeds of ap- * Trans. Acad Sci. St. Louis. 3 : 50. THE YUCCEAE. 69 parently the same thing at Monclova, in the State of Coa- huila. To these latter, Dr. Engelmann attached the manu- script name Y. rostrata, descriptive of the long-attenuate apex of the fruit. While passing between Eagle Pass and Monterey, in company with Professor Sargent and Mr. Canby, in March 1900, my attention was attracted by a narrow-leaved Yucca that was cultivated at C. P. Diaz and in station yards along the Mexican International railroad, and that was found forming a natural low forest about Peyotes, on the water-shed between the Rio Grande and Sabinas, where, on subsequent visits, in April and August, I was able to study it in detail. Among Yuccas this is conspicuously loosely rooted in the soil, so that large plants are easily removed. The trunks vary in height from about .3 m. to an observed maximum of about 3m., the usual height being about 2m., and the wood is extremely soft and spongy. When the old leaves are removed, the diameter of the stem is usually .15 or .2 m., and it is not dilated except where the roots start from the base. Older plants are sometimes branched at the top, but the branches remain short, so that these trees usually possess several subapical crowns of leaves, rather than a series of separated elongated branches, like those of many other arborescent species. The leaves are very numerous, radiating in every direc- tion from the top of the stem in an oblong or usually nearly globose crown some 1.25 to 2 m. in diameter, and, although thin, they are sufficiently rigid rarely to become arched from their own weight, as they are in the species of N^olina, like JW. longi folia, with similar foliage. They are flattened or a little biconvex, quickly contracted from a broad base and then very narrowly lanceolate, measuring about 6 mm. at the nar- rowest point and 12 mm. at the widest, which is about one- third their length below the grooved, acute, pungent apex. They are somewhat glaucous, occasionally slightly twisted 70 MISSOURI BOTANICAL GARDEN. and striately veined, and with a very narrow bright yellow horny margin that bears numerous very minute teeth, like those of Y. rupicolaand Y. rigida. The old leaves, closely reflexed against the stem, persist for many years as a straw- colored thatch-like covering, and the denuded lower stem is lozenge-marked by the leaf-scars and does not develop a thick bark. The glabrous panicle ranges from .5m. long to more than twice that length, and is raised on a stalk 30 to 50mm. thick, which, though sometimes barely protruding from the leaves, is more commonly exserted for a length about equal to that of the branched part, and is sparingly bracteate, the narrow green lower bracts gradually passing into the dingy floral bracts. The common outline of the flower-cluster is attenuate-ovoid, but not infrequently the lower part of the cluster, like the top, is unbranched, the uppermost and lowest flowers then standing in the axils of the bracts of the main stem. The rather large waxen pendent white flowers, which are very rarely somewhat purple-tinged, expand from 50 to 75 mm. They are slightly umbonate at base, on short curved pedicels which rarely reach their own length. The segments of the perianth are lance-obovate, the inner whorl somewhat crenulate, and the outer narrower, thicker and subentire. The stamens, which are somewhat clavately thickened and spreading near the top, are coarsely papillate-pubescent, as in other species of the genus. The narrowly oblong conical ovary is green, and the attenuate white style con- siderably surpasses the stamens and ends in three slightly notched lobes. The erect or suberect very firm-walled capsule, measur- ing about 25 X 50 mm., is oblong-acuminate with the atten- uate upper third of the convex carpels somewhat spreading in dehiscence, and is raised on a concavely obconical base, corresponding to that noted for the flowers, from the top of which remnants of the withered perianth commonly de- THE YUCCEAE. 71 pend. The seeds are black, thin, margined, and rather small. Of somewhat the aspect of Y. radiosa, but with more rigid and denticulate not filiferous leaves, this species rivals in gracefulness of habit the Nolinas of Mexico and the grass-trees (Xanthorrhoea) of the South Sea, both of which it far surpasses in beauty of inflorescence, and it should prove a desirable addition to regions like California, Madeira and the Mediterranean countries, where it will prove hardy, and to some of the gardens of which I have been able to send viable seed. AA. Fruit indehiscent (so far as known). B. Fruit soon drying, erect, spreading or pendent. Seeds thin, flat, slightly margined: albumen not ruminated (but surface of seed often somewhat grooved). Heteroyucca. 1. Leaves finely denticulate, softly green-pointed. Large tree. Y. GIGANTEA Lemaire, 111. Hort. 6. Misc. 91. (Nov. 1859). 13:92. Rev. Hort. I860 : 222. Engelmann, Trans. Acad. St. Louis. 3 : 212. Baker, Gard. Chron. 187O: 1184. Journ. Linn. Soc. Bot. 18 : 224. Hemsley, Garden. 8: 134. Trelease, Kept. Mo. Bot. Gard. 9:141. pi. 40-42. At length a rough-barked branching tree 10 m. or more high. Leaves rigidly spreading or somewhat flexuous, green, glossy, plicate, with soft green tip, over 1 m. long and often 100 mm. wide, scabrid margined. In- florescence compact, close to the leaves. Flowers resembling those of Y. gloriosa. Fruit apparently soon drying. This species, if more than a form of Y. elephantipes, was first described from young specimens cultivated in European gardens, and again, in mature form, from a large tree cul- tivated in the Azores. It does not appear to be known in a state of nature. In habit and foliage, except for larger dimensions, it resembles Y. elephantipes, but if the notes on the spontaneous Azorean fruit are accurate, possesses fruit comparable with that of Y. gloriosa, and it may be a hybrid, Y. elephantipes being doubtless one parent, in this x;ase ; but it is very doubtful as anything but a form of Y. elephantipes. 72 MISSOURI BOTANICAL GARDEN. 11. Leaves 'at most sparingly denticulate or flliferous, pungent. Lower plants. 2. Leaves broad, rigidly ascending or spreading. Y/ GLORIOSA Linnaeus, Sp. PL 319. (1753). Walter, FL 'Carol. 124. Michaux, Fl. 1 :196. Duhamel, Arbres et Arbustes. 3. pi. 35. Bryant, Flora Diaetetica. .16. Pursh, Fl. 1:228. Elliott, Bot. S. C. & Ga. 1 : 400. Baker, Gard. Chron. 187O : 1184. Engel- 'mann, Trans. Acad. St. Louis. 3:38, 211, 213. Koch, Dendrologie. 2 2 :343. Carriere, Rev. Hort. 49:287. /. 48. Watson, Proc. Amer. Acad. J14 : 251. Baker, Journ. Linn. Soc. Bot. 18 : 225. .Sargent, Silva. 1O:23.^Z. 503. Gard. Chron. iii. '28 : 262. /. 77. Garden. 49 : 218. /. (Y. acuminata Sweet, Brit. Fl. Gard. 2. pi. 195. (1827). Bommer, \Journ. d'Hort. Prat. 1859 : 42. Lemaire, 111. Hort. 13:95. Baker, Gard. Chron. 1870:1123. Ref. Bot. 5. pi. 316. Engel- mann, Trans. Acad. St. Louis. 3 : 40. Garden. 8:133. Gard. Chron. n. s. 4:110. Y. gloriosa acuminata Carriere, Rev. Hort. 1868 : 157. Baker, Journ. Linn. Soc. Bot. 18 : 226. Y. integerrima Stokes, Bot. Mat. Med. 2 : 267. (1812). ~*Yuca, sive lucca Perana. Gerarde, Herball. 1359. /. (1597). Yuca foliis Aloes. Bauhin, Pinax. 91. (1623, 1671). Morison, Plant. Hist. 2:419. Sect. 4. pi. 23. (1680). Pontedera, Anthologia. 295. pi. 6.f. n. (1720). Yuca sive lucca. Parkinson, Paradisus Terrestris. 434. /. (1629). Yucca, sive lucca Peruana. Johnson in Gerarde, Herball. 1543. /. , (1636). Raius, Hist. Plant. 2: 1201. (1688). Juca gloriosa. Hunting,, Waare Oeff. der PI. 471. pi. (1682). Naauw- keur. Beschryv. der Aardgew. 663. (1696.) Yucca; foliis Aloes. Boerhaave, Index Alter PI. Hort. Lugd.-Bat. 2:132. (1720, 1727). Cordyline foliis pungentibus integerrimis. Van Royen, Fl. Leyd. Prod. 22. (1740). Yucca foliis margine integerrimis. Linnaeus, Hort. Cliff. 130. (1737) Hort. Ups. 88. (1748). Shortly caulescent and cespitose or the trunk 3 to 5 m. high and with several branches. Leaves slightly glaucous when young, smooth or the dorsal lines roughened, rather thin but rigid, often concave near the in- rolled purfgent usually dark apex, about .5 m. long and 50 mm. wide, the THE YUCCEAE. 73 usually brown margin at first with a very few distant rarely persistent minute teeth, when developed entire or occasionally with a few detach- ing slender fibers. Inflorescence mostly narrowly paniculate, the base often not exserted, glabrous or exceptionally puberulent. Flowers creamy white, often tinged with red or violet : ovary often with a slight suggestion of basal stipe; style oblong, white, frequently 3-divided. Fruit obovoid-oblong, mostly pendent, with six prominent ridges, the thin exocarp soon drying about the core : seeds glossy, 5 to 6 X 6 to 7 mm., slightly grooved as if the albumen were ruminated. Plates 43-46. 80, f. 4. Coast and " sea islands," from South Carolina to north- eastern Florida, on the sand dunes. Generally planted and in places escaping, in the eastern Gulf region. Plate 94, f-1- The typical form and what is called here variety plicata are the only spontaneous forms of this species of which I have knowledge. It has been in cultivation since 1596 (Gerarde, Herball, 1359. /.), and to-day is represented by a considerable number of garden forms, several of them hardy further North than any other species except Y. flac- cida, Y. filamentosa, and Y. glauca. Some of these approach the following two species while others, scarcely presenting mature characters, are but tentatively placed anywhere; and a number of imperfectly described gar- den hybrids add to the difficulty of properly understand- ing Y. gloriosa. The following key, including these hy- brids, may serve for the naming of the forms : Leaves not or little plicate, usually concave only toward the end. Leaves rigidly spreading. From slightly glaucous becoming green, A to .8 m. long, 40 to 50 mm. wide. Y. gloriosa. Dwarf and smaller-leaved. f . minor. More persistently glaucous. Somewhat falcate. f. obliqua. With whitish median variegation. f. medio-striata. Outer leaves somewhat recurving. Leaves but transiently glaucous. var. robusta. Persistently glaucous. f. nobilis. Leaves narrower. f . longifolia. 