pool, we recorded species and length and 

 then released the fish into either the test 

 or control fishway. When groups of 20 fish 

 were used, releases of one fish at a time 

 were made alternately into the test and 

 control fishways as rapidly as possible 

 until 20 fish had been introduced into each 

 fishway. Species not used in the tests were 

 returned to the collection pool by use of a 

 brail in the release compartment. When 

 individual fish were timed, the fish passed 

 entirely through the fishway before another 

 was introduced. 



Timing of Fish 



Timing started when the fish entered 

 the fishway at weir 54* and ended when they 

 left at weir 60. Individual fish were timed 

 with stopwatches. Two persons tinned fish 

 in each fishway. One recorded total time, 

 and the other recorded the time spent in 

 each pool. Groups were timed with a 20 -pen 

 recorder (figure 3). A push button at each 

 covinting point activated a pen which re- 

 corded the passage of a fish. Counting of 

 fish was done visually. 



Figure 3. — Twenty pen recorder. (Noteaixpensarein operation). 



'Weir numbers refer to elevation above mean sea level. 



Analysis of Passage Time 



Passage time was used as a basis for 

 comparing the 1:8 -slope fishway with the 

 1:16. The median was the measure used to 

 compare the passage times of individuals 

 and a table of confidence intervals was used 

 to test for differences between medians 

 (Dixon and Massey 1951 ).Ameasure termed 

 median elapsed time was used to compare 

 passage times of groups. This measure 

 was determined as follows: time was re- 

 corded for each fish as it entered the fish- 

 way at weir 54, with the recorder starting 

 at zero when the first fish crossed weir 54, 

 The time from zero was also recorded on 

 the same chart as each fishcrossed weir 60 

 when leaving the fishway. The median 

 elapsed time was then determined by sub- 

 tracting the time of the median fish at weir 

 54 from the time of the median fish at 

 weir 60. A t test was used to test for dif- 

 ferences between the means of the median 

 times. 



Use of the measure, median elapsed 

 time, made it possible to terminate a given 

 test even though some of the slower moving 

 fish had not completed their ascent of the 

 fishway. To have obtained an arithmetic 

 mean of the passage tinnes for each group, 

 it would have been necessary to account for 

 the passage of all fish through the fishway, 

 a condition which on occasion would have 

 necessitated considerable delay in testing. 

 In .these experiments, all remaining fish 

 were readily removed from the fishway and 

 a subsequent test was begun as soon as the 

 fishway was cleared. 



RESULTS 



1:16 Slope, 1. 0-foot Rise 



Prior to comparing ascent times in 

 fishways of different slopes, two identical 

 1:16 slope (Type 1) fishways (figure 4) 

 were constructed in the experimental flume. 

 If passage times were comparable in both 

 fishways, then any difference found after 

 slope had been altered would be ascribed to 

 the difference in slope. 



Six groups of 20 steelheadCSa/OTo gairdneri) 

 and 6 groups of 30 steelhead were timed up 

 each fishway (table 2). An analysis of vari- 

 ance indicated no significant difference 

 existed between the means of the median 

 elapsed times of the north and south fish- 

 ways, or between the means of the 20- and 

 30 -fish groups. 



