Significance of the Somiiic Constitution, etc., in the Hirudinea. 153 
We now pass on to consider the possible relation of the foregoing 
remarks to the Arthropoda. 
The question as to whether the phylum Arthropoda with its apparently 
varied segmentation, progoneate and opisthogoneate characters, excretory 
and respiratory systems, constitutes a natural group is to a large extent 
answered by the apparently fundamental similarity in its various classes in 
regard to the number of segments. 
This can be made out in all of the classes, although in all except the 
Insecta we find numerous variations due either to the absorption or 
addition of segments. 
The Malacostraca, Xiphosura, Scorpionida, Insecta, Symphylid Myria- 
poda, and some species of Peripatus (those bearing 14 pairs of walking 
legs) can be made out with good reason as having a body constituted by 
twenty-one segments, together, in many, with a telson. 
This is no dogmatic attempt to state exclusively that all the various 
divisions of the phylum are necessarily monophyletic, but rather an 
attempt to show that there is good reason for believing that a funda- 
mentally similar somiiic constitution traverses the phylum, and conse- 
quently that we can regard the phylum as descended from ancestors 
possessing 21 somites. There is certainly strong difference of opinion 
in regard to the somitic constitution of the Arthropodan head, but as 
this will not leave a bigger indiscrepancy than one segment when the total 
is made up, a discussion of the same can be omitted here since it does not 
fundamentally concern the substance of this paper. 
For a long time it was thought that the Entomostraca — more par- 
ticularly the Phyllopoda — necessarily constituted the primitive types of 
Crustacea, but a study of the variation in segmentation in primitive 
Trilobita passing upwards from the Cambrian, and the early appearance of 
the Merostomatous Arachnida, with their constant number of somites, are 
strong arguments against such a conclusion in regard to segmentation. In 
Copepoda and Cirripedia we get an approach again towards a constant 
number of postcephalic segments, and although this number may not agree 
absolutely with that for the Malacostraca, yet it supports the idea of a 
constant somitic constitution rather than the varied segmentation of the 
Phyllopoda, and, further, the number of thoracic segments is practically 
similar in Copepoda and Cirripedia. We may see in the presence of more 
than one pair of appendages in an abdominal segment of Apus a beginning 
in the increase of the number of segments, andin the Ostracoda and 
Cladocera an absorption of segments just as easily accomplished as the 
degeneration of the abdomen in Cirripedia and the loss of segmentation 
in the parasitic Copepoda. 
The same variation from the fundamental number of segments laid 
down for the phylum will be found in all the other classes. Perhaps the 
12 