74 MISSOURI BOTANICAL GARDEN. Leaves conspicuously plicate toward the end, mostly very concave, not recurved. Rather persistently glaucous. var. plicata. Tall (1.5 to 3 m.) Leaves at last greener. f. superba. Leaves dark green, persistently denticulate. f. maculata. Leaves purplish. Y. X DeleuiU. Leaves greener, very broad. Y. X sulcata. Leaves olive-green, scarcely pungent. T. X Carrierei. Y. GLORIOSA Linnaeus. Synonymy as above. Acaulescent or not tall. Leaves broad, entire, green, neither recurved nor plicate, plane or very openly concave. Plates 43. 44. The most common form of the Sea Islands of South Carolina and Georgia. Y. GLORIOSA MINOR Carriere, Rev. Hort. 1860:361. Truffaut, Rev. Hort. 1869:474. Baker, Ref . Bot. 5. pi. 319. Journ. Linn. Soc. Bot. 18 : 225. ? T. acuminata Garden. 27 :266. /. T. rubra Hort. A garden form, smaller in every way. Plate 45. Y. GLORIOSA OBLIQUA (Haworth) Baker, Gard. Chron. 187O : 1184. Journ. Linn. Soc. Bot. 18 : 225. T. obliqua Haworth, Syn. PL Succ. 69. (1812). Lemaire, 111. Hort. 13:95. Engelmann, Trans. Acad. St. Louis. 3 : 40. Koch, Den- drol. 2 2 :345. A form with glaucous leaves somewhat twisted to one side. Y. GLORIOSA MEDIO-STRIATA Planchon, Fl. des Serres. 23. pi. 23 93-4. (1880). Gard. Chron. n. s. 13: 716. Belg. Hort. 31 : 36. Wiener 111. Gart.-Zeit. 6 : 156.' Y. gloriosa medio-picta Carriere, Rev. Hort. 1880 : 259. A garden sport with a median whitish stripe on the leaves. Y. GLORIOSA ROBUSTA Carriere, Rev. Hort. 1868. 158. ? T. acutifolia Truffaut, Rev. Hort. 1869 : 320. Belg. Hort. 1870 : 24. Y. gloriosa recurvata Baker, Gard. Chron. 1870: 1184. Y. gloriosa Gawler, Bot. Mag. 31. pi. 1260. Redoute, Liliacees. 6. pi. 326-7. THE TUCCEAE. 75 Intermediate between Y. gloriosa and T. recurmfolia, with the outer- most of the evanescently glaucous usually slightly plicate leaves somewhat stiffly recurved. Y. GLORIOSA NOBILIS Carriere, Rev. Hort. I860 : 360. 1868 : 157. T. Ellacombei Baker, Ref. Bot. 5. pi. 317. (1872). Engelmann, Trans. Acad. St. Louis. 3:41. Garden. 4:356. 8:134, 147. 16: 196, 214, 216, 236, 257, 285. Gard. Chron. iii. 2: 111. Y. gloriosa Ellacombei Baker, Journ. Linn. Soc. Bot. 18:226. (1880). Y. gloriosa Gardening 111. 22: 155. /. Leaves scarcely plicate, glaucous, the outer recurved, sometimes twisted to one side. An intermediate form, differing from f . robusta in its more persistently glaucous leaves. M. Carriere (Eev. Hort. I860: 361) recognizes a sub-variety parviflora of this variety. Y. GLORIOSA LONGIFOLIA Carriere, Rev. Hort. 1862 : 234. J. longifolia Hort. in part. r. glaucescens Eev. Hort. 1 : 266. 2 : 111. Baker, Kew Bull. 1892:8. Y. gloriosa glaucescens Carriere, Rev. Hort. 1860:360. Baker, Gard. Chron. 1870:1184. ? F. Brasiliensis Baker, Kew Bull. 1892 : 8. Scarcely differs from var. noUlis except in its leaves when young being narrower, though in age they are said to reach a width of 75 mm. Y. GLOEIOSA PLICATA Carriere, Re v. Hort. 1860:359. Engelmann, Trans. Acad. St. Louis. 3:39, 40. Baker, Journ. Linn. Soc. Bot. 18 : 225. Y. gloriosa Maund, Bot. Gard. 3. no. 2 8 6. Elliott, Bot. S. C. & Ga. 1 : 400. Lemaire, 111. Hort. 13 : 94. Garden. 81 : 16 1. /. 45 : 45. /. 49 : 332. /. Gard. Chron. n. s. 19 : 820. /. 157. iii. 8 : 692. /. 136. iii. 15:304. pi. Amer. Florist. 8:61. /. Rept. Mo. Bot. Gard. 3. pi. 6. Gardiner, Journ. of Hort. 62:487. f. 126. Y. plicata Hort. r. plicata glauca Hort. Y. plicatilis Hort. Y. glauca Hort., in part. Differs from the type in having the more permanently glaucous usually shorter and hence relatively broader concave leaves evidently plicate to- ward the apex. 76 MISSOURI BOTANICAL GARDEN. " Sea islands " of Georgia and South Carolina, with the type. Y. GLORIOSA SUPERBA (Haworth) Baker, Gard. Chron. 1870 : 1184. Journ. Linn. Soc. Bot. 18 : 225. Ella- combe, Garden. 8:147. T. superba Haworth, Suppl. 36. (1819). Bot. Register. 20. pi.. 1 690. Lemaire, 111. Hort. 13 : 94. Engelmann, Trans. Acad. St. Louis. 3:41. Ellacombe, Gard. Chron. iii. 2:111. r. gloriosa Gard. Chron. n. s. 12: 500, 688. /. 118. Kept. Mo. Bot. Gard. 8. pi. 7. Garden. 33 : 202. /. 58 : 446. /. A cultivated form of var. plicata, becoming 3 or 4 m. high, with greener leaves. Plates 46 J.I. 84, f. 4. Y. GLORIOSA MACULATA Carriere, Kev. Hort. 1859 : 389, 430. Koch, Dendrol. 2 2 :345. A low garden form, with the plicate dark green leaves persistently a little roughened on the margin : the varietal name referring to a mottled variation of the usual red tinging of the flowers. 22. Leaves more elongated, recurved. Y. RECURVIFOLIA Salisbury, Parad. Lond. pi. 31. (1806). Nuttall, Gen. 1 : 218. Pursh, Fl. 1 : 228. Elliott, Bot. S. C. & Ga. 1:401. Lemaire, 111. Hort. 13:94. Curtis, Bot. N. Car. 56. Baker, Gard. Chron. 1870:1184. Ref. Bot. 5.pL 321. Hemsley, Garden. 8 : 133, 136. /. Koch, Dendrol. 2 2 : 344. Gardiner, Journ. of Hort. 42 : 246. /. 1". gloriosa recurvifolia Engelmann, Trans. Acad. St. Louis. 3 : 39, 40. (1873). Baker, Journ. Linn. Soc. Bot. 18 : 225. Amer. Garden. 11: 661, 666. /. T.recurva Haworth, Syn. PI. Succ. 69. (1812). Gard. Chron. n. s. 18 : 689. Garden. 16 : 528. 47 : 337. /. Gardening 111. 18 : 230. /. 22 : 485. /. r. obliqua Regel, Gartenflora. 8 : 36. 17 : 161. pL 580. r. pendula Greenland, Rev. Hort. 1858 : 433. /. 128. Carriere, Rev. Hort. 1859:488. /. 104. Annales d'Hort. et de Bot. 2:93. Baker, Kew Bull. 1892 : 8. Garden. 1 : 238. /. J. gloriosa Riddell, N. O. Med & Surg. Journ. 8 : 763. Lloyd & Tracy, Bull. Torr. Bot. Cl. 28: 71, 91. r. gloriosa mollis Carriere, Rev. Hort. 1860: 362. Baker, Gard. Chron. 1870:1184. THE YUCCEAE. 77 Y. gloriosa planifolia Engelm. Trans. Acad. St. Louis. 3 : 39, 41. (1873). Y. filamentosa variegata Park & Cemetery. 11 ; 184. /. Y. variafolia Garden. 16:257. Shortly caulescent, branching. Leaves at first somewhat glaucous, nearly plane, long, flexible, recurved, about 50 mm. wide, often slightly plicate above, narrowly yellow- or brown-margined, often with a very few microscopic teeth, at length entire or slightly flliferous. Panicle narrow, the scape often included. Styles shouldered. Fruit erect, oblong, with 6 winged ribs mostly infolded over the nectarial grooves : seeds rather dull, 6 to 7X 7 to 8 w&-> the surface less grooved. Plates 46. 47. 84, f. 5. * Sea islands " and adjacent coast of Georgia, and on Dauphin, Ship and Breton islands, between the mouth of the Mobile and the mouth of the Mississippi river. Plate 94, f. 2. This species appears to have been in cultivation since 1794, and, like the preceding, is represented by many gar- den varieties, among which some of the described hybrids already referred to are placed in the following key : Leaves neither variegated nor very broadly margined. Y. recurvifolia. Bracts blackish- or purplish-brown. f. tristis. Leaves dark green, 75 mm. broad. Y. X Andreana. Leaves with conspicuous brown margin. f. rufocincta, Leaves variegated. With broad yellow margin. f. marginata. With median yellow band. f . variegata. With median reddish stripe. f. elegans. Short and broad with pale or purplish stripes. Y. X dracaenoides. Y. KECURVIFOLIA Salisbury. Synonymy as above. Leaves soon becoming dark green, greatly elongated, very much recurved. Plates 46, f. 2. 47, f. 1. 84, f. 5. The usual wild form. Y. recurvifolia tristis (Carriere) Trelease. Y. gloriosa tristis Carrifere, Rev. Hort. 1860 : 303. Koch, Dendrol. 2 2 : 345. A form with blackish-purple bracts. 78 MISSOURI BOTANICAL GARDEN. Y. RECURVIFOLIA RUFOCiNCTA Baker, Gard. Chron. 187O : 1184. T. rufocincta Haworth, Suppl. 37. (1819). Regel, Gartenflora. 8: 37. Lemaire, 111. Hort. 13 : 95. Engelmann, Trans. Acad. St. Louis. 3:41. Y. gloriosa rufocincta Baker, Journ. Linn. Soc. Bot. 18:225. (1880). A low form with rather pronounced accentuation of the reddish- brown margin. Y. recurvifolia marginata ( Carriers) Trelease. Y. gloriosa marginata Carriere, Rev. Hort. 1880: 259. T. gloriosa marginata aurea Carrifcre, 1. c. 260. Y. gloriosa elegans marginata Gard. Chron. n. s. 10:667. (1878). Wiener 111. Gart.-Zeit. 5 : 76. Leaves bordered with yellow, and often also rosy tinted. Gardens. Y. recurvifolia variegata (Carriere) Trelease. Y. pendula variegata Carriere, Rev. Hort. 1875 : 400. Y. gloriosa variegata Carriere, Rev. Hort. 1880 : 260. Gard. Chron. 1873:6. iii. 6 : 276, 305. Y. pendula aurea Carriere, Rev. Hort. 1877 : 249. 1879 : 404. ? Y. recurva elegantissima, Wiener 111. Gart.-Zeit. 5 : 460. (1880). ? Y. glaucescens variegata Hort. A garden sport with median yellow stripe. Y. recurvifolia elegans Trelease. Y. gloriosa elegans variegata. Belg. Hort. 1880:63. Gard. Chron. n. s. 16:439. r. gloriosa variegata Belg. Hort. 1884 : 33. Y. gloriosa recurvifolia, fol. var. Rodigas, 111. Hort. 30: 13. pi. 475. (1883). Differs in having the median stripe reddish. Y. FLEXILIS Carriere, Rev. Hort. 1859:398. /. 89. Horticulturist. 14 : 548. /. Lemaire, 111. Hort. 13 : 97. Baker, Gard. Chron. 1870:1183. Journ. Linn. Soc. Bot. 18:224. Engelmann, Trans. Acad. St. Louis. 3 : 41 Koch, Dendrol. 2 2 : 345. Hemsley, Garden. 8:129, 134. /. Y. Mexicana Hort., in part. Shortly caulescent. Leaves mostly transiently glaucous, nearly plane, long, narrow (20 to 40 mm.), little if at all plicate, occasionally a little THE YUCCEAE. 