346 



of which is formed late in life. The only apparent difference 

 between the structure seen in Locusta and that found in 

 Nasonia lies in the fact that the two second appendages do not 

 unite (in Locusta) to form a boring organ, and that the third 

 appendages do not merely act as a protecting sheath for the 

 ovipositor, but actually enter into its formation. 



It is necessary to describe next the formation of the 

 copulatory organs in the male. 



So far as I could observe, no rudiments corresponding to 

 those of the female copulatory organs are present in the male 

 larvae, and no special differentiation of external male organs 

 takes place till very late in larval life. This is intimately 

 connected with certain changes in the last four abdominal 

 segments, which become so disposed as to allow of the 

 eversion of the penis. 



Shortly after the defaecation period the tenth and ninth 

 imaginal discs of the abdomen have grown so as to assume a 

 position at the rear of the animal and at the same time to take 

 only a very small part in the formation of the lateral, ventral, 

 or dorsal walls of the larva. This is brought about by the 

 fact that the growths of these two' abdominal imaginal discs are 

 not very extensive; and that they actually become partly invag- 

 inated into the eighth abdominal segment, which grows much 

 faster than they do. This is clearly seen in the section shown 

 in figs. 25 and 27. The tenth (terminal) segment is quite 

 small, and bears the anus; ventrally it is provided with a pair 

 of appendages, which lie in close contact with the ninth 

 segment, which segment is partly invaginated into the eighth. 

 This invagination is accompanied by a marked cell-prolifera- 

 tion in the integument of the invaginated portion (fig. 25)', 

 and is already clearly seen in the def aecating larva ; during 

 pupal life it develops into the penis. 



The segments then chitinise and are found, in this con- 

 dition, in the early pupa. It is especially worthy of notice 

 that already at this stage the two appendages of the tenth 

 (terminal) abdominal segment have applied themselves very 

 closely to the sternal portion of the ninth (fig. 27) ; indeed, 

 while the segments of the cuticle of the pupa are, in other 

 respects, an exact representation of the shape of the epidermis 

 which has secreted them, yet the appendages of the tenth 

 segment form with the sternal portion of the ninth a cuti- 

 cular covering which is common to them both ; an examina- 

 tion of the epidermal components of this compound cuticular 

 segment, however, reveals its true nature (fig. 27). In the 

 pupa, shortly after its formation, a curious change now takes 

 place, which results in the formation of the penis. Tlie cells 



