40 PROCEEDINGS OF THE ACADEMY OF [Jail., 



In this connection it is to be noted that the axial index shows simply 

 the proportion between the height and the width. It does not neces- 

 sarily show that one race is more conical than the other. It is to be 

 regretted that no satisfactory method for definitely determining this 

 fact was found. In the absence of specific proof, it is only possible 

 to record the writer's impression, derived from a careful study of the 

 material, that as a rule the Lower Michigan shells are not only pro- 

 portionately more elevated, but are actually more conical in shape. 



Conclusions. 



From a careful study of the foregoing data, the following conclu- 

 sions seem to be justified: 



1. That Polygyra albolabris in Michigan exhibits great variability 

 in size. 



2. That both the Upper and Lower Peninsular series show a well- 

 marked division into two parts, the dividing line being at 24.5 mm. 

 in width. 



3. That this dividing line corresponds quite exactly with that 

 separating the several described minor varieties and the typical form 

 as exemplified by the Cincinnati series. 



4. That the evidence tends to show that when depauperization 

 takes place in P. albolabris, whether in individuals or in local races, 

 the results are within certain fairly definite lines. 



5. That when depauperization affects substantially the whole race 

 in a particular locality or district, it is sufficiently permanent to be 

 worthy of varietal recognition. 



6. That the depauperate shells of the Lower Peninsula are on an 

 average smaller and more depressed than those of the Upper Peninsula. 



7. That the major (or typical) series of the Upper Peninsula is 

 smaller and more depressed than that from the Lower Peninsula. 



8. That the major series of the Lower Peninsula is larger and more 

 elevated than the Cincinnati series. 



9. That this is probably to be accounted for by the fact that the 

 Michigan series is not an unprejudiced one and contains a large pro- 

 portion of selected specimens. 



10. That the average shell of both Michigan series combined is 

 substantially the same as that of the Cincinnati series. 



11. That the great variability of the Michigan series, taken as a 

 whole, as compared with the Cincinnati series, is due mainly to the 

 greater diversity of environmental conditions. 



