FISHERY BULLETIN: VOL. 70, NO. 1 



Possible explanations include the misidentifica- 

 tion of tagfs by the counters and multiple obser- 

 vations of the same tagged fish that passed over 

 the dam. fell back, and survived to pass over the 

 dam ajrain. The "falling: back" of salmon over 

 dams is a frequent occurrence on the Columbia 

 River (Johnson, 1965),' although recent studies^ 

 have shown the magnitude of such fallback is 

 not large (usually less than 5'^r ). 



The i^erceiitages of tagged fish recovered by 

 release location (releases pooled by location 

 within years) were not consistent for the 2 years. 

 In 1964 the percentage of tagged fish released 

 on the left bank and subsequently observed ex- 

 ceeded that of the right bank— 95.5 to 81.1. In 

 1965 the comparable percentages were 88.4 and 

 94.9, respectively. 



Distribution by Ladder 



Of the tagged fish sighted in 1964, nearly one- 

 half (46.2 "^r ) chose the left ladder; the center 

 and right ladders lured 32.3 and 21.5^/r , respec- 

 tively. In 1965, 66.3 ^r chose the left ladder, 

 18.2 vr the center ladder, and 15.4% the right 

 ladder. Distribution between ladders was basic- 

 ally the same for each release site within, but 

 not between, years. The percentages observed 

 in the left ladder were 47.6 and 43.3 for the 

 left- and right-bank releases in 1964 but con- 

 siderably higher (67.2 and 65.5) for fish re- 

 leased from the left and right banks, respective- 

 ly, in 1965. However, in both years similar per- 

 centages of the fish not using the left ladder chose 

 the center ladder (60.0 in 1964 and 54.1 in 1965) . 



Between-period comparisons are possible for 

 1965 only when the percentage of tagged fish 

 using the left-bank ladder remained very con- 

 sistent from i)eriod to period, varying only from 

 61.5 to 69.0. 



' Johnson, James H. 1965. Fal'.back of adult chinook 

 salmon at Ice Harbor Dam spillway, May 1964. Final 

 Report to U.S. Army Corps of Engineers for Research 

 Contract No. DA-45-164-CIVENG-63-286. Bur. Commer. 

 Fish., Fi.sh-Passage Research Program, Seattle, Wash., 

 16 p. (Processed.) 



* Personal communication with Charles Junge of the 

 Oregon Fish Commission with regard to experiments 

 with tagged chinook salmon at Bonneville Dam during 

 1970. 



Travel Time from Release to Observation in 

 Fish Ladders 



Travel times — by date of release, release lo- 

 cation, and ladder in which the tagged spring 

 chinook salmon were sighted — are presented in 

 Table 2. Because of the small numbers of fish 

 involved in the 1964 tests, their value is limited. 

 The 1965 experiments provided the most sen- 

 sitive analysis of the time required for tagged 

 fish to pass over Rock Island Dam under en- 

 croachment conditions. 



Results of analysis of variance tests of the 

 hypothesis of no difference in mean travel time 

 between fish released on the right and left banks 

 in 1965 are summarized in Table 3. Regardless 

 of how the data were grouped — whether travel 

 times of the right-bank and the left-bank re- 

 leases were compared period by period, whether 

 adjacent periods were combined, or whether all 

 periods were pooled — no statistically significant 

 diff"erences were found. It is noteworthy, how- 

 ever, that the mean passage time for fish released 

 from the right-bank site was less than for fish 

 released at the left-bank site. Thus, fish released 

 from the right bank were finding and passing 

 over the ladders at least as fast as, if not faster 

 than, those released from the left bank. 



Table 3. — Analysis of variance tests of the hypothesis 

 that spring chinook salmon, tagged and released on the 

 left bank below Rock Island Dam in 1965, traveled over 

 the Rock Island Dam fish ladders equally as fast as fish 

 released on the right bank below the dam. 



132 



