YELLOW PERCH OF LAKE ERIE 



219 



direct-proportion method was therefore employed 

 whenever the calculated lengths exceeded 4.5 

 inches and the empirical curve was used only for 

 the smaller lengths. Since the correction for 4.51 

 inches was less than 0.05 inch, the empirical- 

 curve method was applied only to lengths of 4.46 

 inches and less. 



In practice, all lengths were computed b)' direct 

 proportion, and corrected lengths corresponding to 

 calculated lengths 4.46 inches and less were read 

 directly from table 4, which was prepared with the 

 assistance of the empirical body-scale curve (fig. 

 3). The data for this curve were plotted originally 

 on 1-nim. cross-section paper and the amount of 

 each correction was read directly from this graph. 

 The amomit of correction required for each direct- 

 proportion calculated length is the vertical dis- 

 tance between the extended straight hne repre- 

 senting the body-scale ratio of fish with total 

 lengths of 4.6 inches and more and the empirical 

 line representing the ratio for the shorter fish. 

 The procedure for obtaining the correction for a 

 du-ect-proportion calculated length of 3.25 inches 

 is illustrated in figui-e 3. Line AB is drawn hori- 

 zontally from i=3.25 to B on the straight line 

 representing the body-scale ratio of fish with total 

 lengths of 4.6 inches and more. Line CD is a 



Table 4. — Calculated lengths (inches) of Lake Erie yellow 



perch 



[Total-Iengtb conversion of standard length in millimeters] 



perpendicular that passes thi'ough B from the 

 scale axis to D on the Une representing the bod}'- 

 scale ratio of the smaller fish. The correction is 

 the distance between points B and D. In the 

 present stud}^ only the first-year lengths fell 

 within the range that requu'ed correction. 



CALCULATED GROWTH HISTORIES OF THE AGE GROUPS 



The average weight at capture and the calculated 

 lengths of yellow perch taken from impounding 

 nets in the years 1927-37 are shown by sex and 

 age gi'oup in table 5. Combination of the data 

 for the several years was possible because the 

 corresponding averages varied but little from 

 year to year and the trends in discrepancies 

 between lengths computed from fish of different 

 ages were the same in each of the year classes. 

 The more rapid growth of the females in all years 

 of hfe except the second was evident for each year 

 as well as for the combined years. 



The corrected calculated lengths at the end of 

 the first year of life are seen to be 0.6 inch greater 

 than those obtained by direct proportion for all 

 age groups of both sexes except group I where the 

 difference was 0.4. The smaller amoxint of cor- 

 rection for age-group-I fish is to be expected since 

 they were the larger individuals of their year class 

 and hence their body-scale ratio deviated less 



955513—52 3 



from the straight-hne relation requu'ed for direct- 

 proportion computations. In general, the same 

 remarks may be made regarding the data for males, 

 females, and all fish. Without exception the cal- 

 culated first-year lengths of age-group-I fish were 

 greater than those computed from older fish. 

 The calculated lengths of fish older than age group 

 I revealed a slight tendency for the first-j'ear 

 length to decrease as the fish became older. The 

 discrepancies between the calculated first-year 

 lengths of fish older than group I were small. 

 Comparisons of the calculated lengths for all years 

 of life after the fu-st revealed not only that there 

 was a definite tendency for the lengths to decrease 

 as the fish became older but also that the dis- 

 crepancies each year were larger than in the first 

 j-^ear of hfe. It is to be noted also that, with the 

 exception of group-II fish, the length at capture 

 in the late fall was greater than the corresponding 

 lengths computed from older fish. 



