104 PROCEEDINGS OF THE NATIONAL MUSEUM vol. los 



that the female genitalia are often as useful as the male genitalia in 

 ascertaining the relationship of butterfly groups. 



Concerning the genera now under scrutiny, the soundest early 

 works were those of Butler (Cist. Ent., vol. 1, pp. 33-58, 1870) and 

 Grote (Proc. Amer. Philos. Soc, vol. 39, pp. 4-67, 1900) and were 

 based, according to the practice of that time, mainly upon venational 

 characters. To these studies Klots (loc. cit., pp. 217-220) added 

 a study of the male genitalia, arriving at a fairly satisfactory generic 

 treatment which was superior to all similar generic studies in the 

 Rhopalocera and was certainly an outstanding landmark. 



In preparing the present paper, the male genitalia, venation, and 

 habitus were thoroughly studied and evaluated. Also, the legs, 

 including the claws (which proved to be of great value), and the female 

 genitalia were studied. 



The complexities of relationship among the genera studied were 

 found to be much greater than heretofore believed. The use of any 

 one character (or kind of character) alone was found to produce a 

 different alignment than when some other character was employed. 

 Thus, a consideration of the uncus alone shows Theochila, new genus; 

 Tatochila; Phulia (Infraphulia, new subgenus) ; Piercolias; and Baltia 

 to be quite distinct from one another, while Hypsochila and Phulia 

 (Phulia) would be considered very close to each other and quite 

 like Tatochila. Considering the aedeagus only, one group of the 

 genus Tatochila and the genera Hypsochila, Phulia, Piercolias, and 

 Baltia would certainly seem to be nearly identical, while Theochila 

 and four of the species groups of Tatochila would appear to be quite 

 different from each other and from those mentioned above. On the 

 other hand, Hypsochila, Phulia, Piercolias, and Baltia differ greatly 

 from Theochila and Tatochila in lacldng the tibial spurs of the mid and 

 hind legs. Again, if we placed too much emphasis upon the claws, 

 paronychia, and pulvilli, we would place Hypsochila and Theochila 

 an unreasonable distance from the other genera and from each other. 

 A consideration of the number and fusion of the radial veins taken 

 by itself would give us the unnatural alignm.ent of having Theochila, 

 Tatochila, Hypsochila, and the nominotypical subgenus of Piercolias 

 at one end (with four radial veins), one species of Phulia (Infraphulia) 

 at the other end (with two radial veins), and Phulia (Phulia), a second 

 species of Phulia (Infraphulia) , and Piercolias (Pierphulia, new 

 subgenus), in the middle (with three radial veins). 



It became quite obvious, if we were going to attain a neaily natural 

 classification, that all of these characters had to be considered and 

 used. The result was that sometimes special importance was given 

 to a single structure in one genus (or in several genera) and that 

 much less importance had to be given that same structure in another 

 genus (or in several other genera). 