79 persistently denticulate or filiferous, flexible, at least the outer recurved. Panicle loose, exserted on a long scape. Style somewhat shouldered. Fruit unknown. A many-formed plant, apparently known only in gar- dens. Plate 47, f. 2. The principal forms and the comparable named hybrids may be separated as follows : Leaves plane or little concave, bright glossy green, recurved. T. flexilis. Taller (1 or 2 m.). Leaves pale green. . f. ensifolia. Leaves somewhat falcate. f. tortulata. Leaves evidently flliferous in age. f. Hildrethi. Leaves glaucous, little recurved. f. patens. Leaves concave, pale green. Outer leaves recurved. f. semicylindrica. Leaves all strict. f . Peacockii. Leaves scarcely pungent. f . Soerhaavii. Leaves pale-striate, flliferous. T. X striatula. The following garden hybrids, with flexible leaves less than 25 mm. wide, might be sought here : Leaves flat, entire. ? J. X Massiliensis. ? r. X ensifera. Leaves flat, often denticulate. T. X laevigata. Leaves very concave. T. y^juncea. Y. FLEXILIS Carriere. Synonymy as above. Dwarf. Leaves long and narrow, loosely recurved, bright glossy green. Known only in gardens, where, according to M. Carriere, it is sometimes erroneously called Y. acuminata, Y. sten- ophylla, Y. longi folia, and Y. angusti folia. It is also in part the Y. gloriosa of gardens. Y. flexilis Peacockii (Baker) Trelease. F. Peacockii Baker, Journ. Linn. Soc. Bot. 18 : 223. (1880). Kew Bull. 1892: 8. Wiener 111. Gart.-Zeit. 6 : 320. Garden. 19:226. Scarcely appears to differ except in the numerous leaves being stricter. 80 MISSOURI BOTANICAL GARDEN. Y. FLEXILIS ENSIFOLIA (Greenland) Baker, Journ. Linn. Soc. Bot. 18: 224. (1880). F. ensifolia Greenland, Kev. Hort. 1859: 433. /. 1 29. Baker, Gard. Chron. 1870: 217. Ref. Bot. 5. pi. 318. Engelmann, Trans. Acad. St. Louis. 3 : 41. Hemsley, Garden. 8 : 134. /. Y. Eylesii Hort. Taller (1 to 1.5 m.) with less recurving, soon pale green, somewhat concave, entire leaves. Y. flexilis Hildrethi Trelease. Differs from f . ensifolia chiefly in having its frequently somewhat fal- cate leaves usually finely flliferous in age. Plate 41, f. 2. Cultivated, from unrecorded source, and escaped, at the place of Mr. J. A. Hildreth, at St. Augustine, Fla., where it is said to bloom through the winter and where the spec- cimen photographed was observed in flower at the end of May, simultaneously with Y. aloifolia, though it has never been known to set fruit. Y. flexilis tortulata (Baker) Trelease. Y. tortulata Baker, Gard. Chron. 1870: 1122. Engelmann, Trans. Acad. St. Louis. 3 : 41. Hemsley, Garden. 8: 133. F. gloriosa tortulata Baker, Journ. Linn. Soc. Bot. 18 : 226. (1880). T.falcata Garden. 16:369. (1879). Y. flexilis falcata Baker, Journ. Linn. Soc. Bot. 18:224. (1880). T. undulata Hort., in part. Differs from f. ensifolia chiefly in being shorter-stemmed and with the green leaves flatter and somewhat falcate, and from Y. gloriosa minor in its longer outer leaves being reflexed. Y. FLEXILIS SEMICYLINDRICA Baker, Journ. Linn. Soc. Bot. 18:224. (1880). F. semicylindrica Baker, Gard. Chron. 1870: 1217. Differs from f . ensifolia in its firm and deeply concave narrower leaves (less than 20 mm. wide). Y. flexilis Boerhaavii (Baker) Trelease. r. Boerhaavii Baker, Gard. Chron. 1870 : 1217. Journ. Linn. Soc. Bot. 18:224. Engelmann, Trans. Acad. St. Louis. 3:41. Chiefly differs from the preceding in its flat scarcely pungent leaves. THE YUCCEAE. 81 Y. flexilis patens (Andre) Trelease. T. patens Andri, HI. Hort. 17:120. /. (1870). Gard. Chron. 1871:412. r. pruinosa Baker, Gard. Chron. 1870 : 1122. Garden 8: 133. Y. gloriosa pruinosa Baker, Journ. Linn. Soc. Bot. 18:226. (1880). A garden form, said to have come from China, with less arched glau- cous slightly rough-margined leaves: approaching some of the forms of r. gloriosa. Y. gloriosa, Y. recurvifolia, and Y. flexilis, the last two of which have frequently been treated as forms or varieties of the first-named, present a number of interesting and suggestive peculiarities when studied comparatively. Y. gloriosa occurs spontaneously among the sand dunes of a restricted portion of the southeastern Atlantic coast, where it is often intimately associated with Y. aloifolia and one or more forms of Y. filamentosa. Y. recurvifolia, except for one isolated group of stations, is known from a still more limited part of the same coast. Y. flexilis is known only in gardens, and its source appears to have been as unknown to its describer as it is to those who now cultivate it. About these three so-called species, have clustered in horticultural literature a considerable number of cultivated forms, sometimes treated as varieties of one or the other and sometimes specifically named, all of them entire-leaved with the exception that the margin is more or less persist- ently a little roughened or denticulate or a little filiferous in several of them, and all, so far as I have observed rec- ords, flowering usually in late summer or later, occas- ionally well on to the end of the season. These forms are not infrequently aberrant when placed, from the appearance of a character usually present in some other of the three species than the one under which the given form goes on the general assemblage of its characters. This interblending of characters in some of the variants of plants so distinct in their typical forms as Y. gloriosa, Y. 6 82 MISSOURI BOTANICAL GARDEN. recurvifolia and Y.flexilis are, suggests the possibility that the connecting varieties may really be of hybrid origin. Opposed to this supposition, however, are the absence of any recorded history of their source or origin ; the fact that they have appeared in cultivation and are classed with plants likewise of garden origin or long cultivated and in their other forms giving evidence of considerable variabil- ity ; and, particularly, the facts that, except for Y. aloifolia, the Yuccas spontaneously fruit with extreme rarity away from their native home unless, as seems not to be the case in European gardens where these forms have made their appearance, a moth (Pronuba yuccasella} upon which their pollination almost absolutely depends has been introduced with them, and that most persons who have tried to fertil- ize the plants of this genus have met with little or no sus- cess. Still, suggestion of such hybrid origin has been made,* and the most positive proof is at hand that along the Mediterranean coast, at least, skilful operators can not only intercross these so-called species but can also hybrid- ize them reciprocally with other very distinct species both of the baccate and capsular sections of the genus. Thus, for instance, M. Deleuil, of Marseilles, in and subsequent to 1874, crossed Y. aloifolia variegata and Y. alba-spica (whatever that may be), Y. aloifolia variegata 5 with Y. pendula (or recurvifolia}, Y. plicata (or gloriosa plicata} $ with Y. angustifolia vera (or glauca}, Y. plicata with Y. X laevigata { = aloifolia variegata X alba-spica}, Y. pli- catayvfitkY.jilamentosa, Y. plicata $ with Y. Treculeana, Y. cornuta (or Treculeana} with various species, Y. aloifolia variegata with Y. angustifolia vera, Y. gloriosa longifolia (or Y. flexilis glaucescens ?) $ with various spe- cies, Y. X laevigata % with Y. filamentosa, Y. cornuta and * Ellacombe, for instance, supposed the T. Ellacombei of gardens, which I take to be synonymous with Y. gloriosa nobilis, to be a probable cross between Y. recurvifolia and the garden form known as T. gloriosa euperba. Garden. 16: 257. THE YUCCEAE. 83 Y. plicata, and Y. angustifolia vera$ and Y. TreculeanaQ with various species; and I have knowledge that within recent years a very large series of reciprocal crosses have been effected by Mr. Carl Sprenger between these sub- entire-leaved forms as well as between them and both baccate and capsular species, and within the latter groups.* In Texas, also, spontaneous hybrids between Y. rupicola and Y. Louisianensis appear to occur. Everything considered, therefore, the garden intermedi- ates between Y. gloriosa, Y. recurvifolia and Y. flexilis may at least quite as properly be looked on as being the probable results of occasional unrecorded crossing between these forms as merely very aberrant sports. Few of them appear now procurable, but as far as a knowledge of them can be obtained from the brief descriptions, the known hy- brids of M. Deleuil are capable of natural arrangement under one or the other of these so-called species. With respect to the latter, themselves, the same line of inquiry suggests itself. The garden Y. flexilis , though in its typical form much narrower- and greener-leaved and with more elongately pedunculate and lax panicle, appears mor- phologically to represent only an extreme development of Y. recurvifolia, with which, except that it lends itself read- ily to the coordination of a number of forms in this respect comparable with those similarly grouped under Y. recurvi- foUa, it would logically be connected. The latter itself presents to the eye a blending of the characters of Y. glori- osa and Y. flaccida, which led one of the best students of woody plants, Koch,t to suggest some years since that it may be a hybrid between Y. gloriosa and Y. Jilamentosa, under which name he doubtless meant the recurved-leaved plant here called Y. flaccida. No greater reason exists for * On the results reached by M. Deleuil see Eevue Horticole. 52 : 226. 55 : 109. 58 : 63. 67 : 81. /. 21-23. Gard. Chron. n. s. 18 : 807. t Dendrol. 2 2 : 344. 