DEVELOPMENT OF THE ALIMENTARY CANAL. 427 



(pp. 331-2). This is subsequently converted into an epithelial 

 sac, the epithelial elements of which are apparently derived 

 from three groups of embryonic cells, which can be made out 

 in the alimentary canal of the larva; the histoblasts of the 

 proventriculus, of the chyle stomach, and the ring at the 

 orifices of the Malpighian vessels. 



It is by no means easy to understand how these groups of 

 cells become a portion of the mesenteric sac, but it appears to 

 me probable that the proventricular ring originates the anterior 

 extremity of the epithelial sac ; the manner in which the cells 

 of the chyle stomach form a continuous layer on the inner 

 surface of the sac has been already described (p. 332% but the 

 greatest difficulty is to understand how the ring at the orifices of 

 the Malpighian vessels is related to the mesenteric sac. The 

 sac certainly contains both the distal and proximal intestine, as 

 well as the remains of the Malpighian tubes of the larva. It 

 appears to me probable that the distribution of the tracheal 

 vessels is concerned in the process, and that the new para- 

 blastic tissue is brought into relation with the ring through the 

 medium of the tracheal trunks which supply it. 



It is certain that those branches of the tracheae which are 

 connected with the imaginal rudiments preserve an embryonic 

 structure, and have an external coat, consisting of very small 

 cells, totally unlike the peritoneal coat of the vessels which 

 undergo rapid degeneration during the development of the 

 pronymph. Hitherto, however, I have entirely failed to sub- 

 stantiate by observations the manner in which the Malpighian 

 vessels of the larva are related to their successors, the Mal- 

 pighian vessels of the imago. 



It may be admitted that the ring of small cells discovered by 

 Ganin, and described by Kowalevski, which I will term Ganin's 

 ring, are the rudiments of the Malpighian vessels and meten- 

 teron of the imago, as this appears most probable, but I have 

 entirely failed to trace the changes which this ring undergoes. 



When the mesenteric sac is fully developed, no traces of 

 intestine or Malpighian vessels are seen in sections outside the 

 sac, and the whole of the structures concerned are so friable 



