OBSERVATIONS ON CISSUS QUADRANGULABIS. 247 



on a cursory examination, I took to be a variety of C. quadrangularis, 

 Roxb. PL Ind., the Vitis quadrangularis. Wall, (in Wight and Arnott's 

 Prodromus). If however the descriptions in these works are perfectly 

 correct, it can scarcely be that species. The Guzerat plant is invariably 

 wingless, the fruit ovoid and pointed (not globose), twice as large as a 

 pea ; and although it leaves a slight biting sensation on the tongue, still 

 it is not " very acrid/' as described by the authors alluded to, and, as we 

 are informed by Duchesne, in his 'Plantes Utiles, 5 is eaten in Senegambia 

 under the name of " Raisin de Galani," which would not be the case 

 if the fruit were " very acrid." Further, Roxburgh describes the roots 

 as fibrous, Willdenow and De Candolle as tuberous, a discrepancy 

 which, with those already mentioned, leads to a suspicion at least that 

 two distinct species of Cissus are confounded under the name of Cissus 

 quadrangularis, even after making due allowance for a considerable 

 amount of variation induced by differences of climate. In the older 

 descriptions of this plant by Linnaeus, Forskael, Persoon, etc., there is 

 no mention of wings to the stem, while in Roxburgh's c Flora Indica' 

 and Wight and Arnott's • Prodromus ' the wings are a prominent part 

 of the diagnosis, while the tubers on the roots are not mentioned at all, 

 that is, virtually denied. There is no reason to think that any of these 

 distinguished authors have described the plants before them incorrectly, 

 and therefore we are forced to adopt one of two conclusions, either that 

 two distinct species are included under one name, or that the influence 

 of climate in this particular instance deprives a plant of what have been 

 regarded by excellent botanists as essential specific characters. I am 

 inclined to the former of these, and if my surmise is correct, the two 

 species may be distinguished as follows. 



Cissus quadrangularis* Linn. Mant. 39 (non Roxb.); glaberrimus, 

 scandens, radice tuberosa, caule herbaceo dichotomo obtuse quadran- 

 gulari tumidiusculo articulato, articulis constrictis, stipulis adnatis 

 rotundatis bi-auriculatis caducis, foliis petiolatis crassis carnosis cor- 

 dato-rotundatis integris vel saepius trilobatis margine serrato-denti- 

 culatis, umbellis in pedunculo dicbotome ramoso abbreviate pauci- 

 floris, fructu ovoideo lam rubro 1-spermo, semine subconformi 

 Willd. Sp. PL i. p. 657. Saelanthus quadrangularis, Forsk. del 



33. Ic. t. 2. 



rainulorum 



e est eadcm 

 J.D.H. 



