genus are most difficult of discrimination, owing to the 
necessity of having for this purpose flowers of both sexes 
and also ripe fruit, and because in foliage very different 
species resemble one another. The male inflorescence 
appears to me to afford the best sectional characters, but 
it may have to yield to carpological ones, when the fruits 
are better known. The male flowers of individual species 
are very inconstant as to the number and form of sepals, 
and number of stamens. The disk-glands when present 
are too minute and, I think, variable as to presence or 
absence, to afford aid in Indian species; nor do I find the 
anthers truly extrorse in any, the slits being more or less 
lateral when not truly introrse. 
As mentioned betore, while at least 20 species of the 
genus have edible fruits, only four are known to be in 
cultivation to any considerable extent. Two other spe- 
cies have been mentioned (Duct. sci. nat., 1816) as being 
cultivated; the rest are still wild, even though the fruits 
of some of them are collected and sold in the markets 
and, in some cases, even surpass in quality those of the 
cultivated species. Besides being sought for their fruits, 
the trees yield a good timber, and it has often been re- 
corded that it is used for building houses and boats, for 
posts, and other light constructions. Other parts of the 
plant, such as the leaves, the bark, and the roots, have 
likewise been employed in native medicines, for dyeing 
and even as vegetables. A brief discussion of these uses 
will be given at the end of the paper. All the species 
which thus far have been recorded as of economic value 
are presented in Table 1. The following discussion will 
be restricted only to the four cultivated species men- 
tioned above. 
Betore proceeding with the species discussion, how- 
ever, it may be helpful to consider the vernacular names 
of the species. In this case, we have a chaotic, though 
[ 70 ] 
