90 ANNALS OF RORAL BOTANIC GARDEN, CALCUTTA. 
incised, lacinule unequally serrate; serratures callose-apiculate; upper leaves sine 
less divided; uppermost often reduced to a small, ovate-lanceolate, very acute езіп Е 
Flowers unknown. Infructescence racemose, 20—30 ст. long. Вгасіз qs m р 
of follicles about 6, lower consisting of 3, upper often of / or и ollicle. 3i 
duncles elongating, at first about 5, finally 10—15 mm. long, а ч و‎ = 
terete, torus forming a small knob. Follicles stipitate; stipes 2. 8 ES ed, : 
finally 10—12 mm. long; follicle proper cylindric, attenuated into the we no 
by the short persistent style, without stipes 3—4 em. long, membranous, finely ration айе; 
sutures narrow, not prominent; persistent style little more than 1 mm. long, stigmatic 
surface thick, extending along the ventral suture, slightly recurved at the apex, seeds 
numerous, obovoid. 
Eastern Sikkim: at Nütat; Phari (Dungbu ?). 
Flowers just fallen in July; fruit ripens in August. 
This species agrees with Coptis brachypetala, Sieb. et Zuce., in the biternately divided leaves 
with pinnntifid segments, the partitions being again pinnatifid or ineised-serrate, the serratures being 
acute. The authors state, however, that this specios has long pedicelled flowers and that “ habitus 
omnino ut in 0. asplenifdit, DC., sed segmenta foliorum profundius incisa, lobis plerumque 
augustioribus et minoribus." In our species the lobes are much larg Ы than in C. aspionijolia ; in fact т 
regards the vegetative parts it stands nearer C. Tecta than C. asplenifolia. From both it differs widely 
in the size and shape of tho follicles. 
C. anemonefolia, Sieb. et Zucc, differs in the leaves being ternatisect instead of biternately divided, 
andthe peduncles being twice as long as the fruit. 
Indeed, an exam nation of the figures in Phonzo Zoufou, Vol. VII, proves clearly that our plant 
does belong to neither C. anemonefolia nor to С. brachypetula. It is, however, very closely related to 
Coplis orientalis, Maxim. To judge from Maximowiez’s description and Phonzo Zoufou's fig. tab. II 
versa, which Professor Huth doubtfully refers to C. orientalis, the leaves of the two species are nearly identical. 
A further point of agreement between the two consists in the size of the follicles and the relative length 
of stipes and follicle proper. Professor Huth states that the follicle of C. orientalis is about 30 mm. long 
of which nearly 10 mm. are accounted for by tho stipes. The longest follicles in our species are about 
45 mm. long, the stipes being 12—15 mm. The only essential difference between C. ospriocarpa and 
C. orientalis rests on the comparative length of fruit and pedicel, Maximowiez lays considerable stress 
on the relative length of fruit and pedicel. For he says :--“О. anemonefolia a С. orientali differt 
pedunculis fructum ad summum duplo, in nostra (C. orientali) triplo quintuplove superantibus. In 
C. ospriocarpa the longest fruit-bearing pedicels are 17 mm. long, and about one-third the length of 
| а stipita"e follicle, but usually they are only 3 or less. С. orientalis appears to have five or more carpels 
in one head. Our species has never more than three. There may, of course, exist other differences in the 
parts of the flower. 
Prate 115. Coptis ospriocarpa, Р. Brühl, fruiting specimen from Sikkim ; 1, very immature follicle ; 2, ovary eut open ; 
3, seed; 4, scale from below inflorescence. ' 
DELPHINIUM. 
| who unites the two. On the one hand the posterior sepal of the extreme forms of Aconitum Lycoctonum 
is truly spurred, whilst in Delphinium Hohenackeri and allied forms it is more helmet-shaped than in many 
wn aconite. The only character which can be used to separate all the forms of Aconitum from those of 
developed claw in the nectaries, the claw being nover shorter than 
