272 PROCEEDINGS OF THE NATIONAL MUSEUM vol. ii4 



if for no other reason than the position of the group on the borderline 

 between commensahsm and true parasitism, which ought to provide 

 good clues about the evolution of the latter. It is obvious, however, 

 that such studies can hardly be made until the classification of the 

 group has been worked out, and specific identities firmly established. 

 So far, no extensive work has been done in this direction. 



The present study was rather in the nature of a test to see if the 

 species of a widespread and somewhat heterogeneous branchiobdellid 

 genus might be defined on the basis of such characters of the repro- 

 ductive systems as have been found reliable in many other groups of 

 invertebrate animals. Traditionally, genera as well as species in 

 the Branchiobdellidae have been based on such characters as body 

 form, shape of the jaws, even size of the animals. Clearly it would 

 not take long to exhaust the possible combinations of such features, 

 and by 1950 the taxonomy of American branchiobdellids had reached 

 virtually an impasse. The major consideration underlying this 

 investigation, therefore, has been the determination of what con- 

 stitutes a species in the genus Cambarincola, and the development of 

 coherent diagnoses by which these units may be subsequently iden- 

 tified. I feel that these objectives have been achieved with a fair 

 measm'e of success. 



An inquiry into characters of sj'stematic value reveals that the 

 most important appear to be the gross form as well as histological 

 nature of the spermiducal gland and the adjoining prostate gland, 

 and that these structiu-es afford the basis for division of the genus 

 into sections and groups. The two largest sections seem mamly to 

 be composed of structurally generalized species on one hand and 

 presumably more speciaHzed forms on the other. That a satisfactory 

 grouping of species into a system approximating the evolutionary or 

 phylogenetic trends within the genus has been established is by no 

 means asserted, however. That future studies will impose changes 

 and modifications seems inevitable, yet a start must be made! 



The species of Cambarincola, as now defined on the basis of the 

 male reproductive organs, are remarkably constant for soft-bodied 

 animals, and well-preserved material can be identified with vu-tually 

 100 percent confidence. Most of the species that are known from 

 sufficient material are structurally miiform over their entire area of 

 distribution, and the geographic ranges are m all cases entirely 

 consistent with established biotic or physiographic provinces. The 

 continent-wide distributions published by previous workers are 

 largely the result of misidentifications. 



In addition to the purely taxonomic aspect of this work, it has been 

 possible to draw some inferences from present knowledge of structure 



