" T T i T T n "
9 2 A K K Ä L S OF ROYAL BOTAKIC GARDEN, CALCUTTA.
AB speoimens witli ripa fruits are yot rather Bcaroe, partly in oonsequenoe of the tendency with
p a i d collectors and amateur botanists to negleof; gather ing fruits and ripe seeds, partly from the ueoesaity
of leaving less nooessihle regions before the fruits have attained maturity, it is not yet possible to establish
i a all cases the limits -within whioh the structure of the seed has to be considered as variable. Por
instance, it would be very interesting to know whether of two species like Delphinium lirunomamim
and Dclphhmm Jncqaemontianim, whioh agree with each other neai'ly in every detail, one is a direct
o f f s p r i n g of tbe other, or, what in some respacts is more probable, whether they constitute corresponding
members of two parallel series, one with its winged seeds a deseendent oí D. chtum, the other a
modificaiiou of D. spcciosum. There is no genus among Raniinculaccce, except perhaps Isnpyrum, where
a n accurate knowledge of the structure í>í the seeds is of saoh importance for a correct discriminat
i o n of species as Delphinium, including Aeonitnm. Fortunately the collections of Indian plants made
since thi publication of the first volume of the Flora of British India have furnished the writer
with siitBoient material to settle several doubtful points and to assign to certain forms their tight place
b y making us© of characters derived from the seeds. Thus it has been found that D. ramneulifoliim
of Wallich is not a variety of D. chtum, but belongs to the group of D. fpecio sum; that D. Jncquamont'cnum
cannot be referred to D, Brunoninmm, but has either to be considered a subspecies of
J), liaihmiyinmim, or taken as on independent species; tliat the Khasian forms of D. altis'iimam must
b e separated from the Himalayan forms and raised to the rank of s species, to which the name of
J). stopeHiisiiitim has been given by the writer.
But although a detailed and rather laborious investigation has led the author to the oonchision
t h a t it is always unsafe to separate or to unite two otherwise closely allied forms unless theii' seeds be
known, too much importance should not be attributed to difEerences in minut e details of their structure.
T o cite two examples:—1 ) . dimiicalum., Ledeb., is stated in Boisder's Flom Oriniialis to have seeds
with naiTow continuous wrinkles, whilst Hohenacker's specimens from Georgia have seeds with deeply
notched, nearly sepai-ate scales. J), riigiilomm, Boiss,, is distinguished from IJ. peraicum, Boiss., by the seeds
o£ the former having "distinct adpressed scales," whilst the seed-scales of the latter are "abbseviated and
n n i t e d into transverse wrinkles." As a matter of fact the case is sometimes revtrsed, as an inspection of
figs, 11—14 on plate 119 will show, where 11 and 14 are ñ-om spenimens of the true D. pcrsiom, whilst
12 and 13 are from plants which otherwise are good examples of D. ru^idosiim. Drawings of the
fieeds of praofically all the Delphinia from the region which is here considered are found an plates 119,
121 and 1"¿7. The only Indian spec'.es of which the seeds are yet unknown is D. pachyventrum, with
is more wes'ern subspecies ieangense; and this belongs very probably to the pferosjiermous larkspurs.
Of no less importance than in Delphinium are the seeds in Acoiticiim, a point overlooked by Regel
i n his excellent enumeration and definition of Russian aconites contained in "Reisen in dem Süden
\ o n Ost-Sibirien." The seeds of the various spe¿os of Jconiium are sufficiently well known to utilise
tliem in woiking out a classi Ii cation, of which the following is an attempt:.—
Section I.—LYuocToxoiDEa;- Posterior sepal with a cylindrical, club-shaped, conical or spur-like,
p r o t u b e r a n c e ; seeds transversely rugose; flowers yellow, purplish, or red.
Section II.—CAM MARO IDE^. Postsrior sepal more or less \-aidted, sometimes thick-cyl in d i j c and
i n c u r v e d ; seeds lamellate-rugose; flowers blue or variegated, rarely daik-recl, purplishbrown,
yellowish or white.
Section III.—ÍÍAREI, 1,011)7,.®. Posteriors epal more or loss vaulted, stcds three-sided, pyiamidnl,
more or less winged along the edges, otherwise smooth ; flowers usually blue, or variogaled,
r a r e l y yellow.
ESAMPI.ES,—I. Lycoctonoidese : A. J/!/coctoniim with its innumerable varieties.
