Mr. H. W. Bates on the Endomychidae of the Amazon Valley. 171 



simple ones of variety and species. There is, first, the case of 

 individual differences which arise amongst the members of one and 

 the same species in one and the same locality : these can scarcely 

 be called varieties, as they may exist amongst the offspring of the 

 same parents. Secondly, there is the case of species which are tole- 

 rably constant to their type in one region whilst extremely variable 

 in another. There are others which, in a distant locality, produce 

 a variety which embraces all the individuals of the species existing 

 there : in some cases there is an intermediate space, between the 

 ranges of these varieties and their typical forms, which is unpeopled 

 by either the species or its variety : in numerous instances, however, 

 there exist no apparent natural barriers. The origin and main- 

 tenance of these defined local varieties is a highly interesting ques- 

 tion : some of them have a great resemblance to their typical forms, 

 whilst others differ greatly, although often more in appearance than 

 in reality. It can be shown in some instances, where the local form 

 is considered on all hands to be a perfectly distinct species, that all 

 the points of difference from its parent form can be paralleled sepa- 

 rately by instances in undoubted varieties of species of the same 

 group. The first step in specific dissimilarity is afforded us by 

 Nature in those instances where two closely allied forms, each 

 inhabiting its separate area, live together in an intermediate district 

 without amalgamating. The dissimilarity is proved to be incom- 

 plete when two forms, apparently specifically distinct, intermingle 

 and produce connecting links when they meet together on the 

 frontiers of their respective ranges. It is, however, I think, very 

 desirable that the two or more forms in such cases should be treated 

 separately in our books and placed separately in our collections, and 

 therefore C. interruptus and C. cinctus may require their distinctive 

 appellations. The tendency in systematic zoology to treat geographical 

 forms as distinct species, has the advantage of exciting attention 

 to the exact determination of the localities of specimens. A vari- 

 ation, which in one case might be an individual difference of little 

 importance, becomes in another a scientific fact of the highest 

 significance. 



§ C. Elytra very slightly convex ; joints of the antenna short and thick ; 

 abdomen of the male with conspicuous sexual marks on the ventral 

 surface. 



16. Corynomalus dentatus, Fab. ; Gerstaeeker, Mon. p. 166. 

 I obtained, at Ega, one example only of this species, a male. It is 

 darker in colour than any of the varieties mentioned by Gerstaeeker 