84 MISSOURI BOTANICAL GARDEN. the rejection of this supposition than in case of the similar one that intermediates between IT. gloriosa, Y. recurvi folia and Y. flexilis may be the results of various intercrossing, since the possibility of crossing Y. gloriosa and Y. flaccida has been demonstrated by some of the experiments referred to above; and M. Deleuil's selection of 150 very diverse seedlings from a single one of his crosses gives reason to suppose that on the one hand a number of different aber- rants of these species might have come from even one cross seeding, while on the other hand several well verified hybridi- zations between Y. gloriosa and Y. flaccida might perhaps fail to produce typical recurwfotia. The occurrence of the latter along the South Atlantic coast of the United States, while it suggests the spontaneous hybrid origin of the typi- cal form of this species, does not preclude the possibility that the same form, and particularly its aberrant varieties, may have originated by a comparable process in gardens, where, in fact, they are alone known at present. Though Y. gloriosa and Y. fllamentosa are typically very dissimilar in aspect as well as in technical characters, I have seen side by side on the sand dunes of Tybee Isl- and, Georgia, an acaulescent plant of the spontaneous variety plicata of the former and a normal plant of the form of the latter known as var. concava, so similar in foliage appearance that it was only on close approach that the thinner texture and freely filiferous margin of the leaves of the latter served for its recognition, and I should be even more disposed to believe Y. gloriosa plicata a hybrid between Y. gloriosa and Y. filamentosa concava than to accept the suggestion of Koch concerning Y. re- curvifolia. As to Y. gloriosa, I have long thought that I saw in its characters somewhat of a blending of those of Y. filamentosa and Y. aloifolia,ihe leaves having something of the firmness and thickness of texture of the latter, and something of the thinness and concavity of the former or its variety, with. THE TUCCEAE. 85 frequent vestiges of the marginal characters of both ; while in the color, shape and texture of the perianth, the slight stipe at base of the ovary, the sometimes rather short shouldered style, the mostly pendent indehiscent fruit with thin exocarp drying about a papery core, and the often venously grooved if not truly ruminated seeds, Y. gloriosa holds even more nearly the mean between the two species named. The suggestion of a spontaneous hybrid origin of Y. gloriosa offered by this blending in it of the characters of the two other species with which it is most closely associ- ated, would be less strong if Y. gloriosa behaved in general like a normal species of the genus, if it were of greater geographic distribution, or if it occurred in places thor- oughly isolated from the assumed parents. As has been said, though locally rather abundant, Y. gloriosa as a spontaneous plant is limited, so far as is now known, to a very restricted region about the Carolina and Georgia coast. It is, moreover, a very unusual species in its life processes. In the arid region of the Mexican table-land, the Yuccas are known to be largely dependent for their blooming season upon necessary rainfall, so that a given species, though usually fairly regular, may bloom in aberrant years at any time between midwinter and mid- summer, and the Pronuba moth which serves as pollinator appears to show a similar susceptibility to moisture in the soil, and commonly emerges from the pupa state synchron- ously with the flowering of the Yuccas. Y. gloriosa, how- ever, growing in a region where the other Yuccas bloom pretty regularly during a rather limited part of the spring, when the Pronuba flies, differs from these species in flowering usually in late summer and autumn, though exceptional flower clusters appear to be developed at almost any season of the year, and the only instances that I cer- tainly know of in which its fruit has been observed were once when early blooming plants cultivated in Washington gg MISSOUKI BOTANICAL GARDEN. bore fruit,* once when Dr. Mellichamp found fruit on a plant which had bloomed simultaneously with T. filamen- tosaj and a third instance observed by me on Tybee Island in May last, (Plate 44, f. 2} on a plant which must have bloomed just about as T '. filamentosa was coming into flower. The species, therefore, is all but restricted for its propagation to vegetative methods, by which its present dis- tribution along the sand dunes can fairly well be explained, since the well-budded thick subterranean shoots possess great vitality. What has been said of the ecology of Y. gloriosa might be repeated almost verbatim for Y. recurvifolia, which is likewise autumnal-flowering, and the fruit of which, barring several rather questionable statements in gardening journals, to my knowledge has never been observed until Dr. Mellichamp, in the summer of 1901, found plants fruit- ing in cultivation in the neighborhood of Charleston, and furnished the material from which the description and illustration here published were drawn. The occurrence of Y. recurvifolia on several islands between the delta of the Mississippi and the mouth of the Mobile river, which is not connected with the present question, may, perhaps, have been brought about by currents transporting rhizome fragments derived from plants cultivated somewhere along one of the rivers opening on the northern shore of the Gulf. These ecological considerations suggest with force that if species in the time-honored use of that term, Y. gloriosa and Y. recurvi 'folia, so far as their spontaneous forms are concerned, are of unexpectedly restricted distribution in a region where their congeners are widespread, and that they manifest a surprising disharmony with their surroundings which, because of the rigid pollination requirements of all of this genus but aloifolia, has thrown them into almost * Engelmann, Trans. Acad. Sci. St. Louis. 3: 211. t Kept. Mo. Bot. Gard. 4 : 199. THE YUCCEAE. 87 absolute dependence upon vegetative methods of propaga- tion; though they continue to flower profusely, and because of the unusual if aberrant period over which their bloom- ing extends they now and then fruit, and they are shown to be so fertile under skilful artificial pollination that there is little reason to doubt that they would fruit regularly if they bloomed when the Pronuba was about ; while over the great territory lying between the Atlantic and Pacific and the big bend of the Missouri river and central Mexico, the other Yuccas have held so close a relation with their pollinators as to be very fruitful under all ordinary circum- stances. The ecological facts stated, however, are con- sistent with the morphological suggestion that Y. gloriosa may be a hybrid between Y. aloifolia and Y. filamentosa, and the two considerations appear to constitute so strong an argument for the acceptance of the a priori theory advanced, as to throw the burden of proof upon any who would still regard gloriosa as a species in the ordinary sense, though for purposes of classification it, as well as recurvifolia and flexilis, may continue to be treated as species.* 222. Leaves crowded, regularly and rigidly arcuate. Y. DE SMETIANA Baker, Gard. Chron. 187O : 1217. Joura. Linn. Soc. Bot. 18 : 222. Kew Bull. 1892 :8. ? Y. Helkinsi Hort. Caulescent, at length with a trunk 2 or 3 m. high. Leaves rigid, evenly and stiffly recurved, becoming .4 m. long and 25 mm. or more wide, pur- ple tinged, entire or slightly rough-margined at base, not pungent. Flow- ers and fruit unknown. Plate 48. A garden plant ascribed to Mexico, which when small is very suggestive in appearance of a lily because of its crowded arching not at all concave leaves : quite unlike any other Yucca, and perhaps not of this genus. No positive record exists of the source of the plants of this species cul- * The substance of these conclusions was presented at the Denver meeting of the Botanical Society of America, in August 1901. 88 MISSOURI BOTANICAL GARDEN. tivated at the Missouri Botanical Garden, but they are be- lieved to have come from northern Mexico, many years ago r through Dr. Parry. B. B. Fruit pendent, fleshy and edible : seeds thick, often convex, nearly without a thin border; albumen evidently ruminated. Sarcoyucca. I. Fruit coreless, purple-fleshed. Leaves with denticulate horny border. Y. ALOIFOLIA Linnaeus Sp. Plant. 319. (1753). Walter, Fl. Carol. 124. Michaux, Fl. 1 : 196. Pursh, Fl. 1 : 228. Nuttall, Gen. 1 : 218. Riddell, N. O. Me australis, are described and figured in several places. Associated with them are numerous specimens of Y. radiosa and, in smaller numbers, the true Y. macrocarpa of the great bend of the Rio Grande, which, as has been shown above, is a well-marked species and preserves all of the floral characters of a true Yucca; and, as indicative of their probable range to the southward, it may be mentioned that they are accompanied by Agave applanata, which, in its typical form, is not known elsewhere in the United States. 118 MISSOURI BOTANICAL GARDEN. As it occurs from a little way east of Sierra Blanca to the vicinity of Malone, this tree is usually 2 or 3m. high, rarely reaching 5 meters, and the thin-barked stem, which may reach a diameter of about half a meter, very rarely branches, though occasionally one or two ascending branches are produced. Well developed plants, even if small, differ conspicuously from those of Yucca macrocarpa in their rounder head and the usually greater number of their spreading leaves, which, smooth or at most slightly roughened on the occasional dorsal angles, are of a crab- apple green, openly concave to the very short stout spine, and though at first coarsely filiferous, later have only a few short pectinate thickish fibers toward the tip, while the remainder become detached to the base, where they remain in a loosely cobwebby mass between the leaves, which in age become reflexed and normally persist as a thatch on the trunk even to its base. On vigorous plants the leaves attain a width of 75 mm. and a length of 1.25 m. This species, which is well described by Professor Sar- gent, under the name Yucca macrocarpa, I take pleasure in dedicating to Mr. C. E. Faxon, whose excellent figures of it in the Silva faithfully represent its technical characters. S. Carnerosana Trelease. A simple or rarely slightly branched tree, 1.5 to 6 m. high, at length .7 m. in diameter. Leaves as in the last. Panicle on a stout white- bracted stalk, densely branched close above the leaves, glabrous or exceptionally tomentose. Flowers expanding 75 to 100 mm. ; the cylin- drical tube 12 to 25 mm. long. Fruit oblong, 50 to 75 mm. long, 40 mm. in diameter : seeds 7 to 9 X 8 to 10 mm. Frontispiece to article and plates 72-75. 76,f. 1, 77. 81, f. 12. 