24 DONALDSON BODINE. 



same position usually occupied by the single one. It is perhaps 

 worth noting here that, while the females are usually behind the 

 males in the development of pectinations, they are not at all deficient 

 in the number or the perfection of the cones. 



There is yet another feature which is of importance in determining 

 the relationship of an antenna. This is the character of the surface 

 of the chitinous exoskeleton of the clavola. In the Jugatps the sur- 

 face as a whole is even, but has numerous slender prolongations 

 which have received the name of fixed hairs, Figs. 51 and 53. 

 These are not articulated with the chitin as are the true hairs. They 

 are simply points which are continuous with and parts of the general 

 surface and do not originate at the ends of pore canals. In the 

 Fi'enatj« these are not found, but the surface of the whole clavola is 

 divided up into more or less hexagonal areas bounded by thin con- 

 tinuous plates set normal to the surface, or into a series of more or 

 less imbricated plates with various surfaces and outlines. These two 

 distinct kinds of surface covering are of great importance, as they 

 represent specializations differing in kind, and therefore indicate a 

 dichotomous division in the line of descent. There are other char- 

 acters which would indicate the same thing, but no other autennal 

 character shows that the Hesperiina and Papilionina branched off* 

 from the other Frenatae after their separation from the Jugataj. The 

 degrees of specialization attained in this character of surface are also 

 worthy of attention as they sometimes afford clues to the real posi- 

 tions of families. 



The surface marking of the chitin of the clavola in the Frenatse 

 presents a great variety of forms, but with the possible exception of 

 a few apparently aberrant cases it is possible to trace a pretty definite 

 line of development })assing through them all. This is important, 

 because it means that the system originated but once. The sim})lest 

 and i)rol)ably the most primitive condition exists in the anteinia; of 

 the Pyralidina. Here we find the general surface smooth, but di- 

 vided up into more or less hexagonal ai'eas, fenced off' as it were by 

 continuous plates of chitin standing nornuxl to the surface. The 

 areas thus demarcated are said by some authorities to correspond 

 with the outlines of the underlying dermal cells, and it may be that 

 the plates represent the edges of separate areas of chitin developed 

 from the cells as centres. In the adult condition of most forms, 

 however, I do not believe that the arrangement has any relation to 

 the underlying structures. Fig. 34 represents a portion of the an- 