I I , Cammaroideje: A. tnrieyaUan, Linn.; A. par.iculahim, Lam.; A. 1-üchcri, Rchb.; A. colulile,
P a l i . ; A. uiieinatu'ii, Linn,; A. Jcrvx, Wall.; A. hridum, H. f. ot T.; A. gymnandrum, Maxim,
I I I . NapelloLdeas: Bubsection 1.—Napello»: sepals deciduous, blue or variegatod with white: A.
iiapellus, Linn.
9 3
d on the acc-om'
DESCRIPTIONS OP NEW AND BARE INDIAN PLANTS.
Subsection 2.—Anthore»: sepals mai'cescent, yellow: A, Anthotn, Linn.
For convenience of reference we append a list of the species of which seeds are
p a n y i n g plates ;—
PFCATB 13, D.unchialam pseadovestilum.
PLATE 117,/^. 10 (a), (b), fe) D. saniculifolium ref,aclum; (D) D. dcnuduium.
J'LATB 318, fig. 10. D. grandiJUrum, (a) var. paradox«, (b) rar. Linnad, (c) 7ar. bhotanica.
PLATE 119, fig. 11. D. persicum, Sliiraz.
12. B. rugùloeum. Mount Slicbo, Belucbistan.
13. D. rugulosum. Hari-rud.
34. D-pei-simm, var. arxaneme P.B., Slicbo.
15. D pcrsuniH, Tar. Stocksianam, Belucliistan.
16. D. persicum, var. afghaniam, Afghauistan.
17. D. incanutì^. Kaslimir.
18. n e «amo.
19. D. denudatum, N. W. Him.
20. D. tanimlifolium, var. kokatensis, "K..,liat.
21. D. vncinatum. Hazora.
22. D. tpedosum Biob. suljsp. ranuncuUfolium, rar. oinophjlUm. Kashmir.
'¿3, D cushmirianum, Laliul.
24. D. Jacqucmontianum. (a) Bashalir, (b) Dhauli valley, Kumaun.
25. H. Brawnianum, Kashmir.
26. D. demifiorum, Baslialir.
27. D. viuosum. Sikkim.
28. D. ««iii:,«, N.-W. Him.
29. D. iiupo/wsnitim, Khasia.
30. D. inci.»«,, Nipal.
31. D. glaciale, Sikkim.
32. D. gravdijlomm, war. Ihotanica, S. E. Tibet.
33. D. diclyocaipum (unripe).
S4. D. da^ycaulon, W. Gliuts.
PHTK 121,/ff. JO. Z>. siapelxosmum, Kbasia.
11. D. aliissimam, E. Nepal.
12. D. cheilaatham schizofhyllam, Kumaon.
PHTB 122,/j. 20. D. deneiflorum, Lebung puss,
PLATE I z i . f i j . 1!. D. vht-usum, Sikkim.
VUTZ m, Jig. 10. B.^^domm. subspems ranunculifolU^m. (a) var. ; (b) rar. („) rar
oinophi/Ua. '
W e proceed now to tbe consideration of those species which are figured on plates
1 1 6 — 1 2 5 , 127, 126.
D E L P H I S I U I I DSCIN.^TUJI, H . / . et T.
A pcpennid herb. Ehizonis d o a c e n dmj or vertical, subcylindrio, 4—8 mm thick
c o v e r e d ml h old leaf-shcaths at the ape.t. Stem solitary or double, simple or scarcely
b r a n c h e d , terete, subflemose to geniculate, 1 2 _ i 3 cm. high, near the base 3—4 mnj
t h i c k , ioternodes very short noar t],e base, upwards 1 - 4 em. long, entirely glabrous or
I t e m a glabresoenl base u pwa r d s densely velvety ivith spreading hairs of a length o(
1—2 mm. Leaves long.petioled, crowded at the base ; petiole« slender, subcannliculate
d i l a t e d at the baso into a lanceolate parallel-velnod sheath, two to three times as Ion»
a s the lamina, glabrous or somewhat hairy; leaf-blade trisect to the base, 1-Ó to 6 om°.
l o n g , 3'0 to 8 cm. broad, glabrous or villose especially w h e n yoiing ; segment s cnneate at
t h e base, lateral ones unequally bisect with the e.Uernal divisions often deepl y bipartito
d i v i s i o n s lanceolate in outliiie, divergent, altoruatoly or suboppositely pinnatsly incised,'