83, f. 2. Northeastern Mexico, from the Carneros pass to about Catorce and Cardenas. Plate 94, f. 2. Some years since, Mr. C. G. Pringle made characteris- tically excellent herbarium specimens of a tree which forms large forests about Carneros, Mexico, which were distributed as doubtfully representing a variety of Yucca THE YUCCEAE. 119 fiaccata. These specimens (nos. 2841, 3912), represent another species of Samuela, which, from near the city of Saltillo extends southwards, on the mountain slopes and in the higher valleys, to some distance below the Tropic of Cancer, and is especially abundant in the higher valleys about Carneros pass, where the Mexican National railroad crosses the mountains south of Saltillo, and about Las Tablas on the Tampico branch of the Mexican Central. Like the preceding species, this is a low round-headed tree, very rarely bearing one or two short branches at the apex, and thus in marked contrast with the branched shorter-leaved Y. australis which accompanies it in small numbers about Carneros and elsewhere. The leaves vary considerably in thickness, and the thinner ones are usually a little plicate though they are still thick and rigid. The very thick fibers of the leaves distributed by Mr. Pringle are exceptional. The axis of inflorescence, which, though usually erect, is sometimes arched over by the weight of the enormous panicle, is unusually succulent and devoid of fiber, so that a stalk as thick as one's wrist can be severed by a single cut of a pocket-knife. A striking feature of both species of the genus, but particularly marked in 8. Carnerosana, is the compact depressed bud, as much as 100 mm. in diameter, in which each branch of the panicle ends until blooming is far advanced. Even from a distance, the pure waxen-white fragrant flowers, which remain expanded to an unexpected degree during the daytime, are marked by their cylindrical tube which gives them the appearance of those of Polianthes, though the ovary is free from the perianth, as in other Liliaceae. The fruit of both species, like that of the baccate Yuccas of the southwest, is usually greenish-yellow, though some- times tinged with red or purple, and the soft sweet pulp is pale. 120 MISSOURI BOTANICAL GARDEN. ECONOMIC USES. In contrast with the Aloineae, the Yucceae possess very fibrous leaves comparable with those of the agavoid Amaryl- lidaceae, and local use is made of the fiber* almost every- where that the plants grow. In the southeastern United States, and as far west as the Indian Territory, the leaves of species of Yucca of the filamentosa group, commonly called " bear-grass," are much used for domestic purposes such as making seats for chairs and especially hanging meat, for which they are so much prized in the country that the plants are commonly tolerated as weeds in cultivated fields from which other wild plants are eradicated. In Mexico and our southwestern states the fiber of several of the bac- cate species is crudely cleaned and put to various local uses, cordage included. f The long leavesof " palma loca " ( Y. Treculeana), with coarse fiber, and "izote" ( Y. Schottii Jaliscensis} , with fine fiber, are apparently of considerable use in this manner, respectively in the eastern and western parts of Mexico. About the Carneros pass, where it is very abundant, Samuela Carnerosanais similarly used, and Dr. Millspaugh informs me that Ifesperaloe funifera is re- ported as planted for its fiber about Bustemente, in the Mexican state of Nuevo Leon. The fiber of Hesperoyucca is said by Palmer (/. c.)to be fine and excellent. Cleaning the fiber of all of these plants appears to be attended with the general difficulties that make the commercial preparation of Agave fibers unsatisfactory, but I have seen machine- cleaned fiber of Yucca australis that appeared fairly good, and it may be that notwithstanding its shortness the fiber of these abundant large palma trees of the Mexican table- land will ultimately be used in quantity for the cheaper kinds of bagging, etc. * See Naudin, Kev. Hort. 1855: 141-9. Porcher, Resources of So. Fields and Forests. 530-1. t Palmer, Amer. Journ. Pharmacy. 50 : 586. THE YUCCEAE. 121 The trunks of the species of Yucca, Clistoyucca and Samuela are occasionally used for palisade construction, and in the Carneros pass I have seen houses built almost entirely of material obtained from S. Carnerosana, the walls of palisade-like trunks set on end, and the roof thatched with the leaves. Attempts have been made to use the fiber of Clistoyucca for paper-pulp,* of which a fair grade can be made notwithstanding the gumminess of the tissues ; and the trunks have sometimes been turned into coarse veneers for wrapping bottles, etc., as is commonly done with soft dicotyledonous woods like the cottonwood. The group generally seems to possess the saponifying properties of the Agaves, so that the stems and root stocks are not infrequently used as amoles,^ and a considerable quantity of vegetable soap is claimed to be made from Y. baccata, Y. glauca, and, judging from illustrations in ad- vertising matter, Y. radiosa. Notwithstanding their stiff-pointed leaves, the species which grow in the south western grazing country are attract- ive to cattle in the flowering season, and the animals often display some dexterity and no little courage in riding down the smaller trees or otherwise getting at their succulent flower -clusters, which are further gathered and carried in to be fed to sheep and other animals in some cases, as, for instance, in the Carneros pass, where I have seen large cart loads of the great panicles of /Samuela Carnerosana being taken to the hamlet for this purpose. In their early stages, too, the inflorescence of Yucca, Hesperoyucca and Samuela is said to be either boiled or roasted and used for human food or even eaten raw.t Like the crowns of '* sotol " (Dasylirion} , the nearly fiberless trunks of the southern Samuela are decorticated or split open so that they can be eaten by stock. * Palmer, 1. c. Shinn, Amer. Agriculturist. 1891 : 689. Land of Sunshine. 10*1, and advertisement, f See Palmer, I. c.- J See Palmer, 1. c. The Garden. 24: 104, frpm N. Y. Tribune. 122 MISSOURI BOTANICAL GARDEN. As a rule, the fruits of the baccate species of Yucca and of Samuela are promptly eaten by birds, rats, etc., but domesticated animals are said to like them, and, being quite sugary, they are enjoyed by the Indian and Mexican chil- dren, who commonly call them figs or dates. All that I have tasted possess, in combination with their sweetness, a characteristic bitterness, which makes them somewhat un- palatable, and those of the Rocky Mountain and Mexican region possess a rather viscid pulp which renders them unpleasant to handle when broken. My friend Mr. Bur- bidge has compared the fruit of Yucca aloifolia with black- currant jam with a little admixture of quinine, its purple color no doubt strengthening the suggestiveness. The seeds of the baccate species are said to be purga- tive, though Palmer (I. c.) says that the seeds of Clisto- yucca and Hesperoyucca are ground and eaten, either raw or as "mush; " and Gambold (Amer. Jour. Sci. 1819: 251) states that the pounded roots are used as a fish poison. It would be interesting to have their active principles de- termined. All of the species, when used in the right way, are of decorative value. Y. filamentosa, Y. flaccida, Y. gloriosa, Y. recurvifolia, Y. glauca, Y. baccata, and Y. Harri- maniae appear to be hardy as far north as St. Louis, and Y. Treculeana is reported frost-hardy at Angers, France (Garden. 12: 369), but the other species, so far as tested, demand a climate scarcely less mild than that of our southern states, California or the Riviera. PHYLOGENY AND ECOLOGY. Little can be said as to the origin or mode of specializa- tion of the Yucceae. They are characteristic xerophytes, even those which grow in the moist climates frequently having a preference for dry places, such as sand dunes. Their underground parts are frequently fleshy and very tenacious of life, their stems hold a considerable amount THE YUCCEAE. 123 of moisture, and their leaves are well guarded against undue transpiration. Like other arboreous Liliaceae, their larger representatives produce the impression of being the culmination of a vegetative type perhaps formerly of wide distribution, but now barely able to hold its own except in desert regions where competition between plants is less than elsewhere, while structural adaptation enables them to endure the rigors of this last resort, in a sense, therefore, recalling the bald cypress (Taxodium) among conifers, which for similar reasons has betaken itself to the other extreme of deep swamps. I know of no ecological explanation of the filif erous shedding of the leaf- margins of many species. The dissemination arrangements of the Yucceae are of the more highly specialized types. Many species, consti- tuting the genus ffesperaloe, Hesperoyucca, and the capsular section of Yucca, are wind-disseminated, with thin flat seeds lifted from time to time out of the suberect capsules by gusts of wind. In Clistoyucca the indehiscent mature fruit is spongy and light and apparently adapted to being blown about by the desert winds after the manner of blad- der-fruits or tumble-weeds. Yucca gloriosa and Y. recur- vifolia possess fruits which do not dehisce, though their seeds are thin and flat ; nor do they become edible in ripen- ing, but dry to a firm almost wooden consistency, out of harmony with any usual mode of dissemination. All of the baccate species of Yucca and the two species of Samuela have fleshy edible fruits at maturity, and their abundant endosperm suggests an adaptation to the dry regions, in which all of them, so far as known, live, with the exception of Y. aloifolia, and, perhaps, Y. elephantipes. That they have been derived from thin-seeded capsular species seems more probable than the reverse, and the coreless fruit of the seaside Y. aloifolia suggests its independent fruit specialization rather than a genetic con- 124 MISSOURI BOTANICAL GARDEN. nection with the desert species, which