302 THE PEARLY FRESH-WATER MUSSELS— SIMPSON. vol. xvm. 



mus of Panama, have radial beak sculpture, wbieli sometimes extends 

 well over the body of the shell. I know of no others having- this char- 

 acter excei^t Unio rotundatns^ Lamarck, of Texas and Louisiana, which 

 occasionally exhibits this peculiarity in a slight degree, and which, 

 singularly enough, by its form resembles many of those of South Amer- 

 ica. The Unios of New Zealand and Australia have, so far as I have 

 been able to observe, curved or imperfectly radial beak sculpture, 

 approaching somewhat that of several of the species of South America. 

 Nearly all the austral forms (excepting those of Africa) have peculiarly 

 compressed cardinal teeth, there being a single one in the right and 

 two in the left valve, sometimes slightly multifid, and between those 

 of the latter valve there is a parallel-sided pit, into which the cardinal 

 of the right valve fits. 



I believe that these characters of the shell and embryo, which seem 

 to be reasonably constant, will justify the separation of the Unios of 

 South America, Australia, New Zealand and Tasmania into a subgenus, 

 for which may be used the name Diplofhrn, applied by Spix to Uuio 

 eUipticus and TJ. rotundns of Brazil.' There can be but little doubt 

 that these belong to a different and perhaps older phylum than the 

 species of Europe, Asia, Africa and North America.' 



The writer has proposed for the American species ' a subdivision into 

 groups, which should contain species evidently allied by conchological, 

 anatomical and embryological characters. Each group he proposed to 

 call after some widely distributed, abundant and characteristic species 

 belonging to it. Thus an assemblage of solid, oval forms with radiating 

 stripes, common in the Mississippi Valley, is fairly typified by the well- 

 known XJnio lujamenUmis of Lamarck; another of large, rather light, 

 inflated forms from the same region, is represented by U. ventricosus, 

 Barnes; a third, consisting of compressed, rhomboid species of the 

 Atlantic drainage, by U. complanatus; and to speak of these difterent 

 divisions as the group of Unto liganientinus, U. occidens, or U. compla- 

 natus group, at once conveys to the mind of the merest novice Just what 

 is meant. 



The arrangement is not at all a new one, having been used more or 

 less by Lea, Lewis, Call, Marsh, and other conchologists. Eecently 

 Fischer and Crosse* in monographing the Mexican and Central Ameri- 

 can Anodontas and Unios, group them in the same way, but apply spe- 

 cial names to the sections. It seems to me that such names merely 

 tend to cumber the literature, and uselessly add to the labor of the 

 conchologist in committing them to memory. 



In arranging the Naiades of the National Museum, I have become 

 convinced that this system of grouping, as I have outlined it, is practi- 



iTest, Fluv. Bras., p. 33, 1827. 



^Lea believed that a natural classificatioa would be fouudetl on the development 

 of the embryos in the internal or external branchiae. 

 sProc. U. S. Nat. Mus., XV, 1892, p. 405, and Amer. Nat., XXVII, No. 31G, p. 353. 

 *Miss. Sci. aux Mex. dans I'Am. Cent., 7th part, II, pp. 517, 555, 1894. 



