8 IOWA STUDIES IN NATURAL HISTORY 



they are more primitive than the Pamphilinae. The close rela- 

 tionship of the Pyrrhopyginae with the Hesperiinae and of the 

 Megathymidae with the Pamphilinae leads to their being placed 

 at the beginning and end of the superfamily respectively. The 

 result is a linear series which is not entirely satisfactory, but 

 since no linear series can represent true phylogenetic relations 

 this must be accepted as the best possible, and it does, at least, 

 correctly indicate the general relations of the several major 

 divisions. 



The separation of genera has been the most troublesome phase 

 of the study of skippers since Hubner's classification was first 

 amplified. I have come to the conclusion that the intermediate 

 position of the group, together with the apparently transitional 

 state of many of the structures, is accompanied by a greater 

 blending of forms than has been recognized in the past, and that 

 the normal genus may present a wide variation of structure, 

 provided that a transition between the extremes be present in the 

 included species. This is nicely illustrated by Thanaos, Hesperia 

 and Poanes (sensu B. & McD., Check List). In Thanaos we 

 have a group of insects of very similar habitus which no one has 

 ever divided, but within the genus are to be found differences 

 in structure which have been made to separate three genera in 

 other cases. The neuration, shape of the wings, palpi and sec- 

 ondary sexual characters very nearly run the gamut of variation 

 found in Group B of the Hesperiinae. Hesperia is similar but 

 shows an even greater range of variation in the antennal club, 

 shape of the wings, and in the palpi. In fact this variation is so 

 great as to occasion some doubt of its unity, but it is impossible 

 to divide the genus without separating some species whose rela- 

 tionship is apparent. 



The matter of secondary sexual characters as a basis for the 

 separation of genera is the greatest bugbear of systematists in 

 this family. Godman and Salvin and Mabille have contributed 

 abundantly to the confusion of genera so based, and in many 

 cases these genera cannot be separated by other means. As far 

 as I am aware the only definite stand taken upon the question is 

 that expressed by Watson in his revision.* He says: "With 



* Since writing this I have found a quotation from Dr. P. L. Sclater by Col. C. 

 Swinhoe in defense of genera based on secondary sexual characters (Ann. & Mag. 

 Nat. Hist, (vn), in, 108, 1899). 



