36 THOMPSON YATES AND JOHNSTON LABORATORIES REPOR T 
The tarsi of the Sarcopsyllidae represents, like the tibiae, various degrees of develop- 
ment from a Pulex-Y\ke tarsus to the highly modified tarsus of Dermatopbilus, There 
are no hairs on the ventral surface of the tarsal segments, except at the lateral edges. 
The proportional length of the segments varies to a certain extent, the fore and mid 
tarsi agreeing with those or Pulex and allied genera in the first segment being shorter 
than the second. The difference in length is, however, very slight in Dermatophilus y 
which has very slender tarsi. The most interesting differences between the various 
species are exhibited by the fifth segment and the claw. 
In most Siphonaptem there is a smaller or larger number of minute hairs on the 
ventral surface of the fifth tarsal segment. These hairs are absent from all the Sarco- 
psyllidae. This negative character, however, is not confined to that family, there being 
only a few such hairs in Pulex irritans, pallidas, and allies, and none in Lycopsylla and 
Malacopsylla. 
The fifth segment bears vcntrally at the apical edge from 4 to o bristles in 
Siphonaptera. The highest number in the genus Pulex is three, one in the middle and 
one on each side or it, the lateral ones being usually somewhat proximal of the 
middle one. Of these three bristles a lateral one has disappeared in most species of 
Pulex, the other two either retaining their position towards each other or the proximal 
(or lateral) one moving distad so as to be on a level with the median one. In the 
Sarcopsyllidae we find never more than two bristles, which stand close together, the 
outer one being usually a little behind the other. This pair of apical bristles is present 
in most Echidnophaga, in which genus the bristles sometimes vary in size according 
to the species. They also occur in two of the three species of Hectopsylla and in 
Derma tophi lus penetrans (the tarsi of Dermatophilus caecata are not known to us). In 
three species of Echidnophaga the outer bristle is lost, the median one alone remaining, 
while in Hectopsylla pulex both bristles have disappeared. Of the three Echidnophaga 
which possess one such bristle only two agree so closely with E. gallinaceus that their 
specific distinctness is not beyond doubt. This being so, it may be concluded that 
among a sufficiently large number of specimens of E. gallinaceus there will be found 
individuals with one such bristle only, at least on one of the tarsi. On the other 
hand, among the specimens of those species which have normally one bristle 
occasionally an individual will possibly be found with two bristles. 
The lateral bristles of the fifth tarsal segment are in Siphonaptera very often of 
importance for the characterization of species, affording in many cases specific distinc- 
tions which are easily recognized. Near the apex of the segment there is always a 
thin hair on each side, placed on the dorsal surface of the segment. This subapical 
hair is usually long, but sometimes rather short. It is present in all the Sarcopsyllidae, 
with the exception of Dermatophilus. Between this hair and the base of the segment, 
but on the ventral side, there is in the Pulicidae a row of four, five, or six bristles, 
which are usually short and stout. The first bristle is often shifted towards the middle 
