DEVELOPMENT OF THE ALIMENTARY CANAL. 431 



the stomodasal portion of the alimentary canal opens into the 

 mesenteron, but the insects drink greedily an hour after they 

 emerge from the pupa. The crop is first distended with fluid, 

 but if the insect is fed on coloured syrup this passes into the 

 chyle stomach as soon, or almost as soon, as the insect is able 

 to fly. 



I believe the lumen of the oesophagus communicates with the 

 chyle stomach as early as the middle of the pupa state, and 

 there is apparently a continuous lumen at least two days before 

 the insect emerges from the pupa, although no fluid ever passes 

 back from the chyle stomach into the oesophagus during the 

 latter days of the pupa stage. 



Development of the Salivary (Sericterial) Glands. So far as I 

 know, no direct observations are extant as to the manner in 

 which these glands are developed in the egg, but they are 

 usually regarded as invaginations of the epiblast of the labium 

 or of that at the root of the maxillae or mandibles, and they are 

 seen as open epithelial tubes at an early stage of development. 



In the pupa, the tubular salivary glands of the imago of the 

 Blow-fly are developed according to Kowalevski [145] from 

 imaginal cells, which, he says, are found at the junction of the 

 duct and gland sac of the larva. Van Rees [147, p. 76] differs 

 widely from Kowalevski in his account of the manner in which 

 the sericteria of the larva disappear, for whilst the latter states 

 that they remain attached to their duct until after the develop- 

 ment of the head, Van Rees figures and describes a new duct 

 at this period which terminates in a blind end in the basal part 

 of the proboscis, and says the degenerated gland masses are 

 already separated from their ducts. 



I have carefully examined the new duct, and in young nymphs 

 it is certainly not seen in transverse sections, except near the 

 distal part of the rostrum. It appears to me to be an entirely 

 new formation, having no connection with the larval sericteria. 



At a later stage, such as is represented in PL XXL, the blind 

 end of this duct is connected by solid cell strings, which 

 surround the cephalo-thoracic nerve cord with two masses of 

 cells, one on either side of the oesophagus. It is from these 



29 



