BIONOMICS OF FLEAS 461 



blood is passed, the insect appearing to take up more food than it 

 can digest. In this way blood is passed in considerable quantities 

 among the debris under the host, and is ingested by the larva. 

 Whether blood so obtained constitutes the sole or the main food of 

 the larva, or whether it is simply taken in when met with, it is 

 impossible to say, but of the actual presence of fresh mammalian 

 blood in the gut, often in a large amount, there is no doubt. 



Regarding the length of the larval stage exact data are not avail- 

 able, as it is a matter of great difficulty to separate out the eggs of one 

 batch. It appears, however, to be about ten days, rather longer in 

 cold weather, in C. felis and X. cheopis. At the end of this time 

 the larva spins for itself a cocoon like that of many Lepidoptera, 

 covering the outer wall with fragments of the debris among which 

 it lies. The pupal cases so formed are minute oval bodies, look- 

 ing just like collections of sand granules stuck together by moisture. 

 When they are opened, as can easily be done with a pair of sharp 

 needles, the pupa (Plate LVII, fig. 4) is found lying free within. When 

 nearly mature it [ resembles the adult ; the legs and 'palps lie folded 

 against the body, and the head is bent ventrally. It is white in colour, 

 and the segmentation of the exo-skeleton can be easily made out. At 

 the posterior end there are two pairs of short processes, on the last and 

 penultimate segments respectively, the nature of w^hich has not been 

 determined. The duration of the pupal stage is normally seven to ten 

 days. 



The larvae of Xenopsylla cheopis and Ceratophyllus alladinis resemble 

 those of C. fells. That of Echidnophaga gallinaceus (Plate LVII, fig. 

 15), on the other hand, simulates a Muscid larva more closely; the body 

 is longer and tapers more towards the anterior end, and has on it only 

 a very few short hairs. The alimentary canal is also longer and more 

 coiled. Specimens which have been examined by the writers, and 

 from which the imagines were bred, do not conform to the description of 

 the larva of Dermatophilus penetrans as given by Newstead, as one 

 would expect from the close relationship of the two genera. 



As is the case with most ectoparasites, each species of flea has a 

 particular host, the blood of which is its normal food. But the rela- 



tion between the different species of fleas and their 



... , c .1 Bionomics Relation 



hosts is not a close one, and m the absence ot their 



ordinary food most of them will feed readily enough 



on the blood of other hosts, either allied species or widely separated 



ones. In Madras, for instance, the flea which commonly attacks 



