26 ILLINOIS BIOLOGICAL MONOGRAPHS {124 
of Naiades to varieties on the basis of compression in the headwaters of 
streams. The same species varies in obesity, but it is the same species, 
whether thin or fat. Other characters are usually present which separate 
the allied species. 
Tasie VI. Spectres or UNIONIDAE FounpD ABOVE AND BELOow Homer Park DAM 
Both Above and Below Dam Below Dam Only stes* 
parva anodontoides 
ferussacianus ellipsiformis 
grandis ligamentina 
imbecillis mulliradiata 
edentulus lachrymosa 
pavonius melanevra 
luteola wardit 
Nenosa R. tuberculata 
complanala T. tuberculata 
rubiginosa cwrculus 
undulata clava 
pustulosa glans 
costata 
marginata 
cocéineum 
ventricosa 
compressa 
In the Big Vermilion this variation in compression is marked in several 
species. Thus Rotundaria tuberculata, Pleurobema coccineum, and Amblema 
are more compressed than are individuals from the Wabash River below 
the junction of the Big Vermilion with that river. Quadrula pustulosa is 
smaller than the same species lower down in the Salt Fork, and the same 
may be said of Alasmidonta marginata and Strophitus edentulus. That the 
rule does not always hold good is shown by the variation of Fusconia rubi- 
ginosa which is abundant in most parts of the Big Vermilion and its tribu- 
taries. Measurements are given in Table VII, showing the length and 
breadth of several species in different parts of the Salt Fork from below 
Urbana to the Big Vermilion. The percentage of width to length is also 
shown. 
It will be seen that the average index for the first lot is 42 per cent and 
for the last lot, 46 miles down stream, is almost the same, 45 per cent. 
These averages compare well with some of those given by Ortmann (1920: 
283). It was observed, however, that in the Salt Fork and Big Vermilion 
the obese individuals occurred with the compressed specimens the former 
increasing in ratios as the distance down stream increased. The variety 
wardii of Quadrula metanevra occurred in two places in Salt Fork but always 
in company with the typical form. In the cases cited above ecological 
features cannot be called into account in locating the cause of the com- 
