MOLLUSCA. 141 



constant supply of water, have now been cleared off. This probably 

 accounts for the fact that the bottom of this portion of the stream was 

 strewn with the dead shells of Goniohasis and a few of the larger uniones, 

 which are not now found alive in this part. 



On a plank floating along the marshy shore of this part of the river, 

 there were found a few specimens of Sphoerium solidnlum, Segmentina 

 crassilabris, Succinea avara, Lymnaea palustris michiganensis, and L. 

 humilis. Among algae, in situations similar to those in the river, 

 Amnicola cincinnatiensis, Planorhis trivolvis, Physa integra, Agriolimax 

 agrestis, Succinea retusa, Campeloma integrum, Sphaerium solidulum, 

 S. striatinum, and a small form of Lymnaea obri/ssa occurred in larger 

 numbers. In addition, along the bottom were obtained specimens of 

 Planorhis trivolvis, a form of Planorhis campanulatus similar to that 

 found in the bayou, Sphaerium solidulum, and Lymnaea palustris, and 

 some dead shells of Goniohasis livescens and Quadrula undulata. 



24. Dead waters. About five miles up stream from its mouth, the 

 river widens and deepens, so that there is formed a long "pickerel 

 pond" or dead water, a stagnant pool about ten yards wide by one 

 hundred yards long. This pool is bordered by swampy, deciduous 

 w^oods, and there is a marginal zone of button-ball bushes {Cephalanthus 

 occidentalis) growing out over the water. The edges of the pool are 

 covered with pickerel weed (Pontederia cordata), while outside of these 

 the yellow water-lilies form cjuite a definite zone. The water is black 

 and slimy, and rotting logs almost block up the stream. (PI. XI b.) 



In this habitat, on the under-side of lily-pads, a few specimens of 

 Ancylus parallelus, Lymnaea palustris. Planorhis parvus, Physa gyrina 

 and P. integra, and single individuals of P. elliptica, Planorhis trivolvis, 

 and Succinea retusa were collected. 



This dead w^ater is so deep and wide (in comparison with the small 

 stream w^hich flows into and out of it) that there is little current even 

 during flood seasons, and the quiet water causes most of the sediment 

 carried in by the stream to be deposited. In addition, the broad zones 

 of pickerel w^eeds and water-lilies along the shores must deposit con- 

 siderable organic material, so that it appears that it will be simply a 

 matter of time before this body of water is reduced to the size of the 

 creek. 



La7id Hahitats. 



Hahitats of the Satid Dunes. — 25. Sand heach. Along considerable 

 stretches of the lower and middle beaches of protected shores, the waves 

 have washed up masses of decaying vegetation. These are kept moist 

 by the capillary action of the sand; the masses themselves are often in 

 direct contact with, and even floating on, the water. Among these de- 



