monids. Some of these are discussed 

 briefly in the following section. 



Length of fishway.- -It was theorized 

 that if length of fishway influenced time 

 per pool, a fish would spend a progres- 

 sively longer time in each pool as it 

 ascended the fishway. An analysis of vari- 

 ance was made on the pool tinries of indi- 

 vidual Chinook salmon and steelhead tested 

 in August and September in both the con- 

 trol and test fishways. There was no 

 significant difference between the meajis 

 of the pool times (appendix tables give 

 mean pool times). This is not conclusive 

 but admits two possibilities: (1) the higher 

 variances within pools obscured any dif- 

 ferences occurring between pools, or 

 (2) the fishways were too short to detect 

 a tiring effect. Additional studies are 

 planned to explore both possibilities. 



Seasonal variation.- -Data on seasonal 



variation in passage time are limited to 

 three periods throughout the season when 

 groups of steelhead were tested in the 

 control fishway (July 24-27, August 7-10, 

 and 15-20). There are passage times of 

 12 groups of steelhead in each of the 3 

 periods for testing (table 5). An analysis 



3000 



2400 



■ 1800 



• 1200 



- 600 



Figure 9.--Mean of median elapsed time (round dots) of three 

 series of groups of steelhead in the l:16-slope fishway plotted 

 against time of year. Also indicated is the daily count of 

 steelhead (solid line) in the Washington shore fishway for the 

 period July 15-September 1, 1956, 



of variance shows a significant difference 

 between the mean times. 



Table 5. — Passage times of groups of 

 steelhead in the control fishway (Type 1), 

 July and August 1956. 



Median elapsed time 



July 24-27 



August 7-10 August 15-20 



Mean 8.37 



12.73 



15.42 



The means of the nnedians of the 

 three series were plotted against time 

 of year (figure 9). The three points indi- 

 cate a direct correlation between the 

 median passage time and the number of 

 days elapsed from the start of the run; 

 i.e., steelhead moved progressively slower 

 as the season advanced. Daily counts of 

 steelhead passing through the Washington 

 shore fishway are presented to show the 

 distribution of the run with respect to 

 season. 



Time of day. --The median elapsed 

 times ol group tests of steelhead made 

 in July and August were divided into 

 morning (8:00 a.m. to 12:00 noon) and 

 afternoon (12:01 p.m. to 4:30 p.m.) periods 

 (table 6). 



Although differences existed between 

 fishways, and within fishways at different 

 times, no differences can be shown bet^veen 

 morning and afternoon periods in the same 

 fishway on the same dates. 



Rise between pools.- -There is some 

 evidence that the magnitude of rise be- 