possess a firm, parchment-like core immediately about the seeds. The pollination relations of nearly all of the group are among the most peculiar and exclusively restricted thus far discovered. Hesperaloe secretes much nectar and appears adapted to birds, as are the Cape aloes, to which it bears no inconsiderable resemblance in its flowers. The other genera are sparingly if at all nectariferous, though all have septal glands, which are rather small in Clistoyucca, but verv large in the others. Yucca aloifolia, again in an exceptional way, appears to be freely self -fertile, but self- seeding is very unusual with all of the other species of this genus, as it appears to be with Hesperoyucca, Clisto- yucca and Samuela. These, so far as known, depend for their pollination upon small moths belonging to the tineid genus Pronuba, of which one species (P. syntlietica) is known only in connection with the single species of Clisto- yucca, one (P. maculata, and its variety aterrima), with the single species of Hesperoyucca, and the only other known species (P. yuccasella) accompanies the various species of Yucca across the continent and has a known north and south range from the great bend of the Mis- souri river to central Mexico. These moths are not known to feed, in the larval stage, on anything but the developing seeds of the plants named ; so that the mutual dependence of moth upon plant and of plant upon moth appears to be absolute, no doubt, taken in connection with the other ecological peculiarities of the yuccoids, a fact of the greatest suggestiveness, but the bearing and meaning of which has as yet escaped both botanists and entomologists. That the flowers were formerly pollinated otherwise appears to be indicated by the presence of nectar-glands, which now appear to be useless. The Jong perianth tube of Samuela, a type of struc- ture usually connected with pollination by some insect of corresponding tongue-length, for which the nectar is thus THE YUCCEAE. 125 kept from shorter-tongued insects, is so closely applied about the lower part of the ovary, as, apparently, to make it impossible for any insect to reach the bottom of the latter, with even a very slender tongue. Though the actual pollination of this genus is yet to be observed, it is effected by Pronuba yuccasella, at least in 8. Faxoniana, in the flowers of which pollen-laden females of the moth were discovered by my son and myself in April, 1902, and the only explanation of the highly specialized tubular peri- anth I can suggest is that, restricting the access of the ovipositing moths to the upper half or two-thirds of the ovary, it may limit the number of eggs that they can lay in a given pistil, to the advantage of the plant. EXPLANATION OF PLATES. Unless otherwise stated, the illustrations are from pho- tographs by the author. Where two illustrations occur on a plate, the upper or left-hand is referred to first. Frontispiece to article. Samuela Carnerosana, in the Carneros Pass, Mexico. Plate 1. 1, Hesperaloe parmflora, cultivated in San Antonio; 2, H. jparviflora Engelmanni, cultivated at the Missouri Botanical Garden. Plate 2. Flowers of Hesperaloe parviflora Engelmanni, natural size. from the Missouri Botanical Garden. Plate 3. Hesperaloe funifera, at Peyotes, Mex. Plate 4. 1, Hesperaloe funifera, capsules from Peyotes, natural size ; 2, Hesperoyucca Whipplei, capsules from Arrowhead Springs, Cal., natural size. Plate 5. Hesperoyucca Whipplei, and its flowers, reduced, at the sum- mit of the Cajon Pass, California. Plate 6. Clistoyucca arborescens, at Hesperia, California. Plate 7. Clistoyucca arborescens, flowers, reduced, and fruit, natural size, at Hesperia, Cal. Plate 8. Yucca filamentosa, at Sanford, Fla., and flowers, natural size. Plate 9. Yucca filamentosa bracteata, cultivated at Brunswick, Ga. Plate 10. Yucca filamentosa concava, in sand dunes, Isle of Palms, S. C. Plate 11. Yucca filamentosa media, cultivated in Tower Grove Park, St. Louis. Photographed by P. T. Barnes. Plate 12. Partly grown fruit of Yuccas cultivated in the Missouri 126 MISSOURI BOTANICAL GARDEN. Botanical Garden, natural size. I, Y. filamentosa ; 2, Y. flaccida glau- cescens. Plate 13. Yucca flaccida glaucescens, cultivated in the Missouri Bo- tanical Garden. Producing racemose secondary inflorescences, in addi- tion to the central panicles. Plate 14. Yucca flaccida glaucescens, cultivated at the Missouri Botanical Garden, showing thermotropism of inner leaves. 1, Normal position of leaves, at a temperature slightly above the freezing point ; 2, Leaves inrolled, at 26 F. Plate 15. Yucca flaccida glaucescens, cultivated at the Missouri Botanical Garden. Photographed by P. T. Barnes. Plate 16. Yucca flaccida, near Anniston, Ala. Plate 17. Capsules, natural size. 1, Yucca flaccida glaucescens, cultivated at the Missouri Botanical Garden; 2, Y. tenuistyla, Industry, Tex., Lindheimer. Plate 18. r?(cca tenuistyla, near Sealy, Tex. Plate 19. Yucca tenuistyla from near Sealy, Tex. Small sized flowers, natural size. Photographed by P. T. Barnes. Plate 20. Yucca constricta. 1, Cultivated at the Missouri Botani- cal Garden from Seward Co., Kas. ; 2, Near Uvalde, Tex. Plate 21. Capsules, natural size. 1, Yucca constricta, Cline, Tex. ; 2, Y. radiosa, Benson, Ariz. Plate 22. Yucca radiosa, at Benson, Ariz. Fruiting plants and an exceptionally symmetrical young plant. Plate 23. 1, Yucca angustissima, a type sheet in the Engelmann her- barium ; 2, Yucca glauca, near Albuquerque, N. M. Plate 24. Capsules, natural size. 1, Yucca angustissima, from near the Grand Canon, Ariz. ; 2, Yucca glauca, from Mauitou, Col. Plate 25. Yucca glauca, cultivated in the Missouri Botanical Garden. Plate 26. Yucca glauca stricta, cultivated in Tower Grove Park, St. Louis, from Seward County, Kas. Plate 27. Flowers of Yucca glauca stricta, natural size, from the preceding. Photographed by P. T. Barnes. Plate 28. Yucca Harrimaniae, at Helper, Utah. Plate 29. Yucca Harrimaniae, Helper, Utah, Capsules, natural size. Plate 30. Yucca Arkansana, near Fort Worth, Tex. Plate 31. Yucca Arkansana, near Dallas, Tex., fruiting plants. Plate 32. Yucca Louisianensis. l, Near Jefferson, Tex.; 2, Near Texarkana, Tex. Plate 33. Yucca Louisianensis, Hughes Springs, Tex. Plate 34. Yucca Louisianensis, Hughes, Tex. 1, Form with slen- derer, paler style; 2, Form with very tumid dark green style, slightly reduced. Plate 35. Yucca rigida, near Picardias, Mex. Plate 36. Capsules, natural size. 1, Yucca rigida, from Picardias, Mex. ; 2, Yucca rostrata, from Peyotes, Mex. THE YUCCEAE. 127 Plate 37. Yucca rupicola? Aberrant sheet of Wright, no. 1909, in the Torrey herbarium. Plate 38. Yucca rupicola, the more normal Gray herbarium sheet of Wright, no. 1909. Plate 39. Yucca rupicola. 1, Flowering plant, on limestone hills a few miles west of Fort Worth, Texas; 2, Flowers, slightly reduced, of plant cultivated by Mr. J. Reverchon, from same locality. Plate 40. Yucca rostrata, at Peyotes, Mex. Flowering plants. Plate 41. Yucca rostrata, at Peyotes, Mex. The upper figure show- ing the lozenge-shaped leaf-scars. Plate 42. Yucca rostrata, at Peyotes, Mex. fruiting plants. The foreground is occupied by Agave heteracantha. Plate 43. Yucca gloriosa, in the sand dunes of Tybee Island, Ga. Plate 44. Yucca gloriosa, Tybee Island, Ga. 1, Smooth-barked trunk, with roots, exposed by the shifting of the sand; 2, With partly grown fruit, photographed in May. Plate 45. Yucca gloriosa minor, cultivated in the Missouri Botanical Garden. At the left are Y. aloifolia, with narrow leaves, and Y. elephan- tipes, with broad more flexible leaves. Plate 46. 1, Yucca gloriosa superba, fruit and cross section, natural size, cultivated in Washington, D. C. (Schott); 2, Yucca recurvifolia, fruit, natural size, cultivated at Bluffton, S. C. (Mellichamp, in 1901). Plate 47. 1, Yucca recurvifolia (2 m. high), cultivated in the National Cemetery, Vicksburg, Miss. ; 2, Yucca flexilis Hildrethi, escaping, at St. Augustine, Fla., photographed in May. Plate 48. " Yucca De Smetiana," cultivated in the Yucca tower of the Missouri Botanical Garden. Plate 49. Yucca aloifolia. 1, Associated with Ipomoea Pes-Capreae, on the dunes of South Beach, St. Augustine, Fla. ; 2, Overgrown with SmilaXf on the dunes of Tybee Island, Ga. Plate 50. Yucca aloifolia Menandi, type plant cultivated at the Mis- souri Botanical Garden. Photographed by P. T. Barnes. Plate 51. Yucca elephantipes. 1, A large tree, at El Florido, Guate- mala; 2, The dilated base of a tree, at Chiuautla, Guatemala. Plate 52. Yucca Treculeana. 1, In flower, cultivated at C. P. Diaz, Mex. ; 2, In fruit, near Peyotes, Mex. Plate 53. Yucca Treculeana canaliculata. Cultivated in the Alamo Plaza, San Antonio, Tex. Plate 54. Yucca Treculeana canaliculata. Cultivated at the Mis- souri Botanical Garden. Photographed by P. T. Barnes. Plate 55. Yucca Schottii, west of Nogales, Ariz., photographed in August : the second figure, from near the boundary monument in the Sierra del Pajarito. Plate 56. Yucca Schottii Jaliscensis. " Izote ", in the suburbs of Zapotlan, Mexico, photographed in September. Plate 57. Yucca brevifolia. The mixed type-sheet, in the Torrey her- 128 MISSOURI BOTANICAL GARDEN. barium. The leaves are representative of Y. brevifolia, and the inflor- escence, apparently, of Y. Schottit. Plate 58. Yucca brevifolia, toward the Santa Cruz river, to the northeast of Nogales, Ariz. Plate 59. Yucca brevifolia. Tree about 2 meters high, with panicle axis from preceding year, near Nogales, Arizona. Plate 60. Yucca australis. The original sheet of Thurber's collection from Parras, Mex., in the Torrey herbarium. Plate 61. Yucca australis. 1, In fruit, at Parras, Mex.; 2, In flower, near Topo Chico, Monterey, Mex. Plate 62. Yucca valida. Old hedgerows, near Durango, Mex. Plate 63. Yucca valida, near Gutierrez, Mex. Plate 64. Yucca valida. 