28 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE 



same size category. Other European investigators, however, have not found that the 

 same phenomenon applies in other species of fish in the North Sea. It is possible that 

 environmental factors, such as low temperatures in the spring and early summer of 

 1934, played some part in the smaller-than-average size of the members of the 1934 

 dominant year-class of striped bass. 



It will be noted in figure 18 that the growth rate of the 2- and 3-year-olds m 



1936 and 1937 was fairly steady over the period from June through October. In 

 general, the modes of the length-frequency curves for the 2-year-olds progressed about 

 2 cm. each month. In October 1936, however, the 2-year-olds appear to have shown 

 an increased growth rate, the mode for this curve having progressed 3-4 cm. beyond 

 that for September. In October 1937 the fish of this age did not exhibit a similarly 

 increased growth rate, but the mode for this length-frequency curve progressed about 

 2 cm . — an amount about comparable to the growth during the summer months. 

 Since the temperature fell sharply in late September and October in both 1936 and 



1937 (see fig. 30), the normal expectation would be that the increase in length at this 

 time would have been less than in the summer months, assuming that the food sup- 

 ply remained constant over this entire period. There are a number of possible ex- 

 planations of this apparently higher growth rate in October. There is some chance 

 that errors in sampling were responsible. Thus it is known that the population was 

 starting to change late in October (see Migrations, p. 37), and there is a slight pos- 

 sibility that fish that had summered farther north, where they apparently grow faster 

 despite somewhat lower average temperatures (see fig. 19) were included in the 

 samples at the end of this month. This does not seem likely, however, for the con- 

 sistent recapture of individuals tagged in this area from June through October gives 

 good evidence to the contrary. Another explanation of the apparently greater growth 

 rate in the fall is suggested by the skewness of the length-frequency curve for October 

 1936. It will be noted in figure 18 that in all curves for the 2-year-olds, except that 

 for October 1936 the peaks come about midway between the two extremes of the 

 range in size, or below that point. In October 1936, however, the peak falls well 

 above the midpoint between the extremes of size, and there is also a tendency toward 

 the same situation in the curve for October 1937. It may be, therefore, that this 

 apparently greater growth rate is possibly the result of "compensatory growth," the 

 name given by Watkin (1927) to the phenomenon of the smaller fish of a single age 

 group making up a deficiency in size between themselves and the larger fish of the 

 same age group in a relatively short period after having lagged behind for some time. 

 The most probable explanation of the increased growth rate in the faU, however, is 

 that the food supply or its availability increased at this time. The analysis of the 

 stomach contents of striped bass is discussed in a later section of this paper, but for 

 the present it is interesting to consider the fact that this species is voracious in its 

 feeding habits and that it preys on small fish, particularly young menhaden {Brevoortia 

 tyrannus) and shiners (Menidia menidia notata) in Connecticut waters. Both of 

 these species spawn in the spring and early summer, and during July the young are 

 still so small and stay so close to shore that they do not form a large part of the diet 

 of the bass. But by late summer, and particularly early fall, they have increased in 

 size to such an extent that they have added enormously to the available food sup- 

 ply. (For information on the growth rate of Menidia, see Food of the striped bass, 

 p. 53, and fig. 36.) The analysis of stomach contents during September showed 

 that striped bass continually gorged themselves on these small fish to the virtual ex- 

 clusion of other tvpes of food. Furthermore, judging from the relative numbers 

 taken in seine hauls in 1936 and 1937, and from the statements of local fishermen, 

 young menhaden were unusually abundant in Connecticut waters in the latter part 

 of 1936. It is likely that these juvenile menhaden were responsible for the greater 

 growth rate of the striped bass in the fall of 1936, and that the increased availability 

 of the food supply in the late summer each year accounts for the maintenance of or 

 increase in the growth rate through October despite the sharp drop in temperature 

 at this time. 



As wdl be shown subsequently, there is evidence that the growth rate of the 

 striped bass varies considerably in different localities along the coast. It has already 

 been pointed out, however, that there was a great vaiiation about the mean in measure- 



