154 ALASKA FISHERIES AND FUR INDUSTRIES IN 1919. 



above question. It was most unfortunate, for this reason, that 

 Wood River was selected for the purpose. Its disadvantages were 

 clearly seen and were discussed by Marsh and Cobb, in their report of 

 the operations during 1908. The Wood River is not the only red- 

 salmon stream tributar}" to Nushagak Bay. The main Nushagak, 

 the Igushik, and the Snake also contribute their quota to the schools 

 of fingerlings passing seaward through Nushagak Bay, and each of 

 them receives its percentage of the escapement from the fisheries 

 conducted in the open bay. It is not known what their respective 

 percentages amount to. The streams have never been racked and 

 the escapement tallied. It is currently believed that their runs, 

 individually or even combined, are less important than the Wood 

 River run, but the a})proximate number that must be added to the 

 Wood River tally to produce the annual escapement from the entire 

 fishing district is wholly a matter of conjecture. The salmon bound 

 for these four streams enter Nushagak Bay togetlier and are indis- 

 tinguisliable. Those captured by the commercial fisheries include 

 members of all four colonies commingled, in their due proportions. 

 What these proportions are no one knows. The chronicled escape- 

 ment is for Wood River alone. It is evident that the total es- 

 capement from the commercial fishery and the proportion which this 

 escape bears to the total run can only be obtained by assuming 

 certain values for three unknown streams. Marsh and Cobb have 

 done the best that can be done with a bad situation by assuming 

 what they consider maximum and minimum values for the escape- 

 ments to the unknown rivers. Interesting results are obtained in a 

 field in which previously there has been no information, but the 

 elements of uncertainty which reside in all conclusions based on the 

 incomplete data of this experiment unfit them for either practical or 

 strictly scientific purposes. Had it been practicable to rack each of 

 the rivers and to obtain a census of the escapement into each, there 

 would now be something certain on which to build. But the enter- 

 prise would be unnecessarily arduous. The desired results coidd be 

 more easily achieved b)^ abandoning the Wood River experiment 

 and choosmg for the purpose some river like the Chignik or the 

 Karluk, where all the spawning fish of a given fishing district enter a 

 lake or lakes through a single channel. But the Wood River census 

 has not given us with any certainty the number of spawning fish 

 which are necessary to maintain the Nushagak run, nor is the number 

 known even approximately. 



Marsh and Cobb have stated that although the census figures may 

 not have absolute value, they can safely be used for purposes of 

 comparison between one year and another. If this were true, we 

 could still use them for purposes of prediction, as discussed in a 

 preceding ])aragraph. If the total escapement from the Nushagak 

 fishing grounds was always, year after your, the Wood Jiiver escape 

 multiplied by a constant factor, the Wood River figures would be 

 just as valuable for purposes of prediction as though we knew what 

 the factor in question was. If the Wood River es(;ape in a given 

 year were half that of the previous year, we could then be assured 

 that the entire escape for the district was half that of the previous 

 year. But unfortunately, even for this purpose, the figures arc 

 unreliable. No constant factor can be assumed. Such assumption 

 would register belief that the runs to the four rivers woidd in diflercnt 

 years alwa\'s vary in the same direction and to the same amount. 



