The origin and development of the wings of Coleoptera. 537 



or Lepidoptera. During the first instar the disc thickens slowly, but 

 after passing into the second instar, it invagiuates slightly, then 

 evaginates into the narrow cavity thus formed, and at the following 

 ecdysis becomes an external structure. It should be noted that the 

 wings are in no sense "dorsal backward prolongations of the tergum" 

 as Packard, Graber and others have argued, but arise as true lateral 

 structures which, for the mechanical convenience of the larva, are 

 shifted in later moults to another position. 



In Microcentrum latifolium (Orthoptera), the wings arise in the 

 same way as described for Änasa, but in the embryo, so that at hat- 

 ching the embryo shows well developed wings already evaginated lying 

 beneath the cuticula of the first larval stage ready to become external 

 at the beginning of the second larval stage. 



If the development of the wings in Anasa or Microcentriim could 

 be retarded so that the wings should not begin to grow until the 

 last larval stage, which should also be prolonged, then the development 

 of the wings of the Holometabola and Heterometabolous species would 

 not difier. Even as it is the development of the wings in Anasa or 

 a beetle like Fhymatodes is alike even in details except in the time 

 when they become external, as far as the type of wing development 

 is concerned there is no ditïerence. In this connection it is of interest 

 to note that in some of the Tettigidae (Tettigidea), — so I am informed 

 by Dr. Hancock, — the wings remain beneath the cuticula during the 

 larval life and then at the final transformation emerge as fully de- 

 veloped wings. In this interesting case we see in a Heterometabolic 

 insect a condition of wing development that is exactly like that of 

 the Holometabola. In this group also there is a more or less pro- 

 nounced change of form, a loss of larval characters and the assumption 

 of adult ones. In the face of such examples as these one is led to 

 doubt seriously the accuracy of the wide application of the terms Hetero- 

 metabola and Holometabola. Such conflicting phenomena are not con- 

 fined to the Heterometabola, but occur in insects with supposedly 

 complete metamorphosis, there being many records of beetle larvae 

 with wings, Dewitz (1883) and others. In forms like the meal 

 worm a constant percentage have the wings external in the larva, 

 while in forms like the Staphylinidae there is scarcely any external 

 change from larva to adult, certainly no more than occurs in the 

 last moult of many insects with an incomplete metamorphosis. 



CoMSTOCK & Needham (1899) have advanced the view that the 

 ■wings of all insects are developed upon one fundamental type, and 



