MUSSELS OF CUMBERLAND EIVER AND TRIBUTARIES. 63 



the ventral margm, leaving thin tips. The shell is easily distinguished from any- 

 other species in the river except from old examples of Q. suhrotunda, which are always 

 more elongate and always have yellow flesh. 



Q. ebena would probably thrive only in the lower parts of the river, although when 

 the propagation of this species becomes feasible it may be worth trying in the upper 

 river. 



78. Quadrula tuherculata Rafinesque. Purple warty-back. 



A careful study of our material, as well as of the evidence at hand from the literature, 

 convinces us that Q. granifera and Q. tuherculata, though quite markedly distinct in 

 typical cases, are really connected by intermediate forms. In some rivers, like the 

 Tippecanoe at Delong, Ind., only strongly marked tuherculata are found. In others, 

 like the Mississippi about Fairport, Iowa,, only well-marked granifera are found. 

 In such streams or portions of streams as contain both species they are indistinguish- 

 able, or so connected by intergrades that no clear liae of demarcation can be drawn 

 between them. In the Cumberland, the first shells seen, in the lower part of the 

 river, were identified provisionally as granifera; as we ascended the river some doubts 

 as to the species began to appear, while in the upper tributaries the shells were pretty 

 clearly identified as tuherculata. This naturally introduces the question as to influence 

 of envu'onment on shell form, which may be touched upon briefly here. 



The most striking and essential difference between tuherculata and granifera is one 

 of degree of inflation, tuherculata being a flat form and granifera much inflated. We 

 have a number of cases among the Unionidae where two otherwise similar sliells are 

 distinguished by this feature; among these are: Q. plicata, inflated, Q. undulata, 

 compressed; D. dromas, inflated, D. caperatus, compressed. From our experience 

 we are inclined to believe that one usually finds the compressed species in small 

 streams, while the more inflated forms are found in large rivers. Often Avhen a main 

 stream has plicata, the little tributaries \\\\\ have undulata, especially if they are 

 rather shallow and swift streams with gravel bottoms. The more compressed form is 

 better adapted to plow into the gravel or crawl under rocks and hold its position in a 

 swift current, where the inflated form would present too much surface to the force of 

 the water. In the softer mud and weaker current of larger streams an inflated form 

 would be advantageous, heli^mg to buoy up the animal. 



To state the situation precisely as we have found it, if one takes one of the larger 

 rivers from source to mouth, and finds both tuherculata and granifera or plicata and 

 undulata in the stream, the compressed form is likely to be in the upper stretches of 

 the river while it is a small swift stream, and the more inflated form farther down in 

 the main body of the river where the bottom contains more mud and the current is 

 slower. Extreme forms of either species, so far as we know, are never found in the 

 same bed, but where both are represented the forms run together. 



The literature relating to granifera and tuherculata is exceedingly interesting, but 

 too long to give in detail. To imderstand the present status of the group, however, it 

 is necessary to state that Simpson in his Synopsis removed these two species from the 

 Quadrula pustulosa group, where they had been previously placed, making of them 

 the subgenus Rotundaria on the basis of a "well -developed sulcus on the posterior 

 5lope and remarkable beak sculpture." The beak sculpture is well marked on tuher- 

 culata but not so well, or almost absent, on granifera. Ortmann, finding only the outer 

 gills used as marsupia in tuherculata, raised Rotundaria to generic rank. We have 

 usually found only the outer gills of granifera at Fairport marsupial, although we have 

 a record of one example with marsupia in all four gills. 



The species does not reach a very large size in the Cumberland. On account of 

 its purple nacre it is of no value for buttons. 



