282 THE AMERICAN MIDLAND NATURAIJST 



it might be fairly assumed that any remaining coloration, making 

 allowance for the possible effects of stream pollution was character- 

 istic for the shell at the locality. Finally, the natural conditions 

 of the Upper Ohio are also largely true of the streams draining 

 into L. Erie, and similar explanations may be advanced for the 

 shells living them. 



2. The pollution of streams by sewage and industrial wastes 

 presents a most unnatural factor affecting the epifiermal color of 

 shells.* While as Ortmann and Baker have independently pointed 

 out, pollution from either source may be so extensive as to ulti- 

 mately kill the animals, for the purposes of our problem we may 

 only consider their possible relations to epidermal color. Sewage 

 is largely organic matter and would seem first hand to be most 

 largely concerned with the amount of silt in the stream. According 

 to Prof. Karl Phelps of the U. S. Public Health Service, the in- 

 dustrial wastes are largely sulfuric acid and sulfate of iron. Where 

 the former chemical is present in sufficient abundance it would 

 burn the organic matter, (conchiolin) of the shell black and thus 

 be partly responsible for darker colors, while the sulfate of iron 

 might form discoloring deposits. As it happens that the pollution 

 of the water by these wastes increases going down stream, un- 

 doubtedly some of the change of color indicated is due to it, at 

 least in the lower stretches of the Allegheny and the Monongahela 

 as conditions now stand. This deposit of iron is frequently so 

 tenacious as to require acid to dissolve it. 



The fact that each drainage leaves its own imprint on the shells 

 collected from it is well known to experienced collectors. In view 

 of the data previously presented with regard to the great uniformity 

 in epidermal color determinable at a given locality, such seems 

 readily referable to peculiar stages in the development of the 

 environmental conditions outlined. Similarly, causes underlying 

 conclusion 5 may be sought for. The rays of the epidermis may 

 disappear with age on account of the darkening of the epidermis 

 due to the causes suggested. The rays have their widest distribution 

 in small rivers and creeks, .where of course, silt is not in its greatest 

 variety and abundance. The conclusion as stated that age has no 

 relation to a regular sequence of epidermal color change somewhat 

 bears out the opinion ventured concerning the greater effect of the 



* The localities from which my material was collected gave evidence of 

 pollution at the time, and a large number of them are now completely barren. 



