mo 



of Blieede's and E tret's figures until the remaining four plants 



have been disposed of. 



Of the four plants in question two are from the East Indies as 

 understood by Linnaeus; the other two are West Indian species. 

 The former are (1) a Ceylon plant collected by Hermann and 

 published on his authority as Rasa-kinda or Rossa-kinda, which 

 forms the basis of one of the Linnean citations (Flor. Zeyl. 

 n. 358) under D. sativa in 1753 (Sp. PL ed. 1. p. 1033) ; and (2) 

 an Amboyna plant figured by Bunipli as Olus sanguinis, which 

 was referred by Linnaeus to D. sativa in 1754 (Amoen. Acad. 

 vol. iv. p. 133). Both of these Asiatic plants exclude themselves 

 from D. sativa , however it may ultimately be decided to apply 

 that name, because neither of them belongs to the genus 

 Dioscorea. The Ceylon plant is a species of Tinospora (Trim. 

 Joum. Linn. Soc. Bot. vol. xxiv. p. 151); the Amboyna plant is 

 a Cardiopteris (Blume, Rumphia, vol. iii. p. 207. t. 177). The 

 two West Indian plants are both members of the genus Dioscorea ; 

 they are (1) a Martinique wild plant first recorded by Plunder 

 and named by him Dioscorea scandens foliis tamni, fructu 

 racemoso: and (2) a Jamaica cultivated plant named by Sloane 

 Voluhilis nigra radice tuberosa compressa maxima digitata fari- 

 nacea esculenta folio cordato nervoso. With the earliest citation 

 of the Martinique species in 1703 Plumier published a plate 

 showing the fruit (Gen. p. 9. t. 26); the main-figure by Plumier 

 of the same plant was published in 1755 by Burmann (PI. Amer. 

 Plum. t. 117. fig. 1). The actual specimen which is the basis of 

 Sloane's diagnosis of 1696 (Cat. Jam. p. 46) is still in the Sloane 

 herbarium. A comparison of the material in herbaria tallying 

 with the figure by Plumier and the specimen of Sloane respec- 

 tively, shows that the Martinique plant twines to the left and 

 has reflexed capsules, whereas the Jamaica plant twines to the 

 right and has forward-pointing capsules; the leaves of the 

 Jamaica plant, moreover, while very often alternate on secondary 

 branches are opposite on the main bines and on most primary 

 branches, those of the Martinique plant are always alternate; 

 the flowers, too, differ considerably, those of the Martinique plant 



have membranous bracts and bracteoles and a membranous 



perianth with only three fertile stamens alternating with three 

 staminodes, the anther-cells being discrete,* whereas those of the 

 Jamaica plant have firm almost hyaline bracts and bracteoles, 

 a thick and somewhat fleshy perianth with six fertile stamens, 

 the anther-cells being contiguous. There is therefore no ques- 

 tion that the two West Indian plants included by Linnaeus under 

 D. sativa belong to two distinct species. 



The species figured by Plumier was transferred by Willdenow 

 in 1806 to /). piperifolia. Humb. and Bonpl. AVilldenow's treat- 

 ment was followed by Poiret in 1813 (Encyc. Meth. Suppl. 



Itie original Plumierian drawing (MSS. tom. if. n. 144) includes some at 

 o^ ™! e anal >' tlcal details contained in Plumier \s plate of 1703 (Gen. 

 t 26) These were not mad* use of by Burmann (PI. Amer. Plum. t. 117, 

 tig. J) possibly owing to their having already been published. In the 

 original drawing, underneath the figures of th'e male flower, we find the 

 legend ' fleur blanche, en clochette, decoupee en 6 parties, 3 etamines a 

 double teste. 



