208 
as paired processes, which later on amalgamate; the evidence 
indicates that it represents the entire labium and not the palpi 
only. In many adult Heteroptera there are four divisions, which, 
by their position, are evidently homologous to the submentum, 
mentum, and ligula, the latter consisting of two segments which 
are probably the subgalea and the amalgamated paraglossa and 
lacinia. In Belostoma there is a pair of small, simple processes 
on the subgalea which are considered by some to represent the 
palpi. From the base of the mandibles a sulcus has sunk into 
the head running toward the antennae, and the head capsule 
has grown over the base of the mouth parts. This obscures the 
homologies. Although the order is very large and the head 
undergoes great changes in the various groups, yet the shape, 
arrangement, and function of the mouth organs remain more 
constant than in any other of the larger orders of insects. 
By the character of the mouth organs the Hemiptera are dis- 
tinctly separated from all other insects. The Thysanoptera come 
nearest to them, but they are so distinct that they can have no 
direct phylogenetic connection, only an indirect one, through a 
remote common ancestor from which both may have evolved 
along somewhat similar but distinct lines. The Psocidae pos- 
sess a semi-free maxillary rod, which may be the starting point 
of the maxillary seta, but they can only be related through a 
very distant common ancestor. The similarity between the wings 
of Psocidae and Psyllidae is due to convergence or parallel 
development, and has no phylogenetic significance. If the Mal- 
lophaga have any relationship to the Anoplura, then the latter 
can have no relationship with Hemiptera. The sucking mouth 
of the Anoplura appears to be built on a very different plan 
from that of the Hemiptera, and the Siphunculata most cer- 
tainly are very different. I consider that the Hemiptera are the 
most isolated of all the large orders and their origin is obscure. 
That such highly specialized mouth organs, with their special 
line of development going back into the embryo, could have 
originated independently in two or more different groups, is 
highly improbable, so we must, therefore, consider the Hemip- 
tera to form a monophyletic order, very ancient and very 
isolated. That they are ancient is demonstrated by the fact that 
in the Trias of Australia the divisions of Cicadidae, Cercopidae, 
