The origin and development of the wings of Coleoptera. 561 



III. Conclusion. 



Id the preceding pages I have described [the development of the 

 wings of beetles and of their parts and it now remains to consider 

 briefly the general aspects of this account and its bearing upon the 

 general topic of the origin and development of insect wings. 



In an early part of this paper I have shown that the wings and 

 spiracles arise in homologous positions upon the sides of the segments 

 as determined by the attachment of homodynamous muscles, and 

 that the hind wings are without much doubt derived from the de- 

 generate spiracle disc of the metathorax, but in the elytra the case 

 is different. Whether in the migration of the mesothoracic stigma 

 to its larval position only the opening migrates and the disc remains 

 behind, or whether both the spiracular disc and opening are moved 

 forward is a point upon which I was unable to get decisive evidence. 

 My evidence points strongly to the view that it is the spiracular 

 opening alone which migrates and that the spiracular disc remains 

 behind to form the fundament of the elytron. At present I believe 

 that the evidence points strongly to Verson's (1890) view that the 

 wings of Coleoptera and Lepidoptera are derived from the rudiments 

 of the mesothoracic and metathoracic spiracles. The question is, 

 however, a decidedly open one, and one upon which studies of the 

 development of the late embryo and early larva of Heterometabola 

 will undoubtedly throw much light. 



All of the evidence here presented concerning the condition of 

 the wings of Coleoptera and Heterometabola is most positively opposed 

 to the theory of the origin of the wings of insects as dorsal, backward 

 prolongations of the tergum. When the anatomical position and de- 

 velopment of the wings are considered, as far as is known no support 

 for this theory can be derived from these two important sources. 

 Further, the external position and appearance of the wings in larval 

 stages cannot be criteria of value, for it is well known that the wings of 

 Heterometabolous nymphs often shift their position during the larval 

 stages to one that is adapted to the habits and environment of the 

 larva. Thus the wings in Periplaneta arise on the sides of the thorax 

 in exactly the same place where they arise in Coleoptera or Lepido- 

 ptera, but the wings soon migrate dorsally and posteriorly, and when 

 they become external appear as direct backward prolongations of the 

 tergum. This theory of the origin of insect wings has only the wildest 

 sort of speculation for its support. At present we do not know that 



Zool. Jahrb. XVII. Abth. f. Morph. ßg 



