256 



FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE 



best to poorest condition is: 1930, 1928, 1927, and 

 1929. 



The K averages of 2.24 in April 1932 and 

 October 1934, and 2.18 in November 1937 (p. 255), 

 suggest that the condition of the Lake Erie yellow 

 perch in these 3 years was excellent — probably 

 superior to that in 1927, 1928, and 1929. 

 Fluctuations in value of K with stage of maturity 



Great differences in value of K associated with 

 stage of maturity were foimd in the comparison 

 of the gravid and spent females taken in May. 

 (Unfortunately, gravid and spent males were not 

 recorded separately in the field records.) The 

 detailed information on the loss of weight by 

 the females at spawning is presented in table 32 

 where the data have been arranged to show the 

 average weight in grams and the average K before 

 and after spawning for each 5-mm. standard- 

 length interval. For both gravid and spent 

 females the changes of K with increase in length 

 appeared to be random rather than to exhibit a 

 progressive increase or decrease. Consequently, 

 there was no obvious relation between the per- 

 centage loss of weight and the length of the speci- 

 mens. The loss of weight varied from 3.4 to 

 24.6 percent. The average of the percentages 

 computed from the best-represented intervals, 

 those in which both gravid and spent fish were 

 represented by seven or more specimens, showed 

 a weight loss of 16.1 percent at spawning. A 

 slightly lower estimated loss of weight (15.5 per- 

 cent) was obtained from the weighted-average 

 coefficient of condition. 



Fluctuations in value of K with age 



Data for the study of the variations of K with 

 age are given in table 33. To avoid the distorting 

 effects of monthly and annual variations in con- 

 dition, averages are given for each month's col- 

 lection of each year. Since sex and stage of matu- 

 rity have little influence on the value of K in the 

 summer and autumn, the data include all the fish 

 whose ages were determined. The data of table 

 33 do not point toward any dependence of condi- 

 tion on age. It is true that in 7 of 10 comparisons 

 fish of age group II had lower average coefficients 

 of condition than those of age group I. This 

 difference can be explained, however, as the result 

 of gear selectivity. Since the group-I fish were 

 near the smallest size that could be retained by 

 the impounding nets it is readily conceivable that 

 only the heavier individuals of that age group 

 were retained. There is less indication that gear 

 selectivity affected the K values of age groups II 

 and III although numbers of the group-II yellow 

 perch were below the theoretical maximum length 

 of escape (170 mm.). Gear selection possibly 

 may account for the fact that group II had the 

 larger average Kin 9 of 10 comparisons for months 

 earlier than October (see p. 221). In the later 

 months, after group II has practically completed 

 the third season of growth, age group III had the 

 higher K values in all 7 comparisons (October, 

 November, and December). Comparisons of age 

 groups III and IV reveal that the former had the 

 higher average K 6 times whereas the latter had 

 the higher value 2 times. The two age groups 



Table 32. — Comparison of average weights and condition (K) of gravid and spent female ijellow perch taken by impounding 



nets in Lake Erie, May 1929 



(Number of specimens in parentheses] 



' Unweighted mean, based on those length intervals in which both gravid and spent fish are represented by at least 7 individuals. 



