418 S. H. SCUDDER ON THE STRUCTURE AND 



Vestales most closely, but is so different from the lower tribes of the sub-family, that were 

 we to base our opinion on the caterpillar alone we could hardl^^ help considering the Eumae- 

 ida3 as a group possessing a value equivalent to the rest of the Ephori. This view, however, 

 we shall be unable to maintain after an examination of the other stages of the insect. 



The chrysalis (figs. 10-12) is stout, full, arched and rounded, the mesothorax elevated, 

 domed and mesially subcarinate,^ separated by a pretty deep excision from the very full and 

 arched abdomen, which is much higher than the thorax and rather abruptly elevated at the 

 second segment. The head is small, not at all protuberant, situated entirely upon the under 

 surface of the body, so as not to be seen on a dorsal view, being concealed by the fullness of 

 the prothorax ; this is very broad, somewhat tumid, especially laterally, closely resembling 

 the same part of the larva at its final stage ; the under wings are entirely concealed by the 

 upper. The body, viewed from above, is obovate, very obtuse at either end, but especially 

 in front, broadest from the second to the fourth abdominal segments, the under surface not 

 flattened but broadly rounded transversely, the terminal segment of the abdomen apical 

 though directed downward, and the cremaster reduced to some slight rugosities, wholly des- 

 titute of armature. The body is girt by silken threads Avhich pass between the first and sec- 

 ond abdominal segments dorsally, and then are directed forward (fig. 10) ; it cannot be 

 attached at the posterior extremity of the body, as stated by Westwood, for absolute want 

 of any hooks to phmge into the silk. 



In the Vestales, as far as one can gather from the few meagre descriptions and imperfect 

 illustrations that are published of their transformations, the head of the chrysalis is visible 

 from above, as in all chrysalids of butterflies, excepting those of Ephori, and the chrysalids 

 themselves are often angulated. In the Ephori, on the other hand, the head is always upon 

 the under surface of the body, concealed from view on a dorsal aspect, and so, also, to a 

 greater or less extent, is the last segment. The chrysalids are always rounded and entire. 

 There can be no doubt, therefore, that the affinities of Eunicneus at this stage are decidedly 

 with the Ephori, although the terminal segment of the body seems to be more exposed dor- 

 sally than is usually the case with them. It evidently resembles that of the Armati more 

 closely than it does the other tribes of Ephori, but has a more flattened and laterally ex- 

 panded prothorax. 



The different tribes of Ephori differ from each other in the chrysalis state in a curious 

 manner and one which has never yet been noticed. The dermal appendages of Armati, 

 for instance, consist of equal hairs, tapering only at the tip, profusely provided with minute 

 spicules diverging from the stem at but a slight angle. The dermal appendages of Adoles- 

 centes consist of uniformly tapering hairs, abundantly provided with minute spicules, which 

 diverge from the hairs at right angles. Those of the Villicantes are fungiform. The 

 Eumi^idte, again, are provided with compressed, ribbon-like hairs, equal, tapering only 

 at the extreme tip and slightly twisted throughout their length. In this respect again 

 Eunn\3us is most nearly allied lo the Armati. 



A few years ago in my Systematic Revision of North American Butterflies I indicated my 

 adhesion to the views of a few writers who considered the two groups, known in modern 

 times as Erycinidaj and lycaiuidiJ?, but to which the earlier terms Vestales and Ephori 

 should be applied, as but a single family. New and unlooked for evidence in this direc- 



' It cannot be correctly described, as Westwood has done, as "raised into a slight conical protuberance." 



