584 U. S. BUREAU OF FISHERIES 
formerly less used, or however much the yield may be affected by extension 
of banks. The question arises in my mind as to the result were this increase 
of efficiency and intensity to stop. Would the total yield thereupon enter a 
decline? It would seem to be doing so in the case of the halibut on the older 
grounds despite the increase in efficiency. 
But speculation is rather useless as to the future of our great fisheries— 
we have no real records as yet to use. We must observe what fisheries we 
have before us, and we must recognize, in the continuity of the decline in 
abundance, a very real threat. We must never forget that there may be a 
lower limit of abundance beyond which the species may not be able to breed 
at all; that if there is no limit to efficiency and intensity there is no limit 
to the decline. And the whole yast future of our fisheries may proceed on a 
grossly wasteful and inefficient, as well as dangerous, course if we do not 
solve our problem. Nor will conservation ever be sane until based scientifically. 
I might briefly summarize the points I wish to make, and then take them 
up in turn for discussion. 
1. Our problem, as reviewed above, requires, first of all, adequate studies 
of yield, not merely total but per unit of gear or effort. 
2, It then requires a study of the reaction of the species to fishing. 
RELATIVE ABUNDANCE 

YEARS 
Fic. 14.—Hypothetical fluctuating abundance of a species of fish, with mathematically 
determined trends to show error that might arise from incomplete data 
3. And, finally, our proposal to restore previous levels of abundance will 
require restrictive measures based on the biology of the species and the 
economies of the fishery. 
The scientific investigation of marine fisheries has come to mean,: to me, 
first of all, the establishment of adequate statistics of yield and abundance. 
This is not a simple problem. It requires accuracy, of course, but it also 
requires the discovery of some measure of abundance—as the yield per skate 
of 210 hooks in the halibut fishery. It implies, still more, the careful observa- 
tion of the unit of gear for small changes, such as the use of small hooks 
instead of large, and the recording of the exact localities. But over and beyond 
this, it implies the study of the categories of fish taken, changes in which 
accompany and may exceed in value, to us, the changes in quantity. The 
study of these changes by means of age analyses, ete. is, of course, biology. 
How it can be separated from statistics is unknown to me, 
Metheds of sampling the catch are still in rudimentary shape. Once the 
data are at hand, there arise the difficulties of interpretation, and the scientific 
point of view becomes of exceeding value. To cite but one case as a warning: 
We have found in the herringlike fishes great fluctuations in abundance. These 
extend over a period of 10 years, perhaps. The result to our curve of abundance 
might possibly be thus (fig. 14) : 
