422 THE ALIMENTARY CANAL OF THE IMAGO. 



daeum, since it remains a comparatively simple tube, and 

 exhibits at all times in the life of the insect the characters of 

 an epiblastic structure ; a flat pavement epithelium and a dis- 

 tinct chitinous intirna. It is true that it is somewhat difficult 

 to define the precise limit of the stomodaeum in the proventri- 

 culus, and to determine how far the development of this organ 

 is from epiblastic and how far from hypoblastic cells ; but if 

 we adopt the evidence afforded by the character of the epithe- 

 lium, it is clearly chiefly developed from the hypoblast of the 

 mesenteron a view which is rendered probable by the manner 

 in which the development of the proventriculus takes place. 



It is quite another matter, however, to determine the fate of 

 the proctodaeal involution. Many observers, including myself, 

 have seen a distinct proctodaeal involution in the early stages 

 of embryonic life. In the Blow-fly embryo, however, Graber 

 has evidently regarded my blastopore as an anus and my 

 metenteron as a proctodaeum. If the section from which my 

 figure (Fig. 2) is taken is to be interpreted as I have inter- 

 preted it, there is clearly a proctodaeum, which is quite distinct 

 from the dorsal invagination ; but I have been quite unable to 

 determine its fate. 



I have carefully searched for evidence that the hind-gut is 

 formed from the proctodaeal involution ; and I am not aware 

 that anyone has been more successful than myself in actually 

 tracing the development of the hind-gut (metenteron) to the 

 proctodaeum. Biitschli's figures [126, Taf. XXVII.] are far from 

 convincing as to the real origin of the Malpighian tubes, espe- 

 cially Fig. 31, b, in which these tubes appear to arise from the 

 mid-gut ; and his figures of a still earlier stage are capable of 

 another interpretation (see p. 426). 



The frequently repeated text-book statements that the Mal- 

 pighian vessels originate from the anterior extremity of the 

 proctodaium rest upon indirect, rather than direct, evidence. 



Although Swammcrdam figured the alimentary canal of 

 the Bee larva [4, PI. XXIV., Fig. 6], and represented it 

 as a continuous alimentary tract, it has been more recently 

 described as consisting of a proctodaeum which is blind at its 



