10 U. S. BUEEAU OF FISHERIES. 



completion and in part to the study of conditions causing losses of 

 young fish or of brood stock in fish hatcheries. A flagellate protozoan 

 {Octomitus scdmonis) occurring in the forward end of the intestines,' 

 sometimes in enormous numbers, was deemed responsible for the 

 high death rate among rainbow trout being shipped from the White 

 Sulphur Springs (W. Va.) station. 



In response to a request of Carlos Avery, commissioner of the 

 Minnesota Game and Fish Department, an investigation was made 

 of the cause of the heavy losses of fish during the summer montlis of 

 1921 in several Minnesota lakes. The loss of fish in Lakes Bemidji, 

 Leech, and Mille Lac was very great. Examination of dead and 

 dying fish revealed no bacterial, fimgous, or other parasitic infecting 

 organism. All evidence pointed to the conclusion that the mortality 

 resulted from lowered dissolved oxygen content, occasioned by the 

 want of natural mechanical means for the introduction of oxygen; 

 the great losses occurred during and just after an unusually ex- 

 tended period of great heat, with no wind or rain. The species of 

 fish mainly affected were those that normally seek the lower and 

 cooler and in this case, presumably, the almost oxygen- free waters 

 of the lakes in question. The recommendations offered through B. J. 

 Anson, who investigated the condition, had reference to an appro- 

 priate control of the fish population through the promotion of 

 restricted commercial fishing in these lakes. 



Prof. A. S. Pearse, temporary investigator, completed during the 

 year a report of observations of fish parasites in several waters. The 

 study was originally undertaken as one phase of an attempt to dis- 

 cover why fishes attain only a small size in certain bodies of water, 

 although becoming much larger in other waters near by. The ob- 

 servations were at first restricted to perch from 16 lakes in 3 dif- 

 ferent river systems. Later more extensive observations were made 

 in 5 different types of lakes where the parasites of all available 

 species of fishes were studied. About a dozen species of parasites 

 were discovered on the yellow perch, and approximately 95 species 

 were found on fish of all species examined in the 5 lakes. 



Lakes with the widest range of territory and with opportunity 

 for fishes to invade the greatest variety of habitats are found to 

 have the highest average infection per fish. Taking as a criterion 

 the number of species of parasites showing highest average infec- 

 tions, the lakes rank in the following order : Pepin, Mendota, Michi- 

 gan, Green, Geneva. This order indicates that there is a direct rela- 

 tion betw^een variety of habitat and amount of infection. Variety 

 of habitat is also correlated with a large number of species of para- 

 sites (and of fishes). In other words, the lake with the largest 

 variety of habitats has the greatest variety of fishes and of parasites. 

 It is stated as a corresponding principle that the fish that travel most 

 and thus invade the greatest variety of habitats have the most para- 

 sites. 



Arranged in order of average infection, and weighing the fish and 

 lakes according to a formula given in the paper, the fishes in the 

 five lakes rank in the following order: Dogfish, 90; small-mouth 

 black bass, 78; white bass, 78; rock bass, 73; pumpkinseed, 71; 

 sucker, (58; black bullhead, 66; tadpole cat, 64; wall-eyed pike, 61; 

 speckled bullhead, 60 ; long-billed gar, 56 ; bluegill, 55 ; yellow bull- 

 head, 55 ; all species of cisco, 54 ; large-mouth black bass, 45 ; pickerel, 



