340 



object of arriving at a definite conclusion. The nature of the 

 conclusion to which they have been led is such as to render it 

 desirable to explain how the difficulty experienced by their 

 predecessors has arisen, and to give a brief historical review of 

 the various attempts that have been made to solve it, as pre- 

 liminaries to the statement of the results at which they have 

 themselves arrived. 



Though Linnaeus first applied specific Ames to components 

 of the genus Dioscorea, in 1753 he had already formed concep- 

 tions of the limitation of five out of the eight species then named 

 as early as 1737 (Hart. Cliff, p. 459), and of two of the remaining 

 three in 1747 (Flor. Zeyl. pp. 170-2). The terms in which these 

 conceptions were expressed in 1753 are as follows; the dates of 

 their respective formation are added in each case: — 



pentaphylla. 1. Dioscorea foliis digitalis. Four references 



are given to ussist in the interpretation of this conception; 

 there is no specimen to illustrate it in the proper herbarium 

 of Linnaeus : Hort. Cliff. (1737) n, 5. 



triphylla. 2. Dioscorea foliis ternatis. Two references given; 



no specimen: Hort. Cliff. (17*57) n. 4. 



aculeata. 3. Dioscorea foliis cordatis, caule aculeato bulbifero. 



Four references; no specimen: Hort. Cliff. (1737) n. 3. 



alata. 4. Dioscorea* foliis cordatis, caule alato bulbifero. 



Five references; one specimen: Flor. Zeyl. (1747) n. 360. 



bulbifera. 5. Dioscorea foliis cordatis, caule laevi bulbifera. 



Five references; one specimen: Hort. Cliff. (1737) n. 2. 



xatira. 6. Dioscorea foliis cordatis alter nis, caule laevi. 



Seven references;, one specimen written up by Linnaeus, 

 with another attached but not written up : Hort. Cliff. 

 (1737) n. 1. 



villosa. 7. Dioscorea foliis cordatis alternis oppositisque, 



caule laevi. Two references; no specimen: Sp. PL (1753). 



oppositifoha. 8. Dioscorea foliis oppositis ovatis acuminatis. 



Two references; no specimen: Flor. Zeyl. (1747) n. 361. 



It may assist us in appreciating the situation if the diagnostic 

 characters stated and implied by Linnaeus in this arrangement 

 be presented in the form of a key, implied characters being placed 

 within brackets. 



Key to the Ltnneax species of Dioscorea. 



Leaves digitate : 



[Leaflet^ more than three] 1. D. pentaphylla 



Leaflets three only 2. J), triphylla. 



[Leaves simple] : — - 

 Leaves cordate : — 



Leave- with axillary bulbils : 



Stem armed with prickles 3. D. aculeata. 



[Stem unarmed] : — 



Stem winged 4. J), alata. 



Stem not winged 5. D. bulbifera 



