FISHERY BULLETIN: VOL. 71, NO. 3 



son races. Gard and Drucker (see footnote 5) 

 showed that distinct races of sockeye salmon do 

 exist in the Karluk system. 



The decline in productivity of the Karluk 

 runs was shown in Bulletin 10 of the Interna- 

 tional North Pacific Fisheries Commission 

 (1962) by comparing spawner-recruit curves 

 calculated for the periods of 1870 to 1928 and 

 1929 to 1948. Productivity in the latter period 

 was of course much lower than the former. This 

 is more clearly shown by the returns plotted in 

 Figure 2 for the period of 1945 through 1961. 

 The effect of the weir is also shown by Figure 2 

 which demonstrates the low-level of productivity 

 to which the Karluk runs have been driven since 

 the weir was moved to the outlet of Karluk Lake 

 in 1945. 



Walker in 1950 was the first to record injuries 

 to fry and fingerlings as they struggled to swim 

 through the weir. Bevan and Walker tried to 

 reduce the block by installing a miniature fish way 

 in the eastern end of the weir which they built 

 from a design by M. C. Bell. Bevan also tried to 

 reduce the delay to downstream migrating 

 smolts by having the shape of the weir pickets 

 altered to increase the flow of water through it. 

 Rounsefell's note of the method used in early 

 times to destroy predatory Dolly Vardens by 

 seining at the weir mentions that it was found 

 difficult to do this without injuring larger num- 

 bers of sockeye salmon smolts which apparently 

 were mixed with Dolly Vardens. Such schools of 

 smolts as well as fry and fingerlings above and 

 below the weir must have provided unparalleled 

 opportunity for predation since Rogers (1972) 

 showed that without a barrier such as a weir, a 

 school of 15,000 Arctic char which lay below the 

 outlet of the Agulowak River in Lake Aleknagik 

 must have consumed about 4.000,000 sockeye 

 salmon smolts during the 30-day migration. 

 This was about 27% of the total number of smolts 

 he estimated were produced in the Wood River 

 system above Lake Aleknagik in 1971. 



The effect of the weir on adults has never been 

 measured and may not have been great. But 

 Thompson (1945) and later Talbot (1950) esti- 

 mated that sockeye salmon delayed for 14 days 

 at Hell's Gate never reached their spawning 

 grounds. Moreover, at least half the sockeye 

 salmon held for ripening at the old Karluk hatch- 



ery died before they were ready for spawning. 

 The weir must also have prevented adults from 

 moving back downriver to spawn after ripening 

 in the lake. Such was observed by W.F. Royce in 

 1957 at Brooks Lake on the Naknek system. 

 Bevan and also Walker have seen schools of 

 sockeye salmon move out of Karluk Lake and 

 try to swim downstream through the Karluk 

 weir. Roos has also observed schools of sockeye 

 salmon which had ripened in Chilko Lake move 

 downstream in the Chilko River to spawn. 



It is obvious that the Karluk weir has imposed 

 increased mortalities on sockeye salmon — fry, 

 fingerlings, and smolts — and by blocking the 

 river to free movement of adults has reduced the 

 productivity of the Karluk River spawning 

 ground. 



The effect of the weir on the sockeye salmon 

 runs has never been studied, hence the conclu- 

 sion that it, along with the improperly regulated 

 fishery, has been one of the basic causes of the de- 

 cline of the Karluk runs can only be supported 

 by indirect evidence. Yet, this evidence is so con- 

 vincing to yis that the elimination of the Karluk 

 weir assumes a primary place in any program 

 for restoration of these runs. With other methods 

 for enumeration of salmon runs, such as towers 

 for counting and well-planned tagging opera- 

 tions, much more could be learned about the 

 Karluk sockeye salmon than has ever been 

 learned from the weir. Of course the study of 

 spawning and fry emergence and migration 

 would have to be extended from Karluk Lake 

 into the Karluk River — provided these studies 

 are carried out without interfering in any way 

 with either young or adult salmon. 



Finally, adjustment of the regulations to 

 protect the midseason spawners, recommended 

 by Thompson and Bevan (1954), is a second 

 requirement. The immediate reversal of the 

 downward trend in the returns from individual 

 spawning years when the fishery was closed for 

 extended periods in the Karluk district from 

 1951 through 1956 demonstrated that protection 

 of the midseason fish is essential to recovery of 

 the run. It is impossible to say how far this re- 

 covery would have gone with the weir still in 

 place just below the lake. The present critical 

 state of the run since 1957 has resulted at least 

 in part from increased fishing pressure. 



646 



