TWENTY-EIGHTH BIENNIAIj REPORT. 



63 



It is entirely possible- that some such inadequacy of data lies at the 

 basis of the prevalent idea that "pelagic" species snch as the herring 

 are inexhaustible. 



I would therefore like to propose in this biennial report the use pend- 

 ing the perfection of our knowledge, of the only alternative method 

 possible, which is regrettably crude. This, the direct observation of the 

 fishing methods, with careful record of place of fishing, apparatus, 

 methods, and sizes of catches, should allow us to roughly compare at 

 any desired time the relative difficulty of taking the two classes of fish, 

 old and young. These two classes run at different times of the year, 

 and these observations should allow us to make some estimate of their 

 actual abundance relative to one another. In other words, our work has 

 shown us that instead of observing a single unified fishery for the size of 

 the catches and duration of life, we find ourselves compelled to correctly 

 determine the catches in two more or less independent phases of the 

 fishery, and to compare the course of the two. This is a more funda- 

 mental and more difficult task than one would realize at first glimpse. 

 It throws a very great responsibility upon those having in their care 

 the collection of fishery statistics and the observation of fishery methods. 

 But could it be shown that the fishery for old fish is persistently on a 

 decline in comparison to that for young fish it would be fair to presume 

 overfishing, unless change in methods or apparatus accounted for the 

 increase in young. 



However, to summarize the whole, there can not be a shadow of a 

 doubt that the securing of more refined evidence may be delayed beyond 

 wliat it should be — our work has clearly indicated that to the writer. 

 It may well be that our research, carried further, will clear this matter 

 up and provide means whereby depletion can be detected at an early 

 date. Thus our studies upon the seasonal run of fish may show that at 

 some one part of the season the various classes of fish are all represented 

 at once, and that a correct evaluation of their relative abundance may 

 be obtained. In fact, the development of our knowledge and the per- 



o 



cvj 





cvl 





of Iba. 



100 

 90 

 80 

 70 

 60 

 50 

 40 

 30 

 20 

 10 



Fig. 15. California sardine catch by statistical districts, 1915-1923. 



