in age and growth had been observed in the 

 summer fishery (fig. 6). The increase in 

 numbers of large fish (25 cm. and more, 

 1930 year class and older) was greater than 

 was expected on the basis of their relative 

 importance in the preceding winter. 



Small-sized fish were very numerous, how- 

 ever, in 1934, for considerable quantities 

 both of undersized and of legal-sized fish 

 were reported discarded at sea because no 

 market existed. From fishermen's estimates 

 of size and from a few samples of measure- 

 ments obtained at sea, it was learned that 

 these discards were mostly of the 1932 brood, 

 supplemented by lesser numbers of the 1933 

 brood. 



It was estimated that of a total catch of 

 approximately 2.4 million pounds of scup 

 made by vessels landing at Virginia ports 

 in January, February, and March of 1934, 

 more than 500,000 pounds, or 20 percent, 

 were discarded at sea. Expressed in num- 

 bers of fish, of a total catch of approxi- 

 mately 4.7 million scup, about 1.9 million 

 fish, or 40 percent, were destroyed and 

 discarded. Because of lack of observations 

 in New York and New Jersey, it was not 

 possible to estimate reliably the number 

 of small fish discarded from the catches 

 landed at these ports, but since small fish 

 are less acceptable in these markets than 

 In Virginia, it is probable that the proportions 

 were even higher. 



In the winter of 1935, the age and length 

 composition of the landed catches was similar 

 to that of 1934. The 1932 year class, which 

 accounted for most of the scup discarded 

 in 1934, figured prominently in the catch of 

 1935 under the market cull of "small", and 

 the 1931 year class moved into the "medium" 

 market category after being present in the 

 preceding winter (1935) as "small" (fig. 8). 

 The most striking difference between the 

 1934 and 1935 catch was the relatively small 

 quantity of the smaller sizes of scup reported 

 discarded during winter 1935. In that winter, 

 most of the discarded fish of this size group 

 were presumably the 1933 brood. Hence, the 

 lack of reports of any considerable discarding 

 indicated that the 1933 year class was actually 



less numerous than the 1932 year class in 

 that area at that time. 



The magnitude of these shifts in the size 

 composition of the catches in each of the 

 past several winters is shown more strikingly 

 on the basis of length data weighted to make 

 them proportional to the number of fish 

 landed in total catch and to catch per day 

 (fig. 8, tables 18 and 19). They revealed 

 that the total catch in number of fish increased 

 steadily from 1931 to 1935. Much of this 

 increase was due to increasing fishing ef- 

 fort by a larger fleet. The shifts of fish 

 from larger sizes in 1931 to smaller sizes 

 in 1932 and from smaller sizes in 1933 to 

 larger sizes in 1934 and 1935 are also obvi- 

 ous from figure 8. 



These sharp changes in size composition 

 were reflected in the catch per day's fishing. 

 For example, when the size composition of 

 scup catches shifted from large sizes in 1931 

 to smaller sizes in 1932, the catch per day's 

 fishing also dropped from 6,240 pounds in 1931 

 to 2,643 pounds in 1932 (fig. 9, table 16). 

 Similarly, in 1933 when the small sizes 

 continued to be relatively more important 

 than the larger sizes, the catch per day's 

 fishing remained low and even dropped to 

 1,336 pounds. In 1934 when the number of 

 medium fish increased compared to 1933, 

 the catch per day increased to 4,008 pounds. 

 In 1934, however, as previously described, 

 the small sizes were exceptionally numerous 

 and, for the most part, discarded. If they 

 had been retained as part of the landed catch, 

 the catch per day's fishing would undoubtedly 

 have been much higher. In 1935, although 

 small fish (about 17.5 cm.) were relatively 

 less numerous than in 1934, the number of 

 large fish was relatively more numerous, 

 the latter undoubtedly helped materially in 

 preventing the catch from declining to a much 

 lower point than 2,846 pounds per day. 



From the foregoing discussion, it appears 

 that the yield of the winter fishery in the 

 short period of five winters (1931-35 inclusive) 

 has fluctuated sharply as evidenced by erratic 

 changes in catch per day's fishing and by 

 marked differences in size composition of 

 the catch. If the principal cause of changes 



36 



