cutive seine hauls made along a 600-m. stretch of 

 beach in South Bay on August 4, 1962 (table 3), 

 shows that catches are variable and suggests that 

 pygmy whitefish are frequently grouped in large 

 schools. Approximately 80 percent of the pygmy 

 whitefish collected throughout the Naknek system 

 came from specific sampling efforts that yielded 

 100 or more specimens. 



Table 3. — Age 0+ and older pygmy whitefish caught in 11 

 consecutive seine hauls in South Bay, August 4, 1962 



Underwater observations on the schooling and 

 feeding behavior in lower Brooks River indicated 

 that pygmy whitefish are frequently grouped in 

 schools of several thousand fish. Fishes in these 

 schools were evenly spread out over several meters 

 of stream bottom. Individuals fed independently 

 of other fish, and the undisturbed school of fish 

 would slowly move from one area to another. 

 When disturbed by an observer or predator, schools 

 of pygmy whitefish in Brooks River became tightly 

 grouped. 



The distribution of age 0+ pygmy whitefish, 

 particularly in Brooks Lake, changes in late sum- 

 mer. > Many of these young whitefish remain in 

 littoral areas on the sandy shelf along the north- 

 east shoreline of Brooks Lake throughout much 

 of the summer, but by mid-August they begin dis- 

 appearing. They are scarce in this area in Sep- 

 tember. During the same period, limnetic catches 

 of pygmy whitefish in tow nets increase. Age + 

 sockeye salmon in Wood River Lakes behave simi- 

 larly (Burgner, 1960). Naknek system age + 

 sockeye salmon also show this behavior, although 

 it is not as pronounced. The reasons for this fall 

 change in distribution of young pygmy whitefish 

 are not known. The autumn lake turnover, 

 changes in diet attendant with shifts in food sup- 

 plies, or other limnological changes may be in- 



volved. During the fall, weather on the Alaska 

 Peninsula is characterized by severe storms with 

 gale winds. The resultant heavy wave action may 

 tend to disperse young pygmy whitefish from 

 littoral areas. 



About 95 percent of the pygmy whitefish caught 

 with tow nets in all lakes and basins were age + 

 fish. In 1961 tow nets were used in Brooks Lake 

 from mid-August throughout September, while 

 otter trawls were used primarily from mid-July 

 to mid- August. A comparison of length frequen- 

 cies of pygmy whitefish caught in limnetic areas 

 with tow nets with those caught in benthic areas 

 with trawls during these periods indicates (fig. 3) 



t 70 



I AGE 1+ 

 AGE0+ I AND OLDER 



I 



3 40l 50 60 



FORK LENGTH (MM.) 



Figure 3. — Comparison of length frequencies of pygmy 

 whitefish caught from mid-July to mid-August in 

 benthic areas with otter trawls with those caught from 

 mid-August through September in limnetic areas with 

 3.1-m. -diameter tow nets, Brooks Lake, 1961. 



that while both age 0+ and older fish were caught 

 in the benthic zone, most of those caught in the 

 limnetic zone were age + . Differences in the 

 selectivity of the two gears could have affected 

 these catches; however, we do not feel trawls were 

 selectively collecting older, larger pygmy white- 

 fish since comparative catches of other fishes sug- 

 gest that larger specimens are usually caught in 

 tow nets. Our interpretation of these data is that 

 fewer age 1+ and older pygmy whitefish were 

 present in limnetic areas than in benthic areas. 

 Differences in length frequencies of age 0+ fish 

 shown in figure 3 are due to sampling periods and 

 not to differences in size of fish caught in separate 

 ecological areas or with different sampling gear. 

 In 1962 and 1963 when tow nets were used in 

 Brooks Lake throughout July and August, no 

 pygmy whitefish were caught in limnetic areas 



562 



U.S. FISH AND WILDLIFE SERVICE 



