698 The Food of the Darters. [ October, 
same date and locality usually agreed so closely in food, that the 
study of from two to five gave all the facts obtainable from sev- 
eral times as many. The data here given, therefore, really exhibit 
the food of the family at different seasons in twenty-nine localities 
within the State. 
The genus Plewrolepis is comparatively rare in Illinois, as there 
are few of the sandy streams in the State, which it inhabits. 
Seven individuals were examined—four of P. pellucidus and three 
of P. asprellus. The food of these specimens was remarkably uni- 
form—the only elements found being the larvæ of small diptera 
and Ephemerids. Eighty-one per cent. of the food of all con- 
sisted of the larva of Chironomus,'—a small, gnat-like insect,— 
twelve per cent. of the larva of other small diptera, and the 
remaining seven per cent. of Ephemerid larve (May flies). 
Twelve specimens of the genus A/vordins were studied—seven 
of maculatus and five of phoxocephalus? These represented five 
different localities and dates. This is a larger species than the 
preceding, and to this fact is probably due the predominance 
(seventy-five per cent.) in its food of the larvae and pups of May 
flies (Ephemeridz). These included four per cent. of the larve 
of Palingenia bilineata Say, one of the largest Ephemerids in our 
streams. The remaining kinds were larvz of dragon flies (Agri- 
onini), four per cent.; larve of Chironomus, seven per cent., and . 
Corixa tumida Uhl., thirteen per cent. 
The genus Boleosoma, regarded by Dr. Jordan as the typical — 
darter, was represented by twelve specimens from eight locali- 
ties—nine of maculatum, two of olmstedi and one of camurum? 
These specimens show but slight food differences from other 
darters of similar size; the only notable variation being the 
appearance of fifteen per cent. of case-worms (larvæ of Phry- 
1 The larvæ of Chironomus are among the most important elements of fish food in 
our waters, appearing in abundance in the stomachs of the young of a great variety 
of species. They have been too little studied in this country to allow specific deter- 
| 2 The classification used in this paper is that of the second edition of Jordan’s 
Manual of Vertebrates. ; 
3 Boleosoma maculatum and B. olmstedi should undoubtedly be united. Spec! 
mens in the laboratory collection present the extremes of both. forms, together wiih 
numerous intermediate stages of each character used to distinguish them. 
This whole group exhibits a surprising variability, perhaps due to its compara- 
_ tively recent origin. 
