The aumber of fish migrating might 

 iiffect the proportion moving at night, but 

 no correlation coefficient wets computed for 

 this variable because the components are 

 interrelated (Snedecor, p. l62). The fia- 

 gerllng trap catches may not depict the 

 exact number of migrants that pass Bonne- 

 ville Dam, but it is assumed here that the 

 data approximate the abundance of migrants 

 during the periods studied. In all years 

 except 191+6 the periods were of relatively 

 short duration. Figure 11 shows the leswt- 

 squares regression lines for Chinook 

 flngerlings and yearlings ccmparing aumber 

 of fish and proportion of night migrants 

 in 1952 and 1953- 



On the hypothesis that day-night 

 movement may be associated with time of 

 year, dates were selected when samples of 

 Chinook salmon were sufficiently large to 

 test in 1952 and 1953 (March 1 for finger- 

 lings and April 1 for yearlings). Sampling 

 extended about 95 days for flngerlings and 

 about 35 days for yearlings. Least-squares 

 regression lines were plotted (fig. 12) 

 and correlation coefficients were computed. 

 The correlation coefficients for flnger- 

 lings (0.104; P>0.50) and yearlings (0.352; 

 P>0.1)O) were not significant. This vari- 

 able might be correlated with size of fish 

 because the size of the migrants changes gib 

 the season progresses but seems not to be 

 true. 



2 4 6 8 10 12 14 16 16 20 22 24 26 28 30 

 NUMBERS OF FISH (HUNDREDS)- CHINOOK FINGERLINGS 



1 1 1 1 1 1 1 1 I I r 



J I L. 



30 90 150 210 270 330 390 450 



NUMBERS OF FISH - CHINOOK YEARLINGS 



riguifi 11. --Percent of nl^t-nl^ivnt cUnook flogerllnge aod yearllnge plotted 

 acaljut niaben of flab (19^2 and 1993)- Lines ihovn ai« the 

 leaet-iqu&ze fits. 



10 20 30 40 50 60 70 80 

 DAYS FROM MARCH I 

 CHINOOK FINGERLINGS 



10 20 30 40 50 

 DAYS FROM APRIL I 

 CHINOOK YEARLINGS 



flours. 12--PBrc*Dt of nl^t-mleraDt cMnook fljigorllJigB and yearllngB 

 plotted agalnat daya frca start of saflpllug (19^2 and 1953). 

 Llnea tbcnru are the laaat-squ&re fits. 



River flow and water temperature, 

 which may influence the total number of 

 fish migrating over long periods of time, 

 probably do not change the pattern within 

 24-hour periods. River flow does alter 

 conditions at the dam to the extent that 

 it governs operation of the spillway gates 

 and, to a lesser degree, operation of the 

 powerhouse; in this way it might be a 

 secondary cause of variations in movement 

 of seaward migrants. 



