A NCESTR Y A ND CLA SSIFICA TION. 243 



their exquisite beauty, that their scientific study 

 has been largely abandoned by those who are best 

 fitted for this work by special scientific training. 

 The description of a large proportion of the 

 known species, the characterizing of very many 

 of the genera, involving in either case the desig- 

 nation of each such group in a system, has been 

 done by those who without previous training in 

 any branch of natural history commenced their 

 career by the formation of a cabinet, continued it 

 by the description or illustration of the new 

 species they were able to acquire, then essayed 

 the foundation of new genera, and all without 

 any proper knowledge of other animals, even of 

 those most nearly allied to butterflies. Of course 

 such persons are quite incompetent to judge of 

 the relative subordination of characters in the 

 single group they examine. The result has been 

 exactly what might have been predicted a con- 

 fusion and a contradiction in current classifica- 

 tions that is at war with the harmony of nature. 

 Much is due no doubt to our comparative igno- 

 rance of the early stages of butterflies. Strange 

 as it may seem, amateurs seldom collect, still 

 more rarely preserve for comparative study, eggs, 

 caterpillars, and chrysalids. 



The family groups into which butterflies should 

 be divided have been variously given all the way 

 from two to sixteen. As the structure of the 



