32 THOMPSON YATES AND JOHNSTON LABORATORIES REPORT 
more reduced state ot the bristles two lines of development can be traced. In the genus 
Dermatophilus the dorsal bristles have disappeared, only one bristle above the stigma 
remaining. In Hectopsylla and Echidnophaga^ however, the most ventral bristles have 
been lost, the dorsal ones remaining. In the last two genera, with the exception of 
the $ Hectopsylla psittaci, previously referred to, there are two bristles or only one, 
the first and the seventh tergites having often three bristles on each side. The lost 
bristles are often replaced by extremely minute hairs. The subdorsal bristle of the 
anterior tergites is rather stout in several species (as, for instance, in C. gallinaceus), 
being stouter than the bristles on the posterior segments. The bristles are placed 
usually on or near the middle of the segments ; in some species, however, they have 
become shifted towards the base. While in Echidnophaga and Dermatophilus the 
tergites of segments one to seven bear each at least one bristle, the middle tergites of 
Hectopsylla have lost the bristles entirely. 
Besides these bristles the seventh tergite bears in the Siphonaptera as a rule at 
least one long apical bristle standing usually on a cone, the bristle being rarely missing. 
This apical bristle with its two usual companions is placed in Pulex at a small distance 
from the apical edge of the segment, this edge not being emarginate, as is the case in 
those Siphonaptera in which these bristles are quite apical. Among the Sarcopsyllidae 
only the genus Echidnophaga has preserved a single apical bristle, which is present in 
both sexes, the bristle being sometimes much longer than the other bristles of the 
same segment, sometimes the same size, or smaller, according to the species. 
The bristles on the abdominal sternites of the Sarcopsyllidae are always few in 
number, the proximal sternites being often devoid of bristles, though there are 
generally one or more extremely minute hairs present. 
The swollen abdomen of the pregnant female of the Sarcopsyllidae is such a 
very conspicuous character in Dermatophilus penetrans that in the early history of the 
classification of Siphonaptera it has been advanced as the main distinction from the 
genus Pulex. We now know that the swollen abdomen of the % is not confined to 
Sarcopsyllids, and that not all the Sarcopsyllid females become largely extended. In 
the swollen abdomen of pregnant %% of insects generally it can be noticed that the dorsal 
and ventral chitinized plates of each segment and the segments inter se have become 
separated, the connecting membranes being exposed to view. The last segment or 
segments and also the basal one are usually not extended or not to the same degree 
as the middle segments. This normally extended abdomen of the female is found in 
the family Pulicidae in a number of genera, such as Malacopsylla, Vermipsylla, and 
Chaetopsylla, the dorsal and ventral plates being more or less widely apart. As a 
matter of course the abdomen of the pregnant $ is in all Siphonaptera swollen to some 
extent, but the plates do not become separated except in the comparatively few species 
with sedentary habits. Now, in Sarcopsyllidae the swelling of the abdomen is different 
in each of the three genera into which we have divided the family. The swelling is 
