DESCRIPTIONS OF NEW AND RARE INDIAN PLANTS. 105 
specimen of var. chrysctrichum with the subglabrousleaves and the (sometimes) glabrous ovary of var, 
ciliatum. Ав already mentioned, var. chrysotrichum has sometimes trifid ortrierenate braeteoles ; and with 
the large trilobate bracts occurring in some forms of D. viscosum it would not be astonishing if the bracteoles 
should here and there assume a similar form. The locality (probably the Jongri district) and elevation 
agree well with the supposition that D. trilobatum is probably only an aberrant form of var. ciliatum. 
Prate 123. Dhelphinium viscosum: Н. f. and Th. 1, var. ciliata, upper part of flowering stem; 2, var. gigan- 
tibractea ; 3—7, var. ciliata; 3, nectariferous staminode; 4, petaloid staminode; 5, base of the same; 6, stamens; 
7, pistil; 8, pistil of var. gigantibractea ; 9, stigma; 10, follicle of var. ciliata; 11, seeds of the same. 
DELPHINIUM speciosum М. Bieb., subsp. RANUNCULIFOLIUM. 
Provided Turezaninow assigned to the term lamellato-rugosus the same significance that is attributed 
to it by Boissier and other botanists, he would have been the first to draw attention to the fact that two 
species, very similar in most respects, but radically differing in the structure of the testa of the seeds, have 
been confounded under the name D. elatum, Шап. Ор page 75 of his Flora baicalensis he enumerates 
among the diagnostic characters of D. elatum, “semina lamellato-rugosa lamellis imbricatis" ; and in a note 
on page 76 he says: “ D. cuneatum Stev. semina habet compresso-triquetra angulis alstis et hoc charactere a 
varietatibus D. elati bene distinguitur." It is, however, possible that Turczaninow did refer to seeds which 
are puckered into true transverse wrinkles, as often seen in D. grandiflorum, D. cheilanthum, D. stapeliosmum, 
and especially in various forms of the real D. elatum. Boissier ascribed at first to D. flecuosum, M. Bieb., 
D. dasycarpum, Stev., and D. speciosum, М. Bieb. * semina ad angulos alata”; and it was Ruprecht who first, 
in his Flora Caucasi, showed that the last-named forms have lamellate, rugose seeds, and that it is precisely 
this character by which the group of D. speciosum is sharply distinguishable from that of D. elatum. 
To the former group belongs also the origial D. ramunculifolium of Wallich, in which the epidermis of 
the seed is thrown into transverse wrinkles at an early stage of development, these wrinkles gradually 
developing into imbricate scariose scales like those seen іп D. kashmirianum, D. incanum, or D. camptocarpum. 
Some of the forms, however, included under the varietal name of D. ranunculifilium in the Flora of 
British India are true Delphinium elatum; as, for instance, the D. Hoffmeisteri of Klotsch, which should go to 
form a subspecies of D. elatum co-ordinate with the D. incisum of Wallich. The Indian varieties of 
D. ranunculifelium differ so much in the nature of their indumentum, in the size of the flowers, in the 
position and shape of the bractecles, and in the size of the follicles and tho relative length of spur and sepals, 
that the reduction of D. fleruosum, M. Bieb., D. dasycarpum, Stev., and perhaps of D. formosum, appears 
to be unavoidable. D. Poltaratskii, Ruprecht, is known to the writer only from Ruprecht’s description, 
but it seems closely related to some forms of JD. ranuneulifolium. Dr. Huth had the kindness to point 
out to the writer, who has verified the statement, that the sepals of D. speciosum are hairy on the inside, 
and that this species can thereby be distinguished from allied forms. And indeed all the specimens of 
D. ranunculifolium examined by me share that characteristic with JD. speciosum. The fact, however, 
that in D. incisum, Wall, the sepals are either all of them glabrous, or all of them pubescent, or some of 
them glabrous and some pubescent, on the inner surface, appears to throw doubt on the diagnostic value 
of that character. 
We subjoin a detailed description of D. ranunculifolium, (sp. of Wall.) as far as this form is known 
at present. _ 
A perennial herb. Rhizome descending, subcylindric, with filiform rootlets. Stem 
erect, straight or flexuous, terete or angular, striated or sulcate, fistulose, when flowering 
30 to 100 cm. high (sometimes probably even higher), 4—8 mm. thick near the base, 
glabrescent below, upwards either subglabrous, or hirsute with shining ЧеНехе or 
spreading longish hairs; floral axis usually densely pubescent with white or yeliowish 
crisp hairs usually intermixed with glandular hairs. Branches patulous, the lower ones 
sometimes 40 cm. long. Leaves from very long-petioled below to short-petioled upwards, 
with slender, beneath prominent, nerves; petioles slender, hardly dilated at the base 
Axx. Вот. Вот. Garp. Оліс., Vor. V. 
