BioLorarAL Purvey — ERiE-\iAr,'ARA Watershed 225 



words, mayflies were the most available food and they were con- 

 sumed by trout more than any other food. C'addisflies were second 

 in importance and formed 2(j.:)7 per cent of foods consumed and 

 only 4.1'> per cent of available drift foods. Thus from these figures 

 it is evident that the trout fe<'d on larval caddisflies on stream 

 bottoms since so few were available in the drift. Two-winged files 

 formed 15.39 per cent of the troM djet^and 28.14 per cent of 

 available foods which indicates that many more were available in 

 the drift than were being consumed by the fish. Similarly, plant 

 lice formed only 5.13 per cent of consumed foods and 18.26 per 

 cent of those available. Plant lice and flies w^re much more abund- 

 ant in the drift than in the trout. The reason for this would seem 

 to lie in the fact that most of the flies and lice taken in the drift 

 were quite small and would be hard for trout to s(h\ It has been 

 found by other workers that the smaller insects are most largely 

 consumed by the smaller fishes, 1-3 inches in length, ^lany of the 

 flies which were taken from the stomachs were of good size which 

 the trout could have readih' seen before eating. Beetles and ants, 

 bees and wasps formed 8.15 per cent and 7.33 per cent respectively 

 of consumed foods and 5.71 per cent and 5.26 per cent of available 

 foods which shows a close correlation between the availability and 

 consumption of these two foods. Likewise the stoneflies, true bugs 

 and spiders show a close correlation between availability and con- 

 sumption. Grasshoppers, moths, shrimps, crabs and worms were 

 more abundant in the stomachs than in the drift. This probably 

 is due to the fact that the trout were consuming these foods, and 

 perhaps other foods as well, befoi-e they had time to reach the 

 drift net. Only four fish were taken from the 147 stomachs 

 examined. If larger trout had been taken, doubtless many more 

 fishes would have been found, as it is well known that large trout 

 oftentimes feed voraciously upon minnows, trout fry and other 

 small fish. Metzelaar,* in work upon trout in ^Michigan, found 

 that fish formed 3.7 per cent of the stomach, contents b}^ volume 

 of fishes 7 to 16 inches long. In larger trout, 17 to 28 inches long 

 fish formed 23.8 per cent of the stomach contents by volume. Dr 

 ^letzelaar's results are not tiuly comparable to the results given 

 here, since his are expressed in per cent by volume and ours in 

 per cent by number. However, he does show that fish are con- 

 sumed in greater quantities l)y larger trout and that fish form but 

 a minor part of the diet of snudl trout. Shrimps and crabs were 

 low in number, foi*ming only 0.72 ]^er cent of consumed foods, and 

 likewise they were unavailable to the trout, forming only 0.2 per 

 cent of the drift food. However, in some streams, particularly 

 those in which the Avaters are high in calcium, shrimps oftentimes 

 will be very abundant and will be a principal part of the diet of 

 trout in such streams. 



Chart 1 (derived from Table 4) shows graphically the consump- 

 tion and availability of each food. By reference to Table 5, it is to 



o? 



* Metzelaar, Dr. Jan. " The food of rainbow trout in Michigan." Preliminary 

 report to the Department of Conservation; Lansing, Michigan. Jan. 1928. 



