262 



FISHERY BITLLETIX OF THE FISH AND WILDLIFE SERVICE 



result in effect in a more intensive fishery for that 

 sex. The monthly records of the Ohio Division 

 of Conservation for the years 1930, 1931, 1939, 

 and 1940 '* reveal that during those years the 

 spawTiing-season fishery (April and May) produced 

 28.3 percent of the perch caught diu'ing the entire 

 year in Ohio waters. Fishing intensity during the 

 spawning season, when the males predominate in 

 the catches, is therefore sufficiently great to ac- 

 count for an important differential destruction of 

 the sexes. The apparent preponderance of fe- 

 males during early May would reduce, but prob- 

 ably not eliminate, the effect of the spawning- 

 season fishery on the changes in the sex ratio. It 

 seems impossible to escape the conclusion that the 

 changes in sex ratio with age of the Lake Erie 

 yellow perch were caused in large measure by a 

 differential destruction by the fishery chiefly diu-- 

 ing the spawning season. 



It should be mentioned that the minimum legal 

 size operates to reduce the effect of the differential 

 destruction of the males in the spawning-run 

 fishery. Large numbers of mature but illegal- 

 sized males on the grounds are captured but 



returned to the lake. However, it is known that 

 about 14 percent of the illegal-sized yellow perch 

 are destroyed, or seriously injured, when the nets 

 are lifted. It is not improbable that the total 

 destruction of illegal-sized males during a spawn- 

 ing season is considerable, in spite of the fact that 

 none enters the commercial catch. 



Other factors that might have had aii influence 

 on the changes in sex ratio with age will be men- 

 tioned briefly. Geiser (1923, 1924a, 1924b) con- 

 cluded that the females (of fishes, as of many other 

 animals) are inlierently more viable than the 

 males under adverse conditions. Hile (1936) 

 stated that a differential natural mortality was 

 the most probable cause of the changes in sex ratio 

 of the Cisco, Leucichthys artedi, with age. There 

 is no fishery for the cisco in the Wisconsin lakes 

 whose populations he studied. Hile pointed out 

 further that there was no basis for any assumption 

 of a differential destruction of the sexes by preda- 

 tory forms. Any possible effect of a differential 

 natural mortality of the sexes would be obscured 

 in the Lake Erie yellow perch because of the 

 differential destruction by the fishery. 



SUMMARY 



1. The annual production of yellow perch from 

 the United States waters of Lake Erie fluctuated 

 about an average of 3 million pounds in the early 

 (1885-99) period of the fishery. The average 

 declined to about 2 million pounds in 1900-1927, 

 increased to over 7 K million pounds in 1928-35, and 

 fell to about 2% milhon pounds in 1936-47. There 

 was a definite tendency for the variability in 

 annual production to increase in each succeeding 

 period except the most recent one (1936-47). 

 The trend in average annual production from the 

 Ontario waters was similar to that in United States 

 waters only in the last two periods, 1928-35 and 

 1926-47. The factors of fishing intensity (in- 

 creases in the number of nets, and improvements 

 in nets, boats, and methods of lifting gear) , changes 

 in fishery laws and the administration of the laws, 

 and abundance were considered in evaluating both 

 tlie long- and short-period trends in the annual 

 production of Lake Erie yellow perch. 



2. In this study, age determinations and growth 

 calculations were made from examination and 

 measurement of the scales of 4,377 yellow perch 



" These four years are the only ones close to the period during which sex 

 data were obtained for which monthly records of catch are available. 



taken by trap nets from Lake Erie. In addition, 

 ages were determined of 576 specimens employed 

 in a special study of the relation between body 

 length and scale length, and of 1,566 fish taken by 

 gill nets. Analysis of the length-weight relation 

 was based on 23,158 specimens, and the length 

 frequencies were compiled from the measurements 

 of 59,779 individuals. The materials were col- 

 lected during the years 1927 to 1930, and in 1932, 

 1934, 1937, and 1943 to 1948. Data from the 

 diflerent sections of th(> lake were combined after 

 preliminary examinations revealed the combina- 

 tion justifiable. 



3. Validity of the use of annuli on the scales of I 

 the yellow perch as year marks was established 

 for the first time on the basis of the following 

 observations: (a) The 1927, 1928, and 1929 col- 

 lections were dominated by the same year class 

 that was represented by larger and, according to 

 scale readings, older fish in each succeeding year; 

 (b) the annulus was on the margin of the scale in 

 the early season but was progressively farther 

 from the margin in raid-July, September, and 

 December; (c) the lengths calculated from scale 

 measurements for different years of life agreed 



