RUGGERONE and ROGERS: ARCTIC CHAR PREDATION ON SOCKEYE SALMON SMOLTS 



200 



40 60 



PREY DENSITY 



100 



40 60 



PREY DENSITY 



100 



FIGURE 2. — Shapes of different types of hypothetical functional 

 (left ordinate i and numerical responses (right ordinate) (A) and 

 percent mortality curves (B» (redrawn from Hollmg 1959; Peter- 

 man and Gatto 1978). 



lures from 1976 to 1979. In 1980, the primary 

 method of char capture was a variable mesh size 

 (5.1, 6.4, 7.6, and 10.2 cm stretch measure), mono- 

 filament gill net set across the river and allowed to 

 drift downstream for 15 min. Nonparametic statis- 

 tical analysis of 99 char caught by hook and line or 

 gill net on the same day indicated no significant 

 difference in char consumption of smolts esti- 

 mated by each method (Mann-Whitney U-test: 

 0.10 < P < 0.20). The collected char were anes- 

 thetized with tricaine methane sulfonate (MS- 

 222 ), then their stomach contents were flushed out 

 by a stomach pump. Examination of stomach con- 

 tents from sacrificed char showed that about 909c 

 of the smolts were removed by the pump. Before 

 returning the char to the river, we measured the 

 fork length and placed a numbered Dennison 

 flag-type tag just below the dorsal fin. Stomach 

 fullness was estimated visually and categorized as 

 either full or less full. 



Consumed Smolt Analysis 

 Smolts consumed by char were counted and 



measured after standardizing their length in 10% 

 Formalin 3 for at least 24 h (Burgner 1962). The 

 consumed smolts were measured by one of three 

 methods and converted to fork length using one of 

 the following regression equations (Ruggerone 

 1981): 



1) Fork length (mm) = 0.44 + 1.09 (standard 

 length); r 2 = 0.99; 



2) Fork length (mm) = 3.70 + 1.37 (pectoral fin 

 to hypural bone plate); r 2 = 0.98. 



The preferred method of measurement was fork 

 length; the next preferred method was standard 

 length. When neither of these methods was 

 adequate, the length from the pectoral fin inser- 

 tion to the hypural bone plate was measured. Pre- 

 served fork length measurements were multiplied 

 by a factor of 1.042 to convert back to "live" fork 

 lengths (Rogers 1964). 



Collection and Enumeration of 

 Migrating Smolts 



Migrating smolts were collected and enumer- 

 ated with a winged-fyke net placed in an area of 

 intermediate, but substantial water flow. Smolts 

 trapped in the "live box" were counted and set free 

 every 4 h during the day (0800-2200 h) and con- 

 tinuously during the major migration period 

 (2200-0200 h). Daily smolt abundance was esti- 

 mated by multiplying the fyke net counts by a 

 river width factor. Previous experimentation with 

 two fyke nets indicated an even distribution of 

 smolts across the river. At least one sample con- 

 taining 30 or more smolts was collected each night 

 for length measurements, and when a substantial 

 number of smolts migrated during the day, an ad- 

 ditional sample was collected. Samples to deter- 

 mine a length-weight relationship were collected 

 about every 10 d. Fork lengths were measured to 

 the nearest mm and weights to the nearest 0.01 g. 



Environmental Data Collection 



The water temperature of Little Togiak River 

 was measured to the nearest 0.1°C several times 

 each day. To account for smolt density in the water 

 column, we measured the water level of Lake 

 Nerka as an approximation of the relative water 

 depth in Little Togiak River. The water level was 



3 Reference to trade names does not imply endorsement by the 

 National Marine Fisheries Service, NOAA. 



403 



