398 



FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE 



the range of fluctuation of annual increments in 

 weight was greater than that of length and the 

 curves disagree in certain details. Hile (1954) 

 showed a similar situation in the fluctuation of 

 growth of the walleye in Saginaw Bay. This 

 difference between the annual fluctuations of 

 growth in length and weight depends partly on 

 the nature of the length-weight relation (weight 

 varies approximately as the cube of the length). 

 Thus the growth in weight in a particular year 

 varies both according to the amount of increase 

 in length made in that year and with the actual 

 length of the fish at the time the increase is made. 

 In other words, when two groups of fish in a 

 particular year of life have equal increments of 

 length, their weight increments will differ if the 

 lengths were not equal at the start of that year 

 of life. This influence of length when the growth 

 is made on the value of weight increments can 

 be shown clearly in the fluctuation of first-year 

 growth of Saginaw Bay perch. Here, the growth 

 in weight, as that of length, followed the same 

 strong trend toward improvement in 1942-51 

 (fig. 17). Yet the slow growth in length during 

 the years 1942-45 had reduced the length of 

 perch to an extent that the improved growth in 

 length (1946-51) did not bring growth in weight 

 above average until 1949. 



The annual fluctuations of growth in weight 

 in later years of life followed those of length 

 more closely than did those for the first year 

 (fig. 18). Here again, however, the slow growth 

 in length in 1947 reduced the size of the fish 

 enough to delay the attainment of the maximum 

 growth in weight. Although the maximum 

 growth in length was attained in 1948, the high- 

 est value for weight was reached in 1950. 



Still another factor contributing to the dis- 

 crepancies between curves of fluctuations of 

 growth in length and in weight in the second 

 and later years is the difference in the years of 

 life that predominate in controlling the course 

 of the curves. A curve of fluctuation of growth 

 in length (especially for males) is influenced 

 most strongly by the data for the earlier years 

 of life when growth in length is most rapid. 

 Curves of fluctuation of growth in weight, on 

 the other hand, are affected most by the data for 

 later years when growth in weight is most rapid. 

 Although no differences of trend of annual fluc- 



tuations in growth could be detected beyond the 

 first year, the percentage changes in different 

 years of life were not identical. Only random 

 variation in these changes can be a source of dis- 

 crepancy between curves of fluctuations of growth 

 in length and weight. 



Difference in Growth Rate in 1929-30 and 1943-55 

 and Comparison with Growth from other Great 

 Lakes Waters 



Hile and Jobes (1942) and Jobes (1952) offered 

 detailed comparisons of the growth rate of Sagi- 

 naw Bay yellow perch (collected in 1929-30) with 

 that of perch in Lake Erie, southern Green Bay, 

 and northern Lake Michigan. Since this previous 

 discussion need not be repeated, the present sec- 

 tion emphasizes the changes that occurred in the 

 growth of Saginaw Bay yellow perch population 

 between the 1929-30 and 1943-55 collections. 



The calculated lengths (table 43, fig. 19) reveal 

 a pronounced change in the growth of Saginaw 

 Bay perch between 1929-30 and 1943-55. Sagi- 

 naw Bay yellow perch of the 1929-30 samples 

 were second longest or longest for their age, but 

 those collected in 1943-55 had the slowest growth 

 in length yet reported from the Great Lakes. 

 The legal size (8i/ 2 inches) which was reached 

 by 1929-30 fish during the fourth growing season 

 was not attained by 1943-55 perch until the sixth 

 growing season. 



The change in growth of Saginaw Bay perch 

 between 1929-30 and 1943-55 is shown even more 



Table 43. — Growth in length of yellow perch from different 

 localities of the Great Lakes 



[Sources of data: Lake Erie, Jobes (1952); southern Oreen Bay and northern 

 Lake Michigan, Hile and Jobes (1942); Saginaw Bay 1929-30 samples, 

 Hile and Jobes (1941); Saginaw Bay 1943-55 samples, present study] 



' Unweighted means. 



2 No data for sexes separately. 



