480 SCHAEFER AND BEVERTON [t'HAP. 21 



hitherto independently of each other is largely because of the different situa- 

 tions, both as regards the type of fishery and the kind of data available, with 

 which the investigators were confronted. The reader who has stayed with us to 

 this point will probably see now why this should be. Thus Beverton and Holt 

 were concerned primarily with the development of techniques which could be 

 used for the assessment of the need for, and effects of, regulation of the North 

 Sea demersal fisheries ; assessment of the effect of mesh regulation was one of 

 the main requirements, although this could not be undertaken without ref- 

 erence to the effect of changes in fishing effort as well. For some of these 

 fisheries, such as those for plaice and haddock, good commercial statistics of 

 catch and effort were available for a number of years, but the fisheries had long 

 since passed their developmental stage and, except for the transitory effects of 

 the two wars, the data covered only small changes in population size which 

 could be ascribed to changes in fishing effort. On the other hand, age- 

 determination of these species is relatively easy and extensive age-composition 

 and other research data were available for both of them. In contrast, Schaefer 

 has been primarily concerned with establishing the relation between catch 

 and fishing effort in the Pacific tuna fishery. Good statistics of catch and effort 

 for this fishery are available almost from its inception, and therefore cover 

 that highly critical range of population size near to the virgin state which for 

 most of the North Sea stocks is lost in antiquity. Age-determination, however, 

 is difficult in tropical species, at least by conventional methods on a routine 

 basis, so that most of the data required for the Beverton-Holt approach were 

 not easily available. It will be appreciated that the two approaches would not 

 have been interchangeable between the two situations. 



Of course, if the development of a fishery from its beginning to a grossly 

 overfished state were fully and reliably documented, and the influence of 

 external causes of trend and variability in the data were either small or could 

 be allowed for from other information, the relation between steady catch and 

 fishing effort could be established directly without recourse to theory of any 

 kind. It would still be necessary, however, to use some form of analytical 

 approach to predict the effect of changes in gear selectivity if these had not 

 already been experienced, and perhaps also to detect as quickly as possible the 

 influence of possible changes in the dynamics of the stock due to environmental 

 changes. In practice, however, this is an ideal which no actual fishery ap- 

 proaches ; therefore some kind of theoretical treatment and assumptions are 

 needed to bridge the gap between past experience and future prediction. In 

 the Schaefer approach these assumptions are made directly about the dynamic 

 properties of the stock and fishery as a whole, at the primary level of informa- 

 tion, as it were ; in the Beverton-Holt approach the assumptions are introduced 

 at the secondary level of information and concern the factors responsible for 

 the dynamics rather than the dynamics themselves. Certain assumptions are, 

 however, common to both — notably that of a proportional relationship between 

 fishing effort and the fishing mortality rate it causes. Indeed, for any proper 

 understanding of the dynamics of a fishery it is essential that the characteristics 



