22 MANAGEMENT OF HIGH SEAS FISHERIES 



maximum as the age of recruitment into the fishery (the size of trawl 

 mesh) was increased. The catch then decreased at greater ages of re- 

 cruitment. At greater intensities of fishing they found that this 

 maximum occurred at higher ages of recruitment and from this they 

 developed their theory of eumetric fishing by extrapolating the 

 model to infinite levels of fishing intensity and by permitting the 

 recruitment age to approach a theoretical critical age attained 

 by the fish. Beverton and Holt discussed the effects of changes in 

 growth rate and natural rate of loss on this model, but they could not 

 find sufficient data to provide a real basis for exploring the effects of 

 such changes on the yield curves. They hypothesized that the yield 

 would continue to rise as age of recruitment and fishing rate ^vere 

 increased. Such a continued increase in yield would require rela- 

 tively minor changes in growth, recruitment, and natural losses as 

 the standing crop was increased to very high levels by raising the age 

 of recruitment. At comparable levels in virgin stocks natural mor- 

 tality balances or exceeds recruitment and growth. It appears, there- 

 fore, that the extrapolation of their model beyond observed condi- 

 tions requires more careful investigation before it can be accepted as 

 valid theory applicable to natural populations. 



The deterministic models described above have contributed a 

 great deal to the development of the management of fisheries. The 

 knowledge gained from them has been useful in the management of 

 the northeastern Pacific halibut and the Eraser River sockeye and 

 pink salmon. But it is unfortunate that the theory has occasionally 

 been interpreted too literally in an attempt to generalize the objec- 

 tives of management for all fisheries. 



While under a particular set of conditions, a point of maximum 

 yield probably exists which corresponds to some level of fishing 

 intensity, this level can only be an average value in natural situations. 

 It will have a standard error which will vary in magnitude in pro- 

 portion as all population parameters vary with population size, with 

 changes in environment, and as the species in question reacts to 

 changes in the populations of other species which compose its envi- 

 ronment. A single valued maximum sustained level of yield, there- 

 fore, probably does not exist for any population of any species of 

 fish living under natural conditions. There is only an average value, 

 around which the value for any stock may vary widely. Even for 

 species with a large number of age classes, variations can be consider- 



