370 PROCEEDINGS OF THE NATIONAL MUSEUM. vol.xxvii. 



ill thoracic or abdominal ctenidia is quite likely to vary two or even 

 four ill ,some species, especially in those species in which the ctenidia 

 are laterally reduced. 



The exact homologies of the genital organs have yet to be worked 

 out, though Rothschild has taken a long step in the right direction. 

 In the present paper the terms ''lateral portion of ninth tergite" and 

 "tenth tergite" are used as applied by Rothschild. The upper and 

 most conspicuous pair of genital appendages are called "upper 

 claspers" and the paired organs, immediatel}^ below these, but more 

 inconspicuous and retractile, "lower claspers." 



Of the later authors each has had a ditferent method of stating the 

 comparative lengths of the tarsal joints. It is a matter in which 

 minute exactness is not only undesirable but impracticable, owing to 

 the unevenness of the ends of the joints. In the present paper the 

 proportion is given in terms of the fourth joint, which is given the 

 empirical value of 5, due to its great similarit}' in size in the difi'erent 

 legs and in difl'erent species. The measurements were taken hy a 

 camera lucida and the results reduced to the terms mentioned above. 

 In the author's practice this has greatly simplilied the comparison of 

 species. 



It may be noted that the rather strongly chitinized seminal vesicle 

 in the females, which retains its shape after treatment with caustic 

 potash, possesses a ver}" characteristic form in mau}^ of the species. 

 After mounting, it rests in various positions and this makes its com- 

 parative stud}' very difficult. 



CLASSIFICATION. 



Linnreus began with one genus, Pulex^ and one species, irritans^ 

 which will represent the t3^pe of Pul&x alwa3's. Linnaeus afterwards 

 described P. jyenetrans. The first separation of penetrans as a dis- 

 tinct genus occurred in 1815 « under the name Phynchoprion. I do not 

 know why this name has been rejected. If there is no question as to 

 its application, then it must be used instead of Sarcojjsylla and the 

 family name will also change. 



The dismembering of Pulex began in 1832 with the Ceratoj)hylJn.s of 

 Curtis, and seven generic names were proposed b}- Kolenati. Most of 

 these names represented somewhat artificial groups and were poorly 

 defined. The reaction came with Taschenberg, who disregarded all 

 the latter genera and returned to the Linna^an Pulex for the greater 

 portion of the species, Taschenberg is not, however, altogether con- 

 sistent, although one can readily understand why he was tempted 

 to reject all the inadequate work of his predecessors. However, 

 Taschenberg describes IlystrichopsyUa and establishes a new genus, 

 Tyjjhlopsylla, for three groups com))ined, each of which had previ- 

 ously received a generic name. 



"Oken, Naturg. f. alle Stiinde, III, p. 402. 



