338 



SCIENCE. 



[N.S. Vol. XIX. No. 478. 



possibility of an important action in the in- 

 ternal physiology of plants was suggested. 



The Present Confusion Among the Species of 



Dioscorea: Me. W. E. Safford. 



Mr. Safford became interested in the classi- 

 fication of the species of Dioscorea during his 

 cruises among the islands of the Pacific. On 

 many of them yams are among the principal 

 food staples of the natives, and occur both 

 spontaneously and in cultivation. Many dis- 

 tinct forms occur which have received ver- 

 nacular names on the various islands, but the 

 delimitation of species and varieties is very 

 difficult. The same species varies under dif- 

 ferent conditions of light and moisture ; leaves 

 vary in shape, pubescence, and relative posi- 

 tion on young and old specimens, and, indeed, 

 on different parts of the same plant. Many 

 of the early collectors contented themselves 

 with giving lists of native names together with 

 a brief description of the tubers to which they 

 apply. Many of these names prove to be 

 descriptive, as ' white yam,' ' blue yam,' ' one- 

 head yam,' ' devil yam ' and the like. No at- 

 tempt has been made to bring together the 

 various forms of different island groups for 

 comparison, and no confidence can be placed 

 in existing synonymy. 



On the island of Guam the natives have 

 divided the yams into two classes according to 

 the shape of their leaves, calling all those with 

 broadly cordate or orbicular leaves with a deep 

 basal sinus ' Nika,' and those of which the 

 leaves are more or less sagittate or hastate 

 ' Dago.' Gaudichaud, botanist of the Frey- 

 cinet expedition which visited Guam in 1818, 

 referred the varieties called Dago to Dioscorea 

 alata, and those called Niha to Dioscorea 

 aculeata. In Guam the wild Nika (' Nika 

 cimarron,' or ' Gado ') differs radically from 

 the cultivated form in having a mass of lateral 

 roots about the base modified into sharp, wiry, 

 branching spines. Whatever may have been 

 the cause of their origin, they serve to protect 

 the sweet farinaceous tuber below. Gaudi- 

 chaud referred this species to D. aculeata, but 

 it proves to be D. spinosa Eoxb. 



Linnseus' descriptions are brief and quite 

 insufficient. Many of them were evidently 

 made from type plants in poor condition, and 



in some of them a single description included 

 two or more species. According to Sir Joseph 

 Hooker a part of Linnaeus' description of 

 Dioscorea sativa (' Spec. PI.,' ed. I., 1033) ap- 

 plies to D. spinosa Roxb., to which should also 

 be referred Roxburgh's ovm D. aculeata. The 

 true D. aculeata L. is without the basal spines 

 above described, and D. sativa L. is a glabrous 

 plant with a terete bulbiferous stem. To the 

 latter species should be referred the D. hulbosa 

 of Robert Brown. 



In looking over herbarium specimens it be- 

 comes apparent that yams can not be studied 

 from dried plants. Points of distinction often 

 lie in the flowers or fruit, which are often 

 wanting in herbaria or are represented by 

 only one sex. Cultivated yams are propagated 

 asexually; and many varieties, like those of 

 sweet potatoes, ginger, Colocasia, and other 

 cultivated plants, are seldom seen in flower 

 or fruit. Other species have been differ- 

 entiated according to the form of their tubers ; 

 and these are almost always lacking in her- 

 baria. Still others have been described with 

 reasonable accuracy, but figures of different 

 species have been cited as illustrations. 



Sir Joseph Hooker found the species of 

 Indian Dioscorese in such indescribable con- 

 fusion that, after devoting much labor in de- 

 termining and delimiting them, he had to let 

 them appear in his ' Flora of British India ' 

 in a shape most unsatisfactory to himself, say- 

 ing that he could not hope to avoid errors; 

 that the Roxburg'hian food-yielding species 

 are for the most part indeterminable, and that 

 the Malayan species .are even more loosely de- 

 scribed than the Indian; while in the Wallich- 

 ian collection, which is very complete, the 

 species are often mixed. 



It is evident then that food-yielding 

 varieties of Dioscorea must be studied on the 

 spot where they are cultivated, and not in 

 market places or in museums. Series of 

 complete specimens of the plants should be 

 secured, showing different parts of the stem, 

 basal and cauline leaves, flowers of both sexes, 

 fruit and photographs of growing plants and 

 tubers attached to the stem, together with 

 specimens in alcohol or formalin of the in- 

 florescence and the tubers themselves. 



