interesting comment, since it is impossible to obtain an 
elastic rubber from true Hevea Spruceana; yet the 
quality of the product of H. Benthamiana is usually only 
slightly inferior to (and frequently as high as) that of 
H. brasiliensis, the source of the best rubber. In ascrib- 
ing to the flowers of Hevea discolor a ‘‘reddish purple’’ 
hue, Bentham is very definitely describing a character of 
Hi. Spruceana; H. Benthamiana is known to have bright 
lemon-yellow flowers. Judging from the amount and 
kind of variation which Bentham describes for the anthers 
of Hevea discolor, we might be moved to think more of 
Hi. Benthamiana than of H. Spruceana. Although some 
variation is found in the number and placement of the 
anthers in the latter species, there 1s very much less in- 
stability than is evident in the former. 
Bentham considered Siphonia Spruceana and S. pauci- 
flora directly following the description and discussion of 
S. discolor. Describing Siphonia Spruceana as a plant of 
the banks of the Amazon below Santarem, Bentham 
stated (I. c. 370) that it has ‘‘numerous flowers, purple 
withinside, and much larger than in JS. disco/or,’’ but he 
made no definite statement that he considered it to be 
allied at all closely to S. discolor. On the contrary, and 
rather surprisingly, he wrote (I. ¢. 8370) of Stphonia pauci- 
flora, a species which Spruce collected in rocky situations 
along the Rio Uaupés: ‘‘This is certainly near to S. 
discolor, and may prove a mere variety.’”’ 
Although it is clear that Bentham based his descrip- 
tion of Hevea discolor on the two collections, it is plain 
that the greater weight in both the English and the Latin 
description is given to Spruce 1171 which represents HZ. 
Spruceana. I, therefore, typify Hevea discolor by choos- 
ing Spruce 1171 as the lectotype of this concept. 
We may then treat the binomial Hevea discolor, as 
Ducke definitely has done (in Arch. Jard. Bot. Rio 
[ 253 ] 
