FISHES OF THE GULF OF MAINE 199 



However, this period of multiplication fell far short of equaling the period from 

 18S3 to 1SS5, for the largest catches since the barren years about 1910 were but a 

 fraction of those of the banner years in the eighties. After 1917 the stock once 

 more diminished to such an extent that the catch for 1919 was only about 25 per 

 cent of that of either of the two preceding years. Although 1920 saw a slight 

 recovery, 1921 proved the worst season in the Gulf of Maine region since 1910. 

 That summer, however, must have been an unusually favorable one for the produc- 

 tion of young fish, as little mackerel 4 to 6 inches long appeared in great numbers in 

 the Massachusetts Bay region the following June (1922), swarming in its various 

 estuaries and locally as far north as Mount Desert Island, and raising the catch 

 to slightly more than 3,000,000 pounds throughout the summer, growing meantime 

 to a length of from 7 to 9 inches (p. 205). Their continued growth was probably 

 responsible for the much larger catch in 1923. 



Various far-fetched explanations for these astounding ups and downs in the 

 mackerel catch have been proposed, such as that the fish have gone across to Europe, 

 have sunk, or have been driven away or killed off by the use of the purse seine. 

 However, since similar fluctuations were noticed long before the fishing became 

 intensive, it is safe to say that they are quite independent of the acts of man but 

 bound up with the biology of the fish. There is no reason to doubt that the major 

 fluctuations in the annual catch do actually mirror corresponding changes in the 

 numerical strength of the stock of fish existing in the sea from year to year, just 

 as the annual catches of herring do along the North European coasts. 



In the case of herring the prime factor in determining the abundance of the 

 fish is now known to be the comparative success of reproduction from year to year, 

 years favorable to the survival of the larva? presaging several seasons of abundance, 

 and vice versa. A comparison of the relative proportions of mackerel of different 

 sizes (that is, ages) with the total catches made from year to year, justifies the work- 

 ing hypothesis that this is equally true of the mackerel. 



About 1910, when the stock of mackerel was at its lowest (fig. 93), most of the 

 fish caught were reported to be large, suggesting that few young had survived for 

 several years past. Unfortunately no information is available as to the composition 

 of the catch from the point of view of size for the next three years, when the catch 

 was progressively somewhat larger, but numbers of very small fish, apparently 

 yearlings, were reported in 1912. In 1914 we find fish smaller than l}4 pounds 

 forming nearly 60 per cent of the catch made by the purse-seining vessels in and off 

 the Gulf of Maine, with only about 4 per cent consisting of the large old fish (upward 

 of 2M pounds); and in 1915 small fish formed approximately 80 per cent and large 

 ones only 7 per cent, by weight, of all the mackerel caught in and off the Gulf of 

 Maine, with an even greater preponderence of the former in actual numbers. 



An alteration of this sort in the composition of the stock of any fish, from a pre- 

 dominance of large to a predominance of small, when it accompanies a decided 

 increase in the total weight — still more in the total number — of fish caught, as was 

 the case on the occasion in question, points beyond dispute to an increasing rate of 

 production of young fish, sufficient to much more than offset the annual death rate. 

 As suggested above, 1911 was the first good breeding year in this particular cycle, 

 102274—25 + 14 



