274 THE AMERICAN NATURALIST. 
origin was entirely independent of the parapodia. A study of 
Peripatus appears to throw no light on the subject under dis- 
cussion, beyond indicating that in using the term “ primitive 
metamere °” we must not exclude the idea that such a metamere 
may in turn represent a fusion of an indefinite number of 
annuli. Evidence is thus added to a segment-fusion theory 
rather than to the biramose theory. 
While it appears possible that the exopodite and entopodite 
in Crustacea may have developed from dorsal and ventral para- 
podia, in direct contrast to the manner indicated above for the 
Chilopoda and Hexapoda, further study may show a common 
origin of the two processes. 
SUMMARY. 
In Hexapoda and Chilopoda the “coxa” is composed of two 
more or less fused segments, coxa genuina and meron. 
The antecoxal piece results from the chitinization of the 
membrane connecting the coxa with the sternum. 
The trochantin probably originated from a lateral portion of 
the same membrane. 
Audouin erroneously homologized the lateral margin of the 
posterior coxa in Dytiscus circumflexus with the trochantin of 
the prothorax and mesothorax. s 
The trochanter represents a distinct segment of the legs. 
The meron and cora genuina, together with their corre- 
sponding basal segments epimeron and episternum, give evi- 
dence of a fusion between two primary metameres in the 
Hexapoda and Chilopoda. 
In Hexapoda and Chilopoda the anterior metamere bears the 
functional, and the posterior, the rudimentary leg. 
Of the primitive Hexapoda, Neuroptera [Planipennia] exhibit 
the most generalized condition in the development of the coxa, 
while in Thysanura and ers sel a high pris of speciali- 
zation is shown. 
ANATOMICAL LABORATORY, BROWN UNIVERSITY. 
October 27, 1899. — 
