402 A. SEDGWICK. 
different species-groups that we have before met with in each 
of the species-groups, together with a certain number of 
entirely pecular features. In assigning to Chilio-Peri- 
patus (Peripatus Blainvillei) independent rank I am 
at issue with Bouvier. He associates it with the South 
African species Peripatus cinctipes in Purcell’s genus 
Opisthopatus. He bases this association on the character of 
the ovum (considerably smaller than that of cinctipes), on 
the arrangement of the distal pedal papillae (found also in 
Melano- and Austro-Peripatus), and on the presence of a 
very minute receptaculum seminis. But, as stated above, the 
ovum is much smaller than that of cinctipes, approaching 
more closely to that of Neo-Peripatus, and it is doubtful if 
the very small dilatation called receptaculum seminis is really 
homologous with that structure in other species. JI am quite 
unable, therefore, to admit the association of Blainvillei 
with cinctipes any more than I can admit the genus 
Opisthopatus itself. It seems to me clear that Bouvier, 
in placing it with cinctipes, has been actuated largely by 
his hypothetical views as to the place of origin and evolution 
of the genus. These views, based, as such views generally 
are, upon a priori considerations as to the primitiveness 
of certain characters and the more recent origin of others, 
have no real validity, and tend to obscure the important 
issues, upon which the facts of classification and distribution 
can help to throw light. The futility of this method of 
speculation is further emphasised by the fact that the results 
obtained by it vary directly according to the theoretical bias 
of the author. Bouvier, who regards the presence of the 
maximum number of structures (1907, p. 70), whether legs, 
segments, spinous pads, etc., as a sign of primitiveness, 
places Neo-Peripatus nearest a supposed original form, 
whereas Evans (190la, p. 525), who considers a large and 
heavily yolked ege as an antique character, regards as the 
most primitive group EHo-Peripatus, which by Bouvier is 
regarded as the most advanced. Such diverse results may 
well inspire doubt as to the value of the method employed. 
