67 



SPONOES AND IMtOTOZOA. 



One of tlic Irosli water sponges (Spongilla) had been ealen in consider- 

 able quantities by two examples of the spotted cat taken in »September, 

 but this element was not encountered elsewhere in my studies. 



That the minutest and simplest of all the animal forms, far too small 

 for tlie eye of a fish to see without a microscoi)e, siiould liave been recog- 

 nized in the food of seventeen species of fishes is, of course, to be ex- 

 plained only as an Incident of the feeding habit. It is possible, however, 

 tliat these Protozoa, where es])ecially abundant, may be recognized in the 

 mass by the delicate sensory structures of the fish: and they seem in most 

 cases to have been taken witli mud and slime ricli in orgjinic substances. 

 As most of them are extremely perishable, and can scai'cely leave a trace 

 a few seconds after immersion in the gastric juices of the fish, it is prob- 

 able that they contribute much more generally than our observations in- 

 dicate to the food of some fishes, es])ecially to those which feed upon the 

 bottom. 



Young suckers under six inches in length clearly take them purposely, 

 substituting them in great part for the Entrymostraca taken by other 

 fishes of their size and age. 



I detected Protozoa in the food of several genera of Cy]M-inidie, in the 

 young of buffalo, the river carp, the chub sucker, the red horse, the stone 

 roller', in the common sucker, in a single gizzard shad, in a stone cat, 

 and in a top minnow. The commonest forms, as would be supposed, were 

 those protected by permanent shells; A'iz., Dittlugia, Centropyxis, Arcella, 

 and the like: but occasionally specimens of Actinospha^rium, Euglena, and 

 Dinobryon were present and recognized. 



SCAVENGERS. 



The only scavenger fishes of our collection were three species of the 

 common cattishes: the spotted cat, the yellow cat, and the marbled cat, — 

 all of which had eaten dead animal matter, including pieces of fish, ham, 

 mice, kittens, and the like. A single large-mouthed black bass had like- 

 wise eaten food of this description. 



VEC4ETABLE FEEDERS. 



Considering the wealth of Tegetation accessible to aquatic animals, and 

 Mie fact that few other strictly aquatic kinds have the vegetarian habit, 

 it is indeed remarkable that the plant food of fishes is an unimportant 

 part of their diet. Taking our nine hundred specimens together, the 

 vegetation eaten by them certainly would have amounted to less than ten 

 per cent, of their entire food, and excluding vegetable objects apparently 

 taken by chance, it ])robably would not reach five per cent. 



The greatest vegetarians are among the minnow family, largely in the 

 genera Ilybopsis, >S^otemigonus, and Semotilus, thirteen si)ecimens of the 

 first and twenty-live of the second having taken about half their food 

 from vegetable objects. One hundred and twelve Xotropis. twenty-two 

 Semotilus. eighteen Ilybognatlius. and nine tampostonia. had found in 

 the vegetable l^ingdom a fourtli or lifth of their food. Counting eacli genus 

 as a unit, I find that the family as a whole obtained from plants about 

 twenty-three per cent, of its food. The little riienacobius. already re- 

 ported as strictly insectivorous, was the only onestudicil in wliicli vegeta- 

 tion can scarcely be said to occur. 



The nnid minnows (Umbrida') are also largely vegetarian (forty-one per 

 cent.): and likewise the cyprinodonts, the vegefal>le average in tlie food of 

 thirty-three si)ecimens being seventeen ])er cent. T'lant structures made 

 about one-fourth the food of seven sticklebacks. 



' Hypentelium. 



