480 GEORGE H. CARPENTER. 
states to be always limbless. I confidently anticipate future 
proof that the central portions of the gnathochilarium really 
belong to this post-maxillary segment. In any case, this 
embryonic segment supports the view that the head of a 
Diplopod agrees closely with that of an Insect—an agreement 
now shown more clearly by the presence of maxillule in 
Polyxenus. And considering that Polyxenus, which shows 
this agreement so unmistakably, has exactly the same number 
of trunk-segments as Scolopendrella or a Thysanuran, we 
cannot doubt the near relationship of Insects to Millipedes. 
Isolated Position of the Malacopoda. - 
The Malacopoda (Peripatidee) must certainly be separated 
widely from the tracheates, as a grade of Arthropoda far 
lower than any other class. Belief in the polyphyletic origin 
of Arthropods has been largely supported by a supposed 
close relationship between the Malacopoda, the Chilopoda, 
and the Insecta. But a study of the characters of the 
‘““Protarthropoda” and ‘ Kuarthropoda” as given by 
Lankester (19, pp. 564-6) must surely convince the reader 
that Insects and Centipedes are much more nearly related to 
Crustaceans than to Peripatids. 
Relationship between Crustacea and Arachnida. 
My views on the general subject of the relationship between 
Insecta and Crustacea agree, then, in the main with Lankester’s, 
while we differ on some points of detail. It is even so with 
regard to our interpretation of the kinship between Crustacea 
and Arachnida. He suggests (19, p. 573) that the most 
primitive Crustaceans ‘‘ were not very far removed from the 
aquatic ancestors (Trilobites) of the Arachnida, but differed 
essentially from them by the higher specialisation of the 
head,’ while in my essay (8, p. 349) occurs the statement 
that “there is no difficulty in tracing back the Merostomata, 
the Xiphosura, and the Trilobita to a common ancestry; and 
thus the Arachnida as a class, like the Insecta, have been 
