TRANSFORMATIONS OF THE INSECT. 67 



and thorax together are yet, as still more in the previous 

 stage, much smaller than in the pupa, and there is still a con- 

 tinuous curve from the tip of the abdomen to the head. (0, 

 antenna; /*, lingua, maxilla;, and palpi; i, fore-legs; j, mid- 

 dle legs ; A 1 , meso-scutum ; /, meso-scutellum ; ?i, spiracle of 

 the .propodeum.) 



In a succeeding stage (Fig. 66) of the semi-pupa, the head 

 and thorax together nearly equal in size the abdomen, and the 

 propodeum (c) has become entirely transferred to the thorax. 

 The head has become greatly enlarged ; the rings are very un- 

 equal, the hinder pair are much smaller, and overlaid by the 

 anterior pair ; the three terminal pair of abdominal rings, so 

 large in Fig. 65, have been absorbed, and partially inclosed in 

 the cavity of the abdomen ; and there has been a farther dif- 

 ferentiation of the ring into the sternite ((/), pleurite (e), and 

 tergitc (/). (a, eye; /*, lingua; o, ovipositor, two outer 

 rhabdites exposed to view.) The abdominal spiracles in Figs. 

 65 and f>(5, are represented by a row of dots. In the pupa 

 they are concealed by the tergites, which overlap the sternites. 



Fig. 67 represents the pupa state, where the body has become 

 much shorter, and the appendages of the head and thorax greatly 

 differentiated ; the external genital organs are wholly retracted 

 within the cavity of the abdomen ; the head is freer from the 

 body, and the whole bulk of the head and thorax together, in- 

 cluding the appendages, greater than that of the abdomen. 

 These changes of form, assumed by the insect in its passage 

 from the larva to the pupa state, are nearly as striking as 

 the so-called "hypermetainorphosis" of Meloe and Sitaris 

 described by Newport and Fabre. (/, mesoscutellum ; p, cly- 

 peus ; g, maxilla; with the palpi ; r, lingua.) 



We have also observed similar changes in the semi-pupa of a 

 Tineid larva, which we found in the mud-cells of Odynerus 

 cdbo2)halemtus. There were over a dozen specimens in different 

 stages of growth from the larva to the pupa, which were but 

 partially panily/ed by the well-directed sting of the intelligent 

 wasp, so that some continued to transform into perfect pupae. 



The following changes were noticed: the larva straightened 

 out, and became a little shorter, the prothoracic ring remaining 

 the same ; the head of the pupa being beneath it ; the meso- 



