584 THE AMEBIC AN NATURALIST [Vol. XLIV 



incident should have been closed at this point, but uncer- 

 tainty as to the degree in which the element of time took 

 part in influencing the place-modes of variability led to 

 the examination during the following summer of several 

 thousand more shells. Four convenient localities near 

 Woods Hole (see Fig. 3) where Urosalpinx was abundant 

 were selected and three lots of 1,000 each were collected 

 from each of these localities at intervals of two weeks 

 apart. The data obtained from these shells is arranged 



■ .(K5II7 

 .0:527 



-.0:547 



From Table V it will be seen that in each of the four 

 localities the arithmetical mean (A.M.) increased stead- 

 ily, except for the Penzance Point— July 21— lot of shells. 

 This general increase may be due to the fact that, as the 

 season advanced, there were fewer young shells in any 

 1,000 lot. The young are produced in May and June 

 from individuals that have wintered over, so that early 

 m July the Urosalpinx community is made up of old 

 adults from the preceding year and of young of various 

 sizes. A month later the population is more uniform, 



