520 W. L. TOWER, 



In the Coleoptera, more than in any other group of the Holo- 

 metabola, are found diversity of larval form and metamorphoses. From 

 the generalized, Campodea form larvae of the Camhidae, Dytisddae, 

 Staphylinidae and CoccinelUdqe , to the cylindrical, flesh feeding 

 forms, SilpTia and Dermes^es, toli the elongated Ektteridae, the 

 Scarabaeid grubs, to the phytophagous eruciform Chrysomelid larvae, 

 to the apodus Ceramhycidae and Buprestidae, and finally to the most 

 specialized of all, the Meloidae and Curculionidae, is a series such 

 as no other order of insects can show. Where such diverse conditions 

 of metamorphosis exist one should not expect to find anything like 

 the uniformity in wing development which pervades the Lepidoptera. 

 The problem is further complicated by the fact that only in a few 

 cases do we know the number of stages in larval life, and these vary 

 from three in Phytonomus punctatus Fab. to five in Meloe and seven 

 or more in Dermestes. Many Coleoptera require one, two, three, or 

 even more years in which to complete their larval life and in these 

 species the number of moults is not known. 



In the course of this study, with the exception of a few forms, 

 I have been obliged to depend upon material taken at random; con- 

 sequently, in many cases it has not been possible to obtain complete 

 series of stages. This has been especially true in the case of long- 

 lived species {Ceramhycidae and Scarubaeidae). In determining the 

 larval stages I have been obliged to use the head measurements as 

 the only reliable criteria in determining the larval stages, and in 

 general this method has given satisfactory results. 



1. Origin of tlie Wing Fundament. 



At what time during the development of the individual the wing 

 fundament arises, is a question which has been answered satisfactorily 

 for only a few species of insects. Several authors, it is true, have 

 found the wing discs in the late embryonic stages, but have not 

 traced them back to their origin. Thus Verson (1890) showed that 

 the wing discs were present in Bombyx mori in late embryonic life, 

 and further that these discs had the cells composing them arranged 

 in a "rosette", and that they occupied the place where the spiracles 

 of the mesothoracic and metathoracic segments should have been. 

 He therefore concluded that the wing discs were derived from the 

 spiracles. In very young Lepidopterous larvae, several authors, 

 notably Dewitz (1881), Pancritius (1884), Gonin (1894) and Merger 

 (1900), have found the wing discs, and have uniformly believed 



