Combinations of Tropical American B^dterflies. 593 



An interesting point emerges in connection with the 

 ancestral white of Dismorphias of tliis pattern. Besides 

 the white portion on the upper side of the hind-wing 

 mentioned above, a white patch occupies the inner mar- 

 ginal area of the fore-wing under side. The meaning of 

 this retention of the white on parts of the surface hidden 

 by the overlapping of the wings is very clearly explained 

 and iUustrated by Professor Poulton in his paper " Natural 

 Selection the Cause of Mimetic Resemblance and Common 

 Warning Colours," Linn. Soc. Lend., Journ. Zool. vol. 

 xxvi, p. 606, pi. 40. In addition to the white patches, 

 there are other distinct traces of the ancestral white 

 on the under side of both wings. These persistent traces 

 are formed by a few white scales — easily visible to the 

 naked eye — near the hind margin of the hind-wing 

 and esiDCcially at the apex of both wings. The general 

 appearance of the under side suggests that it may pos- 

 sess a cryptic significance, which is an argument against 

 the unpalatability of this Pierine.* 



Sub-division (b). All the Ithomiines resemble each 

 other except for the fore- wing apical markings (Plate 

 XXXI, figs.1-6). Heliconius narcma (Figs. 7 and 8 ) per- 

 haps follows Mechanitis lysimnia (Figs. 1 and 2) more 

 closely than it does any of the other Ithomiines of this 

 sub-division : its likeness to Melin/Ba ethra of sub-division 

 (a) is far stronger. The Mechanitis does not exhibit that 

 tendency towards transparency which is so characteristic of 

 large numbers of the Ithomiinm, and is readily noticeable in 

 the Cerntinia (Figs. 3 and 4) and Napeogenes (Figs. 5 and 6) 

 belonging to this combination. Heliconius polychrous 

 (Fig. 9) must be considered a rough mimic. Its prin- 

 cipal defect, which is not very apparent in the figure on 

 Plate XXXI, is the great reduction of the tawny colour 

 and the corresponding increase in the black and yellow 

 markings of both wings. Protogonius clrurii (Figs. 10 and 

 11) again is a poor mimic, as is customary in that genus. 

 It is, as is also usual in the genus Frotogonius, the only 

 member of its association with an obviously and strongly 

 cryptic under surface. The yellow band across the hind- 

 wing is easily traceable, although neither sharply defined 

 nor bordered by black, as in the models. The white apical 

 spot is distinct, and the white spots in the hind margin of 



* See R. Shelford in Poulton's " Essays on Evolution," 1908, pp. 

 351, 353. 



