56 MEMOIRS OF THE CARNEGIE MUSEUM. 



Distribution and Ecology in Pennsylvania (See fig. 6) : This species has been 

 found both in the larger rivers and in some of the smaller creeks, but according to 

 my experience, it is distinctly more frequent in the latter. Yet it does not go up 

 into the extreme headwaters. It is not rare in the Beaver-drainage, and also in 

 French Creek, but entirely absent in the upper Allegheny. At all other localities, 

 it is distinctly a rare shell, and the same is true for the Ohio below our state. 



Whenever I found this species alive, it was in swiftly running water, upon bars 

 of gravel, very often in riffles with an abundant growth of Dianthera americana. 

 In fact, on the edge of such patches, I was most successful in taking this species 

 alive. It is mostly not deeply buried, often simply lying upon the bottom. In 

 the Ohio, it is upon the shell-banks, which become accessible only at the lowest 

 stage of the river. This agrees with Scammon (190G), who observed that the 

 "favorite habitat is bars of gravel or shingle in rather swift current." 



General distribution: Type locality, Wabash River (Say). 



Simpson (1900) gives for this species: "Entire Ohio, Cumberland, and Tennes- 

 see river systems; west to Nebraska (doubtful); south to Arkansas and Indian 

 Territory." The first three rivers undoubtedly are the metropolis of this species, 

 and here it is found from western Pennsylvania'*^ through Ohio and West Virginia 

 to Tennessee and northern Alabama (Tennessee-drainage). It is also frequent in 

 the part of Indiana drained by the Ohio, White, Wabash,^*^ and Kankakee Rivers 

 (Call, 1896a and 1900), but in Illinois it is found only in the southern part (Ohio 

 and Wabash, Baker, 1906). Thence, in a northerly and northwesterly direction, 

 it is not found any more (this fact should be noted) , but it is present farther South 

 in the Mississippi-drainage, in southern Missouri (Utterback, 1916), in southeastern 

 Kansas, eastern Oklahoma, and in Arkansas. Call (1895, p. 15) reports this species 

 from the Alabama River, Selma, Dallas Co., Alabama. This is the only locality 

 outside the Mississippi area, and seems to be erroneous, for it has never been found 

 again in this system. 



Genus Rotund aria Rafinesque (1820). 

 Ortmann, 1912, p. 257; Simpson, 1914, p. 903 (as subgenus of Quadnda). 



Type Obliquaria tuberculata Rafinesque. 



Two species have been assigned to this genus, which may be only varieties of 

 the same species. Only one of them is found in Pennsylvania. 



•"^ Call (1895) quotes western New York, but this record has never been substantiated. It is 

 missing in Marshall's list (1895). 



"According to Goodrich (1914), Q. cijlindrica recently has crossed over into the Maumee River, 

 and descended as far as Antwerp, Paulding Co., Ohio. He calls it Ij}' the varietal name sirigillaia, but 

 probably this is an error. 



