MICHIGAN ACADEMY OF SCIENCE. 21 



fifty meters long and twenty centimeters deep; they were too small to be 

 permanent as their depth was less than the annual variation in the level 

 of the lake. In both cases, on account of the method of formation, they 

 were usually long, narrow and shallow, with their long axes parallel to the 

 shore, and at right angles to the wave action. 



II. Sand Spit Tijpes. The sand spit or hook, on the other hand, was 

 usually built up more or less parallel to the direction of the currents, which 

 in our small, fresh-water lakes is largely dependent on the direction of the 

 prevailing strongest winds. They were produced by sand being carried 

 out around some point that crossed the path of the current, and then de- 

 posited in deeper water; this process was carried on until the spit again con- 

 nected with the shore and produced a bar. Their size, and the size of the 

 pools and lakes behind them, was not so completely dependent on the size 

 and strength of the waves and currents, and consequently on the size of the 

 lake in which they were produced, as were the barrier beach pools; they 

 could be started in shallow water along the shore and then be built out 

 gradually into deeper water. The time required for their building, how- 

 ever, was directly so dependent. As a result, the sand bar pools were much 

 larger features than the barrier beach pools, it is true that they were much 

 larger in the Great Lakes than in the smaller, inland ones, but this was 

 largely due to the fact that the few thousand years that have elapsed since 

 the glacial retreat from Michigan was far to short a time for the puny cur- 

 rents of these smaller lakes to construct such large bars, as have been pro- 

 duced, for instance, along Lake Michigan, where many of the lakes so sep- 

 arated are larger than Douglas Lake itself. Off several of the points in 

 Douglas Lake were found very evident signs of large bars, which will, in 

 time, separate off lakes of considerable size from the parent body. 



MOLLUSCAN SUCCESSION. 



A. MOLLUSCS OF DOUGLAS LAKE. 



The molluscs of Douglas Lake, at least of the southwestern portion where 

 the pools were being formed, entirely consisted of gill-breathing species, or 

 of pulmonates which had become adapted to take water into their lungs 

 and breathe in that way. Even the young of Physa ancillaria parkcri 

 (Currier), which lived in water but a few centimeters deep, normally breathed 

 in this way, while the older shells, such as the adults of this same species, 

 were so completely habituated to breathing water that, when placed in 

 small aquaria, they died soon after exhausting the air in the water, with- 

 out even attempting to come to the surface to breathe; altho when once 

 taught to breathe air, by the simple method of exposing them out of water 

 until the water in their lungs partially evaporated or was otherwise replaced 

 by air, they could be kept in a small dish for several months and would 

 come to the surface regularly like any ordinary, air-breathing form. 



The shallow littoral fauna, which was the only portion that directly af- 

 fected the beach pools, could be divided into five groups according to habitat: 

 (1) the shells of unprotected, sandy shores; (2) the shells of more or less 

 protected, sandy to marly shores; (3) the shells of well-protected, mucky 

 shores; (4) the deep littoral shells that came up into shallow water to breed; 

 and (5) the shells of the submerged vegetation zones. 



