30 PROCEEDINGS OF THE ACADEMY OF [Jan., 



aggregated at these points are from twelve different localities, scattered 

 all over the State and in both peninsulas, it is evident that this is a 

 feature of the Michigan race as a whole, and not the result of a few 

 sets of strongly characterized local forms. The same peculiarity- 

 appears in the Isle Royale series (fig. 12), and is intensified when 

 the two series are combined (fig. 5). It is similar to the strong minor 

 mode culminating at 30.25. Indeed, the diversion of the whole series 

 between 24.50 and 34.25 mm. into five well-marked groups is quite 

 striking. In the same way the minor series below 24.5 is divided into 

 five similar groups. 



It is to be noticed that the break in this series at 24.5 is sub- 

 stantially at the same place where the break occurs between the Cin- 

 cinnati and the "minor" series. This is apparently more than a mere 

 coincidence. It is certainly corroborative evidence tending to 

 strengthen the inference drawn from the comparison of those series. 



Of the 225 specimens in the Michigan series, 55 are 24.5 mm. in 

 diameter or less and 170 are more than 24.5. Of the smaller group 

 8 are from the Upper Peninsula and 47 from the Lower. Of the 

 larger series 34 are from the Upper and 136 from the Lower Peninsula. 

 That is, in the Upper Peninsula 19 per cent, belong to the smaller 

 race and 26 per cent, in the Lower Peninsula. 



In fig. 1 the number of specimens of 24.5 mm. or less in diameter 

 from each county are represented by the smaller figure. Thus in 

 Kent County, of the 30 specimens in the series, 9 are of the smaller 

 race. 



In considering the probability of the existence of a small race, it is 

 necessary to distinguish between dwarf individuals, which occur 

 occasionally in all series of any extent, and a race of small individuals, 

 which are all more or less characterized by their diminutive size, It 

 is only the latter, of course, that is entitled to recognition in any 

 proper varietal sense. 



From an inspection of fig. 1, it will appear that, while the small 

 individuals are fairly well scattered over the State and in some cases 

 are no doubt individual dwarfs, they are nevertheless more numerous 

 in the northern part of the State, where the environmental conditions 

 are generally more unfavorable to molluscan life and likely to affect 

 the species as a whole. And it is significant also that of the entire 

 series of 55, no less than 37 occur in the four counties of Charlevoix, 

 Grand Traverse, Huron and Kent. The series from Kent is peculiar 

 in being from so far south and in a county where the typical form is 

 also well represented. The series, however, is well marked and proba- 



