50 THE AMERICAN MIDLAND NATURALIST 



Dactylon, etc., Slcane, hist, i, p. 113, t. 70, f. 3. We can not, there- 

 fore, neglect a study of the West Indian species in endeavoring to 

 ascertain the proper application of the Linnaean name. Mr. 

 Hitchcock in Contributions U. S. Nat. Herb. XII, pp, 134, 142 and 

 209, 1908-9, has shown that the Sloane reference belongs to what 

 is now known as Syntherisma setosa (Desv.) Nash or 5. digitata 

 (Swz.) A. S. Hitch.; he combines under the latter name two forms 

 that have been variously considered as distinct species or as 

 varieties of Panicum sanguinale Linn. Grisebach, Flora Br. W. Ind. 

 544, 1864, has described these West Indian forms as pilose; there- 

 fore, they can scarcely be considered as belonging to P. lineare 

 Linn, which was described as laeves. It has been customary to 

 accept Burmann's Panicum lineareVl. Ind. 25, pi. ro, fig. 2 (fig. 3 

 of the plate) as typifying the Linnaean species but this can not be 

 adhered to as it did not make its appearance until 6 years after 

 Linnaeus had published his P. lineare. Burmann uses the 

 Linnaean diagnostic description verbatim and the references to 

 Sloane mentioned above, with the added remark: — "Specimina ex 

 India missa variant spicis saepe binis linaeribus and flosculis 

 alternis." Since Linnaeus assisted Burmann in the preparation of 

 his Flora Indica, the above phrase seems to indicate that the former 

 thought that the variable specimens from India were not typical of 

 his P. lineare but could be referred to it; also it may be considered as 

 more evidence that the original habitat "in Indiis" as given by 

 Linnaeus did not include India but referred more especially to 

 the East and West Indian Archipelagoes. It seems probable that 

 fig. 2 (which is fig. 3 of the plate due to a transposition of the 

 numbers) represents the variable plants of India mentioned in the 

 supplementary paragraph. Hooker, Flora Br. India VII, 289, 

 1897 refers Burmann's P. lineare to Cynodon Dactylon (Linn.) Pers. 

 I have seen specimens of the latter from India that might be said 

 to be fairly well represented by Burmann's figure if it may be 

 viewed in the light of a very crude drawing. The Index Kewensis, 

 Vol. II, 415, 1895, refers it to Paspalum brevifolium Fluegge; 

 but Hooker, 1, c, 18, describes this species as with geminate flowers,* 

 one sessile and the other pedicelled which is at wide variance with 

 Burmann's figure, and the Linnaean description. The Index 

 Kewensis recognizes Panicum glabrum Gaud, as a valid species; 

 also P. lineare Linn, as a valid species of N. America but I am unable 

 to interpret it. 



