(left or right bank below the dam), ranged from 

 to 42 percent, with mortalities greatest for fish 

 released below the right bank ladder. Tagging 

 data on chinook salmon indicated a loss of spring 

 chinooks released below the right ladder in 1956; 

 no loss was indicated, however, in comparing total 

 returns from below the dam with returns from 

 above the dam. For summer chinooks, tagging 

 data failed to show losses due to the dam. 



Total tag returns at Rock Island Dam, as identi- 

 fied by fish counters, ranged from 64 to 86 percent 

 of the number released below the dam during the 

 experiments. Straying and mortalities may have 

 accounted for many of the missing tags; it was 

 probable also that the counters missed many of the 

 tags, and total returns were greater than indicated. 



Tag returns at Rock Island Dam from below- 

 darn releases showed that although salmon were 

 released at both banks, the majority returned over 

 the left ladder. Only 12 to 14 percent of the 

 tagged salmon returned over the right ladder. 

 The tag returns corresponded closely with the 

 choice of ladders of the untagged populations, of 

 which the majority chose the left ladder for passage 

 and 10 to 17 percent the right ladder. Although 

 the salmon were taken from the ladders for tagging, 

 they did not learn a particlar route in re-passing 

 the dam; neither were they frightened from a 

 particular ladder during repassage of the dam. 



Tagged salmon released above Rock Island Dam 

 arrived at upstream points 2 to 4 days earlier than 

 fish released below the dam. This difference in 

 days out corresponded to the time required for 

 most fish to pass Rock Island Dam after tagging, 

 and is termed the delay at the dam. Fish released 

 below the right ladder apparently were delayed 

 1 to 2 days longer in reaching upriver points than 

 fish released below the left ladder. 



The days-out period at Rock Island Dam, for 

 sockeye released at the two banks below the dam, 

 was essentially the same, with most of the salmon 

 passing in 2 to 4 days, and the peak occurring the 

 third day after tagging. In 1954, right-bank 

 releases were delayed 1 day longer than left-bank 

 releases. This greater delay of right-bank releases 

 was not evident in 1955 and 1956. Spring chinook 

 peaked after 1 day-out, with the majority passing 

 the dam by the fourth day. The peak day for 

 summer chinook was the third day after tagging. 

 No apparent difference in days-out period at 



Rock Island Dam was indicated for chinooks 

 released at the two banks below the dam. 



An increased percentage of the run used the 

 right ladder in 1957, following a change in spillway- 

 gate operating procedures. Evidently attraction 

 to this ladder was increased by opening gates on 

 the right side of the dam next to the fishway. 



On the basis of these tagging studies we found 

 that Rock Island Dam delayed migrating salmon 

 from 2 to 4 days. There is conflicting evidence of 

 a greater delay to fish released below the right 

 ladder than to fish released below the left ladder. 

 Data regarding mortalities gave conflicting results ; 

 some experiments indicated substantial mortalities 

 while others indicated none. While altering the 

 right-bank fishway may attract more fish to the 

 fishway, we cannot say it will necessarily improve 

 passage considering overall loss and delay. The 

 delay of 2 to 4 days may be significant when 

 similar delays at a series of dams are considered. 



ACKNOWLEDGMENTS 



Many persons in addition to staff members at 

 the Seattle Biological Laboratory contributed to 

 the tagging experiments at Rock Island Dam. 

 The Washington State Department of Fisheries 

 helped in planning the project, and provided 

 equipment for the investigations. The Puget 

 Sound Power and Light Company, and Chelan 

 County Public Utility District No. 1 gave valuable 

 assistance by providing equipment and facilities. 

 R. A. Fredin and D. D. Worlund of the Bureau's 

 Biological Laboratory in Seattle assisted with the 

 statistical analyses. 



LITERATURE CITED 



Fish, Frederic F., and Mitchell G. Hanavan. 



1948. A report upon the Grand Coulee fish-mainte- 

 nance project 1939-1947. [U.S.] Fish and Wildlife 

 Service Special Scientific Report No. 55, 63 pp. 

 Fisheries Research Board of Canada. 



1954. Pacific Biological Station. Fisheries Research 

 Board of Canada, Annual Report for 1953, pp. 

 75-99. 

 Howard, Gerald V. 



1948. Problems in enumeration of populations of 

 spawning sockeye salmon. Part 1. A study of 

 the tagging method in the enumeration of sockeye 

 salmon populations. International Pacific Salmon 

 Fisheries Commission, Bulletin No. II, pp. 7-66. 

 Schaefer, Milner B. 



1951. A study of the spawning populations of sockeye 

 salmon in the Harrison River system, with special 



LOSS AND DELAY OF SALMON PASSING ROCK ISLAND DAM 



365 



