110 Dr. H. Eltringham on Specific and 



different from that of any form of melpomene, and, in fact, 

 has a typically silvaniform appearance. It is nearest to 

 that of a species with which one would not at first think 

 of associating it, viz. vetustus : indeed, the only obvious 

 difference between the armature is that in vetustus the 

 extremity of the clasper is rather more densely pilose. 

 That the species is, in fact, closely related to vetustus I 

 have no doubt, and the further evidence for this will be 

 found under the discussion of patterns. It may here be 

 mentioned that the most obvious difference between 

 tumatumari and melpomene forms is the occurrence in the 

 former of a sulphur yellow streak on the underside of the 

 hind-wing, this streak being not on the costa, but below 

 the costal nervure. 



On my pointing out to Mr. Kaye the significance of this 

 yellow line, he kindly brought for examination two mel- 

 pomene-like forms, one of which agrees with melpomene 

 elevatus, Nold., and the other an undescribed female form 

 somewhat like it, from his own collection. Both these had 

 the peculiar yellow line, and without anatomical examina- 

 tion might well have been regarded as geographical forms 

 of tumatumari. Microscopic investigation of the male 

 example showed, however, that though it was apparently 

 not a form of melpomene, it was equally specifically separate 

 from tumatumari. The armature is again of a somewhat 

 silvaniform type, but resembling that of H. sergestus. 

 These two species, elevatus and tumatumari, will be further 

 considered in the discussion of patterns. 



Although they are separable from melpomene it must not 

 be supposed that they are as markedly distinct from that 

 species as are many of the species of Section II from one 

 another. Preparations of the genitalia of melpomene show 

 much individual variation, and whilst there would never 

 be any difficulty in distinguishing the armatures of, say, 

 anderida and melpomene, there might be more difficulty in 

 separating some preparations of melpomene from certain 

 of the Silvaniformes. In other words, the Silvaniformes 

 are not, in spite of their Melinaea-hke patterns, a markedly 

 separate group, and, in fact, it may be said that all the 

 species belonging to Section I are far less satisfactorily 

 differentiated than those of Section II. This fact would 

 seem to lend support to the view that Section I is of more 

 recent development, as we should expect, sinceits members 

 are mimics rather than models. 



