296 I ESHES OF ILLINOIS 



Havana and Meredosia. It is rather disproportionately infre- 

 quent in the waters of the lower Illinoisan glaciation, although 

 not by any means excluded from that area, as a glance at the 

 distribution map for the species will show. We have found it 

 most abundant in the small streams of the Wabash and Kas- 

 kaskia systems, in which it has occurred in 56 and 66 per cent., 

 respectively, of all collections made. 



It is typically a darter of the creeks and small brooks, and 

 44 per cent, of all our creek collections have contained it. It 

 has come from the smaller rivers with about half this frequency, 

 and from glacial lakes with about a fourth. The - average 

 character here ascribed to it is illustrated by the fact that it 

 has been taken by us with darters of other species in almost 

 exactly the average frequency of the associate occurrence of one 

 species with another throughout the whole subfamily. 



It is usually found among gravel and weeds, although not 

 infrequently on a mud bottom, from which situation some 11 

 per cent, of our collections came. Its preference for swift waters 

 is not so marked as in the case of the more typical 4 ai 'ters, 

 nearly a third of our collections having come from standing or 

 quiet water. 



Outside Illinois the species is found from New England and 

 Lake Champlain through the Great Lake region to the Assini- 

 boin River, down the Atlantic slope as far as the Catawba 

 River, and westward throughout the Ohio and Missouri basins 

 to Colorado and Montana. 



Its habits are those of its subfamil}-. It often lies with its 

 head up and its body bent to one side or supported partly by a 

 stone. It can turn its head without moving its body; can roll 

 the eye about in the socket; may rest suspended, as we have 

 seen it do, on the under side of a floating board; and sometimes 

 buries itself, with a whirl, in the soft sand, so that only its eyes 

 are visible. 



The food of a dozen specimens was so uniform that they 

 may fairly be taken as representative. Two thirds of it con- 

 sisted solely of Chironomus larvae, 7 per cent, of other minute 

 larvae of gnats, and the remaining 12 per cent, of larvae of small 

 May-flies. 



The species spawns 'in spring, from the last of April to the 

 first of June. Females were depositing their eggs in our aqua- 

 rium at Meredosia, April 28 and 29, 1899. In the act of spawning 

 th'e male rode on the back of the female, with ventrals astride, 



