•373 



■ 



unfortunate necessity for its employment, sufficiently accounts 

 for the introduction of this species to Guiana and to the Antilles, 

 where it extends from Trinidad, St. Vincent and Martinique to 

 Porto Rico. 



In spite of the identity of Smilas acuminata, Willd., and 



I D. chandrocarpa, (Jriseb., the latter name was still the 



one 



recognised by Fline in 1897 (Engl. Nat. Pfianzenfam. Nachtr. 

 II. -IV. p- 84) and although -we are aware that the name hist 

 used in 1842 cannot be sustained, we do not propose here to 

 suggest the new combination which the identity of these two 

 plants seems to call for, owing to the circumstance that it is not 

 impossible that in the 4 Boyaux du Diable ' we may have only 

 to deal once again with the Polygonatum of Plumier which was 

 in 1789 made by Lamarck the basis of his D. altissima. An 



examination of the two plates published by Burman (PL Amc/\ 



Plum. t. 83 and t. 177. fig. 2) shows that Smilax acuminata and 

 IK altissima have the same foliage; a comparison of the two 



| descriptions by Plumier shows that they have the same under- 



ground stem and the same habit. One very serious difficulty 



p is that in D. altissima as figured by Plumier the leaves are 



opposite. The numerous specimens studied by us show that in 

 Smilax acuminata (D. chondrocarpa) the leaves occasionally are 

 opposite or nearly so, but they are never so strictly and habitually 

 opposite even on the lower stem as in the case of D. alata or D. 

 cayenensis, both of these, but especially the latter, being pro- 

 vided with branches, on some of which all the leaves are opposite 

 while on others all are alternate. There is another, and perhaps 

 more serious difficulty; the male flowers of J), chondrocarpa have 

 a glabrous perianth, whereas the female flowers of I), altissima 

 are described by Plumier as having a pubescent perianth, and 

 until T). (tltissima, Lamk, which has never been met with since 

 Plumier made his drawing and description, can be studied from 



f actual specimens from Martinique, it seems for the moment 



desirable to leave it an open question as to the identity of D. 

 chondrocarpa and D. altissima. This uncertainty, however, does 

 not affect the question immediately before us, which is the 



identity of D. sativa, Linn. [1]. 



Had the Linnean citation of the Hortus Cliffortianus in 1753 

 been identical with the Linnean citation of 1747 (Flor. Zeyl. p. 170 

 and with that of Koyen in 1740 (Flor. Leyd. Prodr. p. 527 ) ; h 

 it been confined to the page (Hort. Cliff, p. 459) and therefore to 

 the plant there somewhat inadequately described; we might have 

 been justified in deciding that T). sativa, Linn. [1], and D. villosa, 

 Linn. (Sp. PL ed. 1. p. 1033), are conspecific. The question for 

 settlement then would have been which of the two names, simul- 

 taneously published by him for the same species, it is desirable 

 to retain; which of the two it is necessary to discard. But in 

 1753 Linnaeus, with less caution than he had hitherto displayed, 

 cited not only the page with his own text but the plate with 

 Ehret's figure, under D. sativa, Linn. (Sp. PL ed. I. p. 1083); the 

 figure in question becomes, as a consequence, at least some part of 

 the authority for his species ; as the figure is a composite one. the 



species *-' is, even at its inception, not a natural species, and 

 the name ]). sativa is without a definite basis. The judgment of 



