200 FLORA INDICA. [Menispermacea . 



shape of the male sepals and of the cup of the corolla, the shape of the bract in the 

 female inflorescence, and the number, size, and degree of hairiness of the flowers, 

 are all extremely variable. The shape of the sepal of the female flower, we thought, 

 might afl'ord characters of importance, but it may be seen to vary to a great extent 

 even in the different flowers in the axil of the same bract, not only in size, but in 

 shape, being sometimes spathulate obovate, sometimes broader than long, and quite 

 entire, at other times emarginate, or even bipartite to the base ; it is either very 

 fleshy or almost membranous, nerveless, or one- two- or three-nerved, with or with- 

 out red oblong dots (intercellular spaces full of a coloured juice), hairy, or almost 

 glabrous ; the bract is equally variable in absolute and relative size. 



An examination of many hundred flowers having shown that no reliance can be 

 placed on characters derived from them, and the shape of the leaves being mani- 

 festly of no value as a character, we have been compelled to conclude that the 

 American and Indian plants are not distinct. Poiret had already anticipated us 

 in this conclusion; and though some of the synonyms which he quotes are un- 

 doubtedly erroneous, it is evident that the Indian plant which he declared so posi- 

 tively to be the same with the Pareira and Caapeba of America, was the 0. con- 

 volvulacea of Willd. and Wight and Ai'nott, and that his mistake lay, not in this 

 conclusion as to the specimens before him, but in supposing that these produced the 

 Oocculus berries of commerce. 



We have carefully compared a very extensive series of specimens of this species 

 from India, Africa, and America, and can with confidence declare that many Ameri- 

 can and African specimens are identical with others from India. There are, no 

 doubt, one or two forms from America for which we have not been able to find exact 

 representatives among our Indian specimens; but their differences are so trifling as 

 to be quite within the limits of variation in this very variable Order. 



It will be seen that we have not quoted many synonyms of American authors. 

 We believe the number might have been very much extended, but the want of au- 

 thentic specimens has prevented us from enlarging the list. The characters dwelt 

 upon for the discrimination of the numerous species described are notoriously variable 

 in all the species of this genus, and we think it very doubtful if more than one scan- 

 dent species exists in America at all liable to be confounded with C. Pardra. 



Cissampdos Voyelii, Miers, from tropical West Africa, is perhaps distinct, the 

 shape of the drupe being much more elongated. The male specimen, which Mr. 

 Miers considered to belong to that species, we have above referred to C. Pareira. 

 C. Vogelii is also an American plant; at least specimens of C.fasciculata, Bcntham, 

 which is perhaps the same as (J. denudata, Miers, and is only a tomentose state of 

 C. andromorpha, DC, are undistinguishable from it. 



Coccuhis niemhranaceus, Wallich, is a curious diseased state of G. Pareira^ in 

 which the branches ai-e covered with multitudes of little pale-coloured leaves. It is 

 not uncommon in shady jungles in the damper parts of the Himalaya. 



13. CYCLEA, Arnott. 



Cyclea et Rhaptomeris, Miers. 



Mas. Sepala 4-8 in calycem oampanulatum vel inflato-subglobosum 

 coalita. Petala totidem, plus minus coalita. AntJierce liorizontales, 

 sepaKs numero asquales, unilooulares, transverse dehiscentes. Pcem. 

 Sepala 3, lateralia, bractea antica sufFulta. Ovarium solitarium, anti- 

 cum. Stigma in segmenta 3-5 subulata radiatim divergentia fissum. 

 Drupa subglobosae. Putamen hippocrepiforme, dorso varie tubercula- 

 tum, lateraliter convexuin, baud excavatum, sed intus loculos 3 vacuos 

 iis Limacia similes continens. — Frutices scandentes, inflorescentia joawi- 

 culala axillari. 



