AND ANATOMY OF CERTAIN HYMENOPTERA PARASITICA. 395 



probability the oesophageal valve can be used effectively to prevent regurgi- 

 tation o£ food, when this is desired. The mesenteron is widest near the 

 thoracic region and tapers backwards imperceptibly, being attached in the 

 ultimate abdominal region to the ventral inner surface of the abdominal 

 vesicle (PI. 24. fig. 6 and PI. 25. fig. 10). There is no difiPerentiation of the 

 hind region of the gut into a proctodseurii (rectum)^ and no apparent 

 Malpighian tubules could be discovered. The gut is drawn in PL 24. 

 fig. 6, M, and in Pi. 26. fig. 24. The hinder region of the gut is attached to 

 the bladder, as shown in PI. 25. fig. 10 in sagittal section, and there is now no 

 vent or anus. In the region of attachment of the hind region of the gut are 

 muscle bannerets which pass from the external muscle-layer of the gut to the 

 connective tissue which lines the inner walls of the vesicle. These muscles 

 are not apparently of the same nature as the circular muscles of the gut. 

 In PI. 25. fig. 11 the gut (GT) is cut in transverse section in the hind-thoracic 

 region. • The black bodies are the nuclei. The gut contains a finely granular 

 mass of food, with oil}^ vacuoles here and there, and occasionally the 

 fragmentary nuclei of the masticated fat-body cells of the host caterpillar. 



In PI. 25. fig. 10, the slightly oblique section of the hinder region of the 

 Microgaster connexus larva illustrates very clearly the disposition of fat in 

 the gut and body. This larva was preserved in Flemming^s fluid, and the 

 osmic acid has blackened the fat. In the gut the finely granular food is 

 seen to have an immense number of fat droplets in it. The gut-wall itself 

 has a large number of big fat droplets in it. The fat-body (FB) of the larva 

 also contains a great deal of fat. Fat-absorption takes p'ace along the whole 

 length of the gut, right to the region of attachment of the hinder gut to the 

 wall of the vesicle. It will be noticed in PJ. 25. fig. 10 that the vesicle 

 contains no fat, the dark spots in it being the large nuclei. 



In PI. 26. fig. 22 (MN) the histological structure of the gut-cells may be 

 seen ; each gut-cell (MN) is very large, extremely vacuolated, and the 

 nucleus is a crenated structure^ as is fairly common in fat-cells but rare in 

 the alimentary-tract cells. 



In most insects though the proctodseum and stomodseum are lined with 

 chitin, the mesenteron also has a modified lining or intima, which, if it is 

 not chitinous, quite often resembles chitin. In some cases, as in the 

 mesenteron of the adult Braconid {Apliidius), I have discovered a lining 

 which is so like a ciliated layer, that all those to whom T have shown the 

 sections have pronounced them to be cilia or wonderfully like them. But 

 when the live animals were teased up in salt solution, though the pseudo- 

 ciliated lining could be seen, there was no ciliary action. Statements with 

 regard to the nature of the lining in the gut of Microgaster are difficult to 

 formulate. The stomodseum is lined by a very thick layer of chitin (text- 

 fig. 14) ; the entire length of the mesenteron also seems to have a 

 chitinous (?) intima of the thickness of about half that of the oesophagus 



