460 MEDICAL ENTOMOLOGY 



of two distinct portions of approximately equal length, the first of 

 which is very narrow, little wider in fact than the oesophagus, while 

 the second and terminal part, which may be taken as the rectum, 

 is wider. The narrow part of the hind gut has on it near its termination 

 a small clump of round cells (fig. 8), of unknown function and 

 origin, placed at one side. This portion does not become directly 

 continuous with the rectum, but joins it a short distance from its 

 anterior end. The wall of the rectum contains many muscular fibres, 

 arranged obliquely. There are no rectal papillae. 



The salivary glands (Plate LVII, fig. 6) also present a remarkable 



similarity to those of the adult. There are two glands on each side ; they 



are tubular, elongate, slightly swollen at the blind ends, 



a ivary g an s o ^ another at their narrow ends without 



larva J 



intervening ducts. From their point of union a fine 



salivary duct arises, and passes to the mouth, near which it unites 

 with its fellow from the opposite side. About midway on the course 

 of each of the salivary ducts there is a fusiform dilatation, equal in 

 width to the widest part of the gland. The walls of the glands are 

 formed of a single layer of cubical cells, surrounding a central lumen. 

 The wall of the dilatation on the duct, on the other hand, is extremely 

 thin, and contains only a few very small and flattened cells, its structure 

 suggesting that it functions as a reservoir. 



The food of the larva is something of a mystery. In a large pro- 

 portion of the larvae which the writers have examined the intestine 



has contained blood, in stages varying from fresh 

 Food of the larva J , . 



unclotted blood to particles of blood pigment ; the 



dark colour of the living larva is due to the presence of this material. 

 Similar observations have been made on the larva of Xenopsylla cheopis, 

 Ceratophyllus alladinis. from the Indian squirrel (Funambulus palina- 

 rum), and in an undetermined species from Gerbillus indicus, so 

 that the circumstance is by no means peculiar to the cat flea. It 

 has never been suggested that the larva is a blood-sucker, and it has 

 no biting mouth parts ; the most probable source of the blood is the 

 adult flea. It is well known that the flea, like most blood-sucking 

 insects, has the habit of defaecating while feeding,* or shortly after ; 

 not only is the hind gut emptied on these occasions, but fresh red 



* Noller, experimenting with the dog flea, found that the insect defaecated from 

 seven to fifteen times during a feed lasting two and a half to three and a half hours, 

 the last material voided being red fluid blood. Pulex irritans was found to defaecate ten to 

 twenty times in half an hour. 