1, Near Gutierrez, Mex., with Opuntia leu- cotricha in the foreground ; 2, Near Camacho, Mex. Plate 65. Yucca valida, near Gutierrez, Mex. The lower partrof the trunk, some years before, had been decorticated without killing the tree, over the lower part of which a new bark has formed. Plate 66. Yucca valida. Flowers, somewhat reduced, from near Gutierrez, Mex. Plate 67. Yucca valida. Type sheet, from San Gregorio, L. Cal., in the Brandegee herbarium. Plate 68. Yucca baccata, in the Grand Canon, Ariz. The fruit is 20 cm. long. Plate 69. Yucca baccata. Fruit of the preceding, natural size (fore- shortened), showing the basal disk. Photographed by P. T. Barnes. Plate 70. Yucca macrocarpa. Flowering plant, near Sierre Blanca, Texas. Plate 71. Yucca macrocarpa. Fruiting plants, in the type region, in the great bend of the Rio Grande. Plate 72. Yucca Mohavensis, at Drake, Ariz. The right-hand figure has a very characteristic plant of Fouquiera in the foreground. Plate 73. Samuela Faxoniana, near Sierra Blanca, Tex. Character- istic round-headed trees. Plate 74. Samuela Faxoniana, a partly sterile fruiting plantj with persistent bracts, and a plant beginning to bloom, near Sierra Blanca, Texas. Plate 75. Samuela Faxoniana. Leaf tips and partly grown fruit (showing the fleshy base and short split tube of the perianth), from Sierra Blanca, Tex., natural size. Plate 76. Samuela Carnerosana, in the Carneros Pass, Mex. Full blown trees, and the foliage head of a young plant. Plate 77. - - Samuela Carnerosana. Flowering and fruiting trees in the Carneros Pass, Mex. The partly sterile inflorescence is conspicuous even in fruit, because of its persistent large bracts. Plate .78. Samuela Carnerosana in the Carneros Pass, Mex. 1, Fruiting tree; 2, Early stage of flowering, showing the large bracts and the buds in which the panicle branches at first end. THE YUCCEAE. 129 Plate 79. Samuela Carnerosana, from the Carneros Pass, Mex., natural size. 1, Inflorescence bud; 2, Flower, with nearer part of perianth removed, and half grown fruit with the persistent split perianth tube upwards of 2 cm. long. Plate 80. 1, Samuela Carnerosana, in the Carneros Pass, Mex.; 2, Yucca flaccida, var., cultivated at the Missouri Botanical Garden. Both reduced. The perianth differences of the two genera are well shown. Plate 81. Samuela Carnerosana, in the Carneros Pass, Mex. 1, A trunk decorticated by slashing it on the two sides and tearing the leaves down, exposing the pulpy interior for stock to feed upon; 2, A fruit, somewhat reduced, showing the split dried perianth tube. Plate 82. Yucca elephantipes, at Chinautla, Guatemala, and Samuela Faxoniana, at Sierra Blanca, Texas. Flowers, reduced about one-third. Plate 83. Seeds of Yuccas, natural size.* 1, Y. Jilamentosa con- cava, Isle of Palms, S. C. (Trelease) ; 2, I", flaccida glaucescens, culti- vated at the Missouri Botanical Garden ; 3, T". tenuistyla, Industry, Tex. (Lindheimer) ; 4, Y. constricta, Uvalde, Tex. (Trelease); 5. Y. radiosa, Presidio, Tex. (Trelease) ; 6, Y. angustissima near Grand Canon, Ariz. (Trelease); 7, Y. Arkansana, New Braunfels, Tex. (Lindheimer); 8, r. Louisianensis, Atoka, Ind. Ter. (Butler) ; 9, Y. glauca, N. W. Mis- souri (Bush); 10, Y. Harrimaniae, Helper, Utah (Trelease). Plate 84. Seeds of Yuccas, natural size. 1, Y. rigida, Picardias, Mex. (Trelease); 2, Y. rupicola, New Braunfels, Tex. (Lindheimer); 3, Y. rostrata, Peyotes, Mex. (Trelease) ; 4, Y. gloriosa superba, cultivated in Washington, D. C. (Schott); 5, Y. recurvifolia, cultivated at Bluffton, S. C. (Mellichamp) ; 6, T. aloifolla, Bluffton, S. C. (Mellichamp) ; 7, Y. elephantipes, cultivated at the Missouri Botanical Garden; 8, Y. Tre- culeana, New Braunfels, Tex. (Lindheimer). Plate 85. Seeds of Yucceae, natural size. 1, Yucca Schottii, Pinal Mts., Ariz. (Pringle); 2, Y. australis, Parras, Mex. (Trelease); 3, Y. valida, Gutierrez, Mex. (Trelease) ; 4, Y. baccata, Grand Cafion, Ariz. (Trelease); 5, Y. macrocarpa, near Presidio, Tex. (Trelease); 6, Y. Mohavensis, Drake, Ariz. (Trelease); 7, Hesperaloe parviflora, Texas (Wright) ; 8, H. funifera, Hacienda de Angostura, Mex. (Pringle, no. 3911); 9, Hesperoyucca Whipplei, Arrowhead Springs, Calif . (Trelease) ; 10, Clistoyucca arborescens (Palmer); 11, Samuela Faxoniana, Sierra Blanca, Tex. (Trelease); 12, S. Carnerosana, Carneros Pass, Mex. (Trelease). Plate 86. Germination of Yuccas, natural size. 1, T. radiosa; 2, Y. macrocarpa. Plate 87. Germination of Yucceae, natural size. 1, Clistoyucca ar- borescens; 2, Samuela Carnerosana. Plates 88-99. Geographical distribution of Yucceae. Stations noted by the author are indicated by a X> aod the general range known to him is shown by horizontal shading. Many gaps require filling. * The figures in this and the following plates are numbered from left to right in the several rows, beginning with the uppermost. 9 INDEX. (Synonyms in Parenthesis. ) Agave applanata 117 Cubensls (114) geminiflora 114 funifera (36) heteracantha 127. pi. 42 Aloe Americana, Comm. (88, 89, 91) Juccae follls, Sloane. (88) purpurea levls, Hunt. (91) yuccae follls, Pink. (89) Aloe yuccaefolia (30, 31) Aloes Floridana, Pluk. (88) Alolneae 27, 28 Astella 27, 28 Beancarnea (27) longifolia (115) Chaenoyucca 43, 44, 46 Cllstoyucca 29, 41, 123, 124 arborescens 41, 103, 121, 122, 125, 129. pi. 6, 7, 86, f. 10, 87, f. 1, 88 Cohnla 28 Cordyline 27, 28 Cordyline foliis pungentibus, Van Eoyen (72, 89) Dasylirion 27, 28, 114, 121 acrotrlchuin 114, 115 aloefolmm (103, 105) graminlfohum 40, 114 longlfohum (115) pitcalrnifolium (115) serratlfollnm 115 Dracaena 27, 28 Ehrenbergli (94) Flntelmannl (94) Lenneana (94) Lennei (94) yuccoides (94) Dracaeneae 27,28 Dracaennideae 28 Draconl arborl affinis, Bauh. (91) Euyncca 43 Fouqniera 128. pi. 72 Fnrcraea Bedinghausii 43, 106 Cubensls 114 gemlnlspina 114 Ilechtia glomerata 115 Herreria 27,28 Hesperaloe 27, 28, 29, 31, 123, 124 Davyl (36, 37) Engelmanni (33, 36) fnnifera 29, 36, 120, 125, 129. pi. 3, 4, f. 1, 83, f. 8, 96 parviflora 29, 30, 125, 129. pi. l,f. 1, 85, f. 7, 83 Engelmanni 33, 125. pi. l,f. 2, 2 yuccaefolia (30, 33) Hesperocallis 27, 28 Hesperoyncca 29, 38, 123, 124 Whipplel 39, 120, 121, 122, 12E, 129. pi. 4, f. 2, 5, 86, f. 9, 88 Heteroyucca 43, 45, 71 lucca, Park. (72) Perana, Gerarde (72) Peruana, Johnson (72) Jnca Americana, Munt. (47) gloriosa, Munt. (72) Lilla regia (105) Lillum regium (105) Milllgania 27, 28 Nollna 27, 28, 71 longifolia 69, 115 Nolineae 28 Pronuba maculata 124 aterrima 124 synthetica 124 yuccasella 82, 85, 87, 89, 124, 126 Roezlia bulblfera (105) regla (105) Samuela 29, 116, 122, 123, 124 Carnerosana 117, 118, 120, 121, 125, 128, 129. Frontispiece, pi. 76-79, 80, f. 1, 81, 86, f. 12, 87, f. 2, 98 Faxoniana 112, 117, 125, 128, 129. pi. 73-76, 82, f. 2, 86, f. 11,98 Sarcoyncca 43, 45, 88 Tacori, Clus. (91) Taxodium 123 THE YUCCEAE. 131 Xanthorrhoea 71 Yuca, Park. (72) ioliis Aloes, Bauh. (72) folila fllamentosis, Moris. (47) Perana, Gerarde. (72) Yucca 27, 28, 29, 42, 120-124. pi. 99. acaulls (114) acrotrlcha (114) acuminata (72, 74, 79) acatifolla (74) agavoldes (96) alba-splca (54, 57) X albella 115 albospica (57, 82, 105) aletrlformls (114) aloefolla versicolor (90) aloifolia (39) 45, 81, 82, 84, 86, 87, 88, 89, 90 (94) 110, 116, 122, 123, 124, 127, 129. pi. 46, 49, 84, f. 6, 95 arcuata 90, 92 Menandi 90 tenulf olia 90 conspicua 89, 92 Draconis 89. 91, 95 conspicua 89 flexifolla (92) marginata 89, 90 Menandi 90, 93, 127. pi. 50 pnrpurea 89, 90 quadrlcolor (91) roseo- marginata (91) stenophylla (88) tennlfolia 90, 93 tricolor 89, 91 varlegata (82, 90) Yucatana 90, 93 X Andreana 77 angustlfolla (54, 56, 60, 79, 82, 83, 114) elata (56) mollls (63) radlosa (56) strlcta (61,64) angustlsslma 45, 58, 126, 129. pi. 23, /. 1, 24, f. 1,83, f. 6, 93 arborescens (41, 89) arcuata (92) argospatha (96) argyrophylla (105) Arkansana45, 53, 54,55, 62, 63, 126, 129. pi. 30, 31, 83, f. 7, 92 armata (88) aspera (96) Atklnsl (90) australls 46, 100, 103, 108-9 (117) 119, 120, 1-28, 129. pi. 60, 61, 85, f. 2, 96 baccata46, 109 (113, 119) 121, 122,128, 129. pi. 68, 69, 86, f. 4, 97 Yucca baccata australls (103, 105-6, 119, 111, 113) clrclnata (103) fraglllfolla (103) gennlna (111) Hystrix (103, 106) macrocarpa (104, 110, 111, 117) periculosa (103) Hcabrifolia (103) Barrancasecca (114) Boerhaavii (80) Boscli (114) Braslliensis (.75) brevlfolla (41) 46, 100 (103) 127, 128. pi. 67, 68, 69, 96 ^ California (39, 95) canallculata (97, 105) fillfera (103) pendula (97) X Carrierel 74 circinata (103, 104, 106) concava (49) conspicua (92) constricia 15, 54 C56) 123, 123. ;..' _.'/, 21,f. 1, 83, f. 4, 92 contorta (67, 9fc) cornuta (82, 9 '.) crenulata (88) XDeleuili 67, 74 De Smetiana 45, 87, 127. pi. 43 X dracaenoldes 77 Draco (91) Draconis (88,91) arborescens (41) elata (56,58) X elegantlsslma 115 elephantipes 45, 71, 92, 94, 123, 127, 129. pi. 45, 51, 82, 84, f. 7 Ellacombel (75) X Elmensls 116 Engelraanni (39) X ensifera 79 ensllolla (80) exlgua (52) Eylesll (80) falcata (80) fllamentosa (39) 44, 46, 47 (48, 49, 60-53, 64) 81, 82, 83, 87, 116, 120, 122, 125, 126. pi. 8, 12, 89, 91 Antwerpensls (51) aurea elegantlsslma (48) blcolor (48) bracteata 47, 48, 125. pi. 9, 90 concava 47, 49, 84, 125, 129. pi. 10, 82, f. 1,90 flacclda (49) glaucescens (51, 82) grandillora (52) 132 MISSOURI BOTANICAL GARDEN. Yucca fllamentosa laevlgata (49) latlfolia (49) major (115, 116) maxima (48, 52) media 47, 49, 125. pi. 11 patens 47, 48. pi. 89 paberula (50) varlegata 47, 48 (77) flllfera(103, 104-6) flacclda 44, 49, 60, 51 (51) 83, 84, 116 122, 126. pi. 16, 91 exigua, 52 glaucescens, 50, 51, 126, 129, pi. 12-15, 17, 80, f. 2,83, f. 2 lineata, 50. grandiflora, 51, 52 exigua, 51 Integra, 51 Integra, 52 lineata 52 orchioides 50, 51 flexllls 45, 78, 79, 81, 83, 87 Boerhaavii 79, 80 enslfolia 79, 80 falcata (80) Hlldrethl 79, 80, 127. pi. 47, f. 2 patens 79, 81 Peacockil, 79 semicyllndrlca 79, 80 tortulata, 79, 80 folils Aloes (72) follls lanceolatis (47) folils margine integerrimis (72) lollorum marg. cren. (89, 92) fragilifolla (103, 104, 106) funifera (36, 38) Ghiesbreghtli (94) glgantea 42, 45, 71 glanca 45 (49, 52, 54,58) 59 (75). 