(Colby and Nepszy 1981) for each area are included 

 in Figure 4 as a measure of solar energy input to 

 the system. Colby and Nepszy (1981) found that 

 walleye growth was directly correlated to ODD >5°C 

 and that the optimum range was from 2,500 to 4,000 

 GDD >5°C. While the GDD >5°C for the John Day 

 pool is within this range, the walleye growth reported 

 here is greater than would be predicted using this 

 variable 



Water temperature may be the most important 

 factor governing the growth of fishes (Brett 1979). 

 Kitchell et al. (1977b) presented a bioenergetics 

 model for walleye growth and indicated that ther- 

 mal optima and maxima for weight specific con- 

 sumption are 22°C and 27°C, respectively and 27°C 

 and 32 °C, respectively, for weight specific respira- 

 tion. Water temperatures in the John Day pool 

 during the growing season remain at or near the 

 thermal optimum for consumption and, perhaps 

 more importantly, do not approach the thermal max- 

 ima for consumption or respiration (Tkble 1). Many 

 northern lakes may not reach the thermal optima 

 (Rawson 1957; Swenson 1977) and the southern lakes 

 or lakes which stratify in the summer may exceed 

 the thermal maxima (MacLean and Magnuson 1977) 

 not only reducing consumption but increasing 

 respiration. Dendy (1948) reported that in June 1944 

 the surface temperature of Norris Reservoir was 

 about 30 °C and that walleye appeared to prefer 

 water temperature of about 24 °C, even though these 

 areas had oxygen concentrations <3.0 mg/L. Con- 

 versely, water temperature of Lac la Ronge did not 

 exceed 20°C (Rawson 1957), well below the thermal 

 optima. 



Exceptions to the north-south trend in high wall- 

 eye growth occur in systems of high exploitation 

 (Forney 1965) and/or where there have been 

 decreases in interspecific competition (Wolfert 1969; 

 Forney 1977) which results in density dependent in- 

 creases in growth rates. The quantity and quality of 

 food are important factors in walleye growth (Kelso 

 1972; Kerr and Ryder 1977; Kitchell et al. 1977b) 

 and fecundity (Colby and Nepszy 1981). Schupp 

 (1978) looked at the growth of walleye from several 

 areas within Leech Lake, MN, and found food of 

 walleye from areas of highest average growth was 

 almost totally young-of-the-year yellow perch, 

 whereas small walleyes from slow growth areas had 

 eaten mostly invertebrates and small minnows. We 

 have found (Maule and Horton 1984) that about 99% 

 by volume of Columbia River walleye stomach con- 

 tents were fish (ag., sculpins, suckers, cyprinids) and 

 that 61% of walleye sampled contained food. 

 Eschmeyer (1950) reported that 89% of the volume 



of stomach contents from Lake Gogebic walleye was 

 fish, but he did not report percent empty stomachs. 

 Dendy (1946) reported that Norris Reservoir wall- 

 eye stomachs contained 99% fish by volume, but only 

 45% of the walleye examined contained food. Rawson 

 (1957) studied Lac la Ronge walleye and reported 

 that fish comprised 97% of the volume of stomach 

 contents and that 39% of the walleye stomachs con- 

 tained food. 



Colby and Nepszy (1981) stated that age to matu- 

 rity is indirectly correlated to growth, but that fecun- 

 dity is probably a function of population density and 

 food availability. They further suggested that the 

 wide variability in walleye fecundities is a mechanism 

 by which walleye can adjust production in response 

 to environmental conditions. Ikble 3 includes fecun- 

 dity data from Norris Reservoir (Smith 1941), Lake 

 Gogebic (Eschmeyer 1950), and western Lake Erie 

 (Wolfert 1969). Based on a comparison of growth, 

 stomach content analysis, and fecundity the 

 mid-Columbia River walleye have a more favor- 

 able food supply than the other areas considered 

 hera 



Hackney and Holbrook (1978) suggested that there 

 is a southern race of walleye that is characterized 

 by rapid, large growth and short life span, and a 

 northern race characterized by slow growth and long 

 life span. They suggested that the pattern of rapid 

 walleye growth seen after the impoundment of 

 southern waters, followed by decreased growth rates 

 some years later is due to a shift from the southern 

 race to the northern race as the result of walleye 

 stocking programs. The movements of young-of-the- 

 year walleye downstream past Columbia River dams 

 has been documented (Brege 1981). Assuming that 

 this is a means by which walleye have colonized the 

 Columbia River, it is biologically similar to impound- 

 ing waters already containing walleye populations, 

 in that new habitat is available for population growth. 

 Although we cannot discount the possibility that the 

 extreme life history characteristics reported here are 

 the result of genetic stock differences, we suggest 

 that they can more reasonably be explained by a 

 favorable temperature regimen and an abundant, 

 high quality food supply. 



Acknowledgments 



We thank Hiram Li and Carl Bond for their 

 reviews of the manuscript. Funding was provided by 

 the U.S. Army Corps of Engineers contract DACW 

 57-79-C-0067, the Oregon Agricultural Experiment 

 Station, and the Milne Computer Center, Oregon 

 State University, Corvallis, OR. 



705 



