( 53.5 ) 



large fla[), the epimerou. The flap, then-tore, has the position which the epimeron 

 of other insects generally has. In many insects the episternum takes part in the 

 formation of the coxal cavity, hnt in others it does not. The plate I have termed 

 the episternnm does not take part in forming the coxal cavity. Its position, however, 

 is dorsal to the lateral portion of the nietasternnm, as is the case in other insects. 

 This plate can therefore very well be tlie homologne of the episternum, in spite of 

 the distance at which it is placed from the coxal cavity. The mesosternum (Fig. 1) 

 of C. elongatus consists of two distinctly separated j)ieces, a ventral and a lateral 

 one, designated in the figure msst and epst + epm. The suture separating the 

 two is only marked ventrally in T. agyt-tes, at the very edge of the plates above the 

 coxa. The large lateral plate is internally divided by a ridge (punctured in Fig. 1) 

 into an anterior and posterior piece ; the ridge resembles those which are observed 

 at all sutures of the thorax. It is, therefore, possible that this internal ridge of the 

 lateral jdate of the mesothorax is an indication of a suture of which all traces are 

 now lost externally. From this we might infer that the lateral plate originated 

 by a fusion of two plates both of which reached the coxal cavity. These two 

 plates would have had the position which the episternnm and epimeron of other 

 insects have. The posterior of the two (epimeron) would correspoml with the flap 

 (epimeron) of the metathorax. The mesothoracic epimeron has no stigma on its 

 upper side, but there is a stigma on the membrane behind the coxa which is covered 

 in a lateral view by the lower edge of the epimeron. This membrane is strengthened 

 by a piece of chitin that rnns from the epimeron to or almost to tlie stigma. It 

 might be urged that this piece of chitin was the homologne of the metathoracical 

 flap, and the circumstance that it is the supporter of the mesothoracical stigma seems 

 to be in favour of this contention. I think it therefore best to leave it for the present 

 undecided whether the plate designated in Fig. 1 as epst + epm, = episternnm + 

 epimeron, corresponds to both the episternum and the epimei'on, or whetlier it is 

 the episternum, and the chitinous piece near the stigma represents the epimeron. 

 The question cannot be solved without a full comparison of many difterent forms of 

 fleas, which I have not yet carried out. 



The epimera of the metathorax extend so far back that the short raetanotum 

 does not cover the space between itself and the metasternum. On this account the 

 first abdominal segment has been pushed forward to form a dorsal covering to the 

 metathorax. The ventral j)late of tlie segment is wanting. Landois, /.r., mistook 

 the epimera of the metathorax for the ventral plate, as I have already mentioned. 



In addition to the first abdominal segment there appear to be nine more 

 segments, of which the last three (or four) are largely modified in both sexes. 

 The first seven tergites, like those of the thorax, have two principal rows of bristles, 

 as shown in f 1. 3. 4. (i. T. H. The lowest bristle of the posterior series is placed 

 below the stigma. The sternites, on the oiher hand, are provided witji a few bristles. 

 In fact the sternite of the second segment of T. agyrtes has one or two bristles 

 only. All these bristles lie close u])on the segments, and the rows of long 

 ones have perhaps the function of preventing the hairs of the host getting 

 between the segments. At the apical edge of the seventh tergite there are dorsally. 

 on each side, one long and two short bristles, which may possibly serve as a 

 protection for the peculiar sensorial plate of the pygidium. In other species of 

 fleas the number and proportional length of these antepygidial bristles varies 

 considerably, and t'nrnish obvious distinguishing characters. In the sexes these 

 bristles are also nftcn different. Ti//ililopsi/Ua pentacanthus. i'or instance, has no 