82, 121, 122, 126, 129. pi. 23, f. 2, 24, f. 2, 25, 83, f. 9,93 mollis (63) stricta (55) 61 (64) 126. pi. 26, 27 glancescens (51,75) varlegata (78) gloriosa 42, 45, 72, 73, 74 (74-6, 79) 81, 84, 85, 87, 88 (95) 115, 116, 122, 123, 127. pi. 43, 44, 94 acuminata (72) elegans marginata (78) variegata (78) Ellacombei (75) glauca pendnla (116) glaucescens (75) longlf olia, 76, 82 macnlata 76 marginata (78) aurea (78) Yucca gloriosa medlo-plcta (74) medio-striata 73, 74 minor 73, 74, 80, 127. pi. 45 mollis (64,76) nobilis 75 parviflora75 obllqna 73, 74 planifolla (76) plicata 73, 74, 75, 82, 84 maculata 74, 76 superba 74, 76 prulnosa (81) recurvata (74) recurvifolla (76) fol. var. (78) robnsta 73, 74, 75 longifolia 73 nobilis 73 rofoclncta (78) snperba 76, 127, 129. pi. 46, f. 1, 84, f. 4 tortulata (80) tristis(77) varlegata (78) gramlnif olia (39, 114) Guatemalensls (94) X Guiglielmi 116 Hanburii (60) Harrimaniae 45, 59, 122, 126, 129. pi- 28, 29, 83, f. 10, 93 Harnckerlana (91) Helkinsi (87) horrida (114) X Imperator 116 integerrima (72) Japonica (106) X juncea 79 X laevigata 79, 82 Lenneana (94) XlHiacea (116) lineata lutea (91) longifolia (75, 79, 96, 115) Lonisianensis 45, 54, 62, 64,83, 126,129. pi. 32, 33, 34, 83, /. 8, 92 lutescens (67) macrocarpa 46 (98) 110 (113, 117), 117, 128-9. pi. 70, 71, 85, f. 5, 86, f. 2, 98 X magniflca 116 X margaritacea 116 X Masslllensls 79 Mazell (98-9) medio-picta (91) Meldensis (50,51) Mexicana(78, 92) Mohavensls 46, 110, 112, 113, 128, 129. pi. 72, 85, f. 6, 98 Mooreana (94) obllqua (74, 76) THE YUCCEAE. 133 Yncca orchioldes (51) major (51) Ortglesiana (39,41) Parmentierl (105) parvlflora (30) patens (81) pavillora (30) Peacockii (79) pendula (76, 82) aurea (78) varlegata (78) periculosa (103. 104, 106) Fernana (72) plcta (91) pltcairnlfolia (115) plicata (75, 82-3) glauca (76) plicatllis (75) polyphylla (57, 103, 104, 106) X praecox 116 Pringlel (43, 106) prninosa (81) pubernla (49, 100) qnadricolor (91, 93) variegata (91) xadiosa 45, 56 (58) 104, 117, 121, 126, 129. pi. 21, f. 2, 22, 83, f.6, 86, f. 1,93 recurva (76) elegantlsslma (78) recurvata (97) recurvifolia 45, 48, 64, 75, 76, 77, 81, 82, 83, 84, 86, 87, 122, 123, 127, 139. pi. 46, f.2,47,f.l,84,/.6,94 elegans 77, 78 marginata 77, 78 ruf ocincta 77, 78 tristls77 variegata 77, 78 revolnta (97) rlglda 45, 65 (67) 106, 126, 129. pi. 35, 36,f.l, 84,f.l, 93 Koezlil (94) rostrata 45, 68, 104, 126, 127, 129. pi. 36, f. 2, 40, 41, 42, 84, /. 5, 93 rubescens (115) rubra (74) rofoclncta (78) rupicola 46, 67, 83, 116, 127, 129. pi. 37, 38, 39, 84, f. 2, 93 rigida (65, 67) Yucca rupicola tortifolia (66-7) scabrlfolia (103, 104, 106) schidigera (113) Schottil 46, 98 (99, 100) 101, 103, 127, 128, 129. pi. 65, 67, 86, f. 1, 96 - Jaliscensiu 99, 120, 127. pi. 66, 96 semicylindrlca (80) serratifolia (115) serrulate (88) argent eo -marginata (90) splnosa (114-5) stenophylla (79, 114-5) Stokesl (91) X stria tula 79 strlcta (61, 64) elatior (64) intermedia (64) X sulcata 74 auperba 76 tenuifolia (93) tenulstyla 45, 63, 62, 126, 129. pi. 17 f.2, 18, 19, 83, f.S, 92 Tonellana (105) tortilis (67) tortulata (80) Trecnleana 45, 82, 83, 96, 97 (99, 103) 106, 112, 115, 120, 122, 127, 129. pi. 62, 84, f. 8, 95. canaliculate 97, 127. pi. 63, 64, 96 glauca (97) undulata (97) X Trelcascl 116 tricolor (91) nndulata (80, 97) valida 46, 107, 128, 129. pi. 62-67, 86, f. 3, 97 Vandervinnlana (96) variafolla (77) variegata (90) versicolor (30) Virglnlana (47) X virldlflora 116 X Vomerensis 116 Whipplei (39) glanca(39) graminlfolla (39) vlolacea (40) Yacatana 93 Yucceae 27, 28 Yuccoldeae 27 KEPT. Mo. Box. GARD., VOL. 13. HESPEBALOE ]' AHVIFLORA AND VAR. ENGELMANNL KEPT. Mo. EOT. GARD., VOL. 13. PLATE 2. HESPERALOE PARVIFLORA ENGELMANNI. KEPT. Mo. LOT. GAKI>., VOL. 13. HESPERALOE FUNIFKRA. KEPT. Mo. BOT. CARD., VOL. 13. PLATE 4. HESPERALOE AND HESPERYOUCCA. KEPT. Mo. Box. GARD., VOL. 13. PLATE 5. HESPEUOYUCCA WHIPPLEI. KEPT. Mo. EOT. GAKD., VOL. 13. PLATE 6. CLISTOYUCCA ARBORESCENS. KEPT. Mo. BOT. GAKD,, VOL. 13. PLATE 7. CLISTOYUCCA ARBORESCENS. KEPT. Mo. Box. GARD., VOL. 13. PLATE 8. YUCCA FILAMENTOSA. KKI J T. Mo. BOT. CARD., VOL. 13. YUCCA FILAMKNTOSA BRACTEATA. KEPT. Mo. BOT. GARD., VOL. 13. YUCCA FILAMENTOSA CONCAVA. KEPT. Mo. Box. GARD., VOL. 13. YUCCA FILAMENTOSA MEDIA. KEPT. Mo. Box. CARD., VOL. 13. YUCCA FILAMENTOSA AND Y. FLACCIDA. KEPT. Mo. BOT. GARD., VOL. 13. YUCCA FLACCIDA GLAUCESCENS. KEPT. >lo. BOT. GARD., VOL. 13. THERMOTROPISM OP YUCCA FLACCIDA. REI-T. Mo. EOT. GAKD., VOL. 13. YUCCA FLACCIDA GLAUCESCEXS. REFT. Mo. BOT. GA.KD., VOL. 13. YUCCA FLACCIDA. BEPT. Mo. Box. GAKD., VOL. 13. YUCCA FLACCIDA AND Y. TENUISTYLA. KEPT. Mo. Box. GARD., VO YUCCA TBNUISTYLA. KKPT. Mo. EOT. CARD., VOL. 13. YUCCA TEXUISTYLA. KEPT. Mo. IJOT. GA.KD., VOL. 13. PLATE 20. 1 YUCCA CONSTRICTA. KEPT. Mo. BUT. GAKD., VOL. 13. YUCCA CONSTRICTA AXD Y. RADIOSA. . Mo. BOT. GARIX, VOL. 13. YUCCA RADIOSA. REFT. Mo. EOT. GARD., VOL. 13. YUCCA ANGUSTISSIMA AND T. GtAUCA. KEPT. Mo. BOT. GARD., VOL. 13. PLATE 24. YUCCA ANGUSTISSIMA AND Y. GLAUCA. KEPT. Mo. Box. GARD., VOL. 13. YUCCA GLAUCA. KEPT. Mo. BOT. GAKD., VOL. 13. YUCCA GLAUCA STRICTA. KKPT. Mo. BOT. GAKD., VOL. 13. PLATE 27. YUCCA GLAUCA STRICTA. PLATE M. YUCCA HARRIM \N1 \l SEPT. Mo. BOT. CARD., VOL. 13. TLATE 29. YUCCA HARRI.MAXIAE. KEPT. Mo. Box. GAHD., VOL. 13. PLATE so. YUCCA ARKAXSANA. IIEVT. Mo. EOT- GAKD., VOL. 13. PLATE si. YUCCA ARKAXSANA. REFT. Mo. BOX. GARD., VOL. 13. YUCCA LOUISIAXENSIS. KEPT. Mo. Box. GARD., VOL. 13. k YUCCA LOUISIANENSIS. KEPT. Mo. Box. CARD., VOL. 13. PLATE 34. YUCCA LOUISIANEN8IS. EEPT. Mo. BOT. GARD., VOL. 13. PLATE 35. YUCCA RIGIDA. KBIT. Mo. BOT. GARD., VOL. 13. PLATE 36. YUCCA BJGIDA AND T. ROSTRATA. KEPT. Mo. EOT. GARD., VOL. 13. PLATE 37. YUCCA RUPICOLA REFT. Mo. EOT. GARD., VOL. 13. YUCCA RUPICOLA. KEPT. Mo. BOT. GARD., VOL. 13. YUCCA RUPICOLA. KEPT. Mo. BOT. GARD., VOL. 13. PLATK 40. YUCCA KOSTRATA. KEPT. Mo. Box. GAHD., VOL. 13. PLATE 41. YUCCA ROSTRATA. REFT. Mo. EOT. GARD., VOL. 13. PLATE 42. YUCCA ROSTRATA. KEPT. Mo. EOT. GARD., VOL. 13. PLATE 43. YUCCA GLORIOSA. KEPT. Mo. Box. GARD., VOL. is. YUCCA GLORIOSA. KEPT. Mo. Box. GARD., VOL. 13. m*s -.,.w , YUCCA GLOKIO8A MINOR. KEPT. Mo. BOT. GARD., VOL. 13. YUCCA GLORIOSA AND Y. RECURVIFOLIA. KEPT. Mo. EOT. GAKD., VOL. 13. PLATE 47. YUCCA RKCrKVTFoLIA AND Y. FLEXILIS. . Mo. HOT. GAKD., VOL. 13. PLATE 48. " YUCCA DE SMETIANA. KEPT. Mo. EOT. GA.RD., VOL. 13. PLATE 49. YUCCA ALOIFOLIA. KEPT. Mo. HOT. GAED., VOL. is. PLATE 50. YUCCA ALOIFOLIA MENANDI. KEPT. Mo. Box. GAKD., VOL. 13. YUCCA ELEPHANTTPES. KEPT. Mo. BOT. GAKD., VOL. 13. PLATB 52. YUCCA TRECULEANA. REFT. Mo. Box. GAED., VOL. 13. YUCCA TRECULEAXA CAXAUCULATA. REFT. Mo. Box. CARD., VOL. 13. YUCCA TRECULEANA CANALICULATA. KEPT. Mo. BOT. GARD., VOL. 13. PLATE 65. YUCCA. SCHOTTTI. REFT. Mo. BOT. GARD., VOL. 13. YUCCA 8CHOTTII JALISCENSIS. KEPT. Mo. EOT. CARD., VOL. 13. YUCCA BREVIFOLIA. REFT. Mo. EOT. GARD., VOL. 13. PLATB 58. YUCCA BREVIFOLIA. REFT. Mo. BOT. CARD., VOL. 13. YUCCA BREVIFOLIA. REFT. Mo. EOT. GARD., VOL. 13. YUCCA AUSTBALIS. KEPT. Mo. EOT. GARD., VOL. 13. YT7CCA AUSTRALIS. KEPT. Mo. Box. CARD., VOL. 13 YUCCA VALID A. KEPT. Mo. Box. GARD., VOL. 13. PLATE 63. YUCCA VALIDA. REFT. Mo. EOT. GAUD., VOL. is. PLATE 64. YUCCA VALIDA. KEPT. Mo. Box. GAKD., VOL. 13. YUCCA VALIDA. KEPT. Mo. EOT. GAUD., VOL. is. PLATE C6. YUCCA VALIDA. REFT. Mo. EOT. GARD., VOL. 13. PLATE 67. YUCCA VALIDA. REFT. Mo. Box. CARD., VOL. 13. YUOCA BAOCATA KEPT. Mo. EOT. GARD., VOL. 13. PLATE C9. YUCCA BACCATA. KEPT. Mo. EOT. GAKD., VOL. 13. YUCCA MACROCARPA. KEPT. Mo. EOT. GAKD., VOL. 13. PLATE 71. YUCCA MACROCARPA. KEPT. Mo. Box. GARD., VOL. 13. PLATE 72. YUCCA MOHAVKNSI8. . Mo. EOT. GAKD., VOL. is. PLATE 73. SAMUELA FAXOXIANA. KEPT. Mo. Box. GARD., VOL. 13. PLATE 74. SAMUBLA FAXONIANA. REFT. Mo. EOT. GARD., VOL. 13. PLATE 75. SAMUELA FAXONIANA. KEPT. Mo. EOT. CARD., VOL. 13. PLATE 76. 8AMUELA CARNKROSANA. KEPT. Mo. BOT. GARD., VOL. 13. PLATE 77. 8AMUELA CARNKROSANA. REPT. Mo. Box. GARD., VOL. 13. PLATE 78. SAMUELA CARNEROSANA. KEPT. Mo. EOT. GARD., VOL. 13. PtATE 79. SAMUELA CARNEROSANA. REFT. Mo. Box. GAKD., VOL. 13. PLATE 80, SAMUELA CARNEROSANA AND YUCCA FLACCIDA. REFT. Mo. Box. GAKD., VOL. 13. SAMUELA CARNKROSANA. KEPT. Mo. EOT. GARD., VOL. 13. YUCCA ELEPHANTIPES AND SAMUELA FAXONIANA. UEIT. Mo. HOT. GARU., VOL. 13. eft* ** ( tftfeft* ^~~^ AfttffttfA M -KKI'S UK YUCCAS. KEPT. Mo. HOT. GAKD., VOL. 13. ** *tt ***<* *** ttttt* 8KKD.S OF YUCCAS. KEPT. Mo. HOT. GARD., VOL. 13. ft : c D c L "- C fe t<& ^-j /' ' & T&J ^~ tt< SBEDS OF YUCCKAK. KEPT. Mo. EOT. CARD., VOL. 13. PLATE 86. GERMINATION OF YUCCEAK. REFT. Mo. EOT. CARD., VOL. 13. PLATE 87. GERMINATION OP YUCCKAB. KEPT. Mo. BOT. GARD., VOL. 13. PLATE 88. 1. HESPEROTUCCA. 2. HSFERALOE PARVIFLORA. CLISTOYUCCA ARBORESCEX8. DISTRIBUTION OF YUCCEAE. KEPT. Mo. Box. GARD., VOL. 13. TUCCA FILAMENTOSA VERA. TUCCA PILAMENT08A PATENS. DISTRIBUTION OF YUCCKAK. EEPT. Mo. EOT. GAHD., VOL. 13. PLATE 90. YUCCA FILAMENTOSA BRACTEATA. YUCCA PILAMBNTO8A CONCAVA. DISTRIBUTION OF YUCCEAK. REFT. Mo. BOT. GARD., VOL. 13. YUCCA FILAMENTOSA and varieties. TUCCA FLACCIDA. DISTRIBUTION OF YUCCKAE. KEPT. Mo. Box. GA.RD., VOL. 13. I-LATE 92. 1. YUCCA LOUISIANEN8IS. 2. T. TENUISTYLA. 1. YUCCA CON9TBICTA. 2. Y. ARKANSANA. DISTRIBUTION OF YUCCKAK. KEPT. Mo. Box. GARD., VOL. 13. PLATE 93. 1. YUCCA ANGUSTISSIMA. 2. T. GLAUCA. 3. T. RADIOSA. 4. T. HABRIMAXIAK. 1. T. RUPICOLA. '2. T. RIQIDA. 3. T. RO8TRATA. DISTRIBUTION OF YUCCKAB. KEPT. Mo. BOT. GARD., VOL. 13. PLATE 94. YUCCA GLO1UOSA. a -5 f * \ M * T i N.BAK. ,. 1 \ { J , . U W 1 . TPCCA RBCURVIFOLIA. DISTRIBUTION OF YUCCBAK. REPT. Mo. Box. CARD., VOL. 13. YUCCA ALOIFOLIA, in the United States. YUCCA TRECUIJSANA. '2. Y. TRKCULEANA CANALICULATA. DISTRIBUTION OF YUCCEAK. KEPT. Mo. DOT. GARD., VOL. 13. PLATE 96. HESPEKALOE FUNIFERA. 2. YUCCA SCHOTTII. 3. T. SCHOTTII JALISCENSIS. fs~- r ^J^ *v_ ^- 3* ^tn ^ i.OFfAUFl