22 THOMPSON YATES AND JOHNSTON LABORATORIES REPORT 
these two different lines of evolution are represented on the one hand by the species 
of the genera Hectopsylla and Dermatophilus, and on the other by the species of the 
genus Echidnophaga. 
In consequence of the great dorsal reduction of the thorax, the distance from the 
anterior edge of the prosternum to the anterior edge of the prothoracical tergite is 
longer than is ordinarily the case in Siphonaptera. The head, being adapted to this 
modification, has grown backwards in Dermatophilus and Hectopsylla, its hinder and 
genal edges having become oblique. This has had the effect of pushing the mouth 
upwards, making it subterminal, and therefore the dorsal line of the head more flat 
than in the Siphonaptera with subventral mouth. This development is most pro- 
nounced in Dermatophilus. The piercing organs of the mouth (labrum and mandibles) 
which are normally directed obliquely downwards in the Siphonaptera, have here 
assumed another position, being directed obliquely forward. 
Now, in Echidnophaga the same result is attained in another way. In this genus 
the reduction in length observed in the thorax takes place also in the occiput. 
The relative length of the frons and occiput is variable in Siphonaptera. In the 
family Pulicidae the frons is as long as, or shorter than, the occiput ; the opposite, 
however, is also found in several species. In the Sarcopsyllids the occiput appears to 
be never longer than the frons, it being shorter than the frons in all Echidnophaga 
and also a little shorter than the frons in Hectopsylla. The occiput is shortest in 
Echidnophaga gallinaceus and its close allies. The difference in length is due to a re- 
duction of the occiput, not to an enlargement of the frons. The reduction proceeds 
from the hinder edge of the occiput, as seems to us to be indicated by the outline of 
this edge. In the Pulicidae this edge is always entire, being somewhat curved ven- 
trally at the antennal groove, and more or less straight further upwards. The strongly 
slanting occipital edge of Hectopsylla and Dermatophilus is also entire. In Echidnophaga, 
however, especially in the species with short occiput, the hinder edge of the latter is 
more or less convex in the middle, the occiput being more reduced in length dorsally 
and ventrally than in the middle, the edge being left standing here. This convex 
portion is, in the $ of Echidnophaga gallinaceus and some other species, concentrated, 
as it were, to a prominent lateral lobe, already referred to above (PI. I, Fig. I and 2), 
Now, the reduction of the occiput has a peculiar influence on the development of 
the frons. As the hinder edge of the occiput and the genal edge of the frons meet 
behind the antennal groove, it is obvious that the shortening of the occiput must be 
accompanied by an alteration in the direction of the genal edge. This edge has 
assumed a more oblique position than it originally had in Siphonaptera, and the oral 
edge has correspondingly become longer. In consequence of this enlargement of the 
oral edge the mouthparts are further away from the prosternum and forecoxae than 
they would normally be, if the shortening of the occiput had not been counteracted 
by the enlargement of the oral edge. 
