260 



FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE 



obtained during the breeding season suggests that 

 spawning in itself may protect immature females 

 since only 3 of 27 females (11.1 percent) in the 



entire sample were immature. The largest of the 

 immature females in this sample was 8}^ inches 

 total leng;th. 



SEX RATIO 



The number of specimens, the sex ratio ex- 

 pressed as the percentage of males in the total for 

 the daily collections, and the ratios for the com- 

 bined collections of each month are shown in table 

 37 for samples containing 50 or more fish. All 

 samples were obtained from commercial impound- 

 ing nets. The sex ratio of the individual samples 

 fluctuated rather widely within each month except 

 August, September, and December, 1929, and 

 April 1932. This wide fluctuation points to a 

 segregation of the sexes throughout much of the 

 year. A segregation may occur, however, in a 

 month in which the sex ratio is not highly variable 

 (as the predominance of males in April 1 932) . 



The wide daily variation in the relative abun- 

 dance of females and males in the samples makes 

 it <lifficult to determine a truly reliable sex ratio 

 for the perch of Lake Erie. The data in table 37 

 suggest that a large number of relatively small 

 samples, preferably distributed thi'oughout the 

 season, will permit a more accurate estimate of the 

 relative abundance of the sexes than may be 

 obtained from a few large samples. Table 37 

 reveals that the average ratio for all samples com- 

 bined, except those taken in April, was 96 females 

 to 100 males (50.9 percent). The April samples 

 were omitted from the computation because the 

 sex ratio obviously was distorted by the presence 

 of disproportionately large numbers of males. 



The April (1928 and 1932) collections were con- 

 sistent in the strong preponderance of males. 

 The males predominated also in both samples ob- 

 tained late in May 1929. On the other hand, the 

 females were relatively more abundant in the 

 samples obtained May 9 and 10, 1929. One 

 sample collected during June 1929 indicated that 

 the sexes were segregated, whereas the other 

 showed no marked preponderance of either sex. 

 The data obtained during July 1929 showed an 

 increasing proportion of males as the month ad- 

 vanced. The males were somewhat more abun- 

 dant than the females in each sample taken during 

 August, but on no date, with the possible excep- 

 tion of August 6, were they strongly predominant. 

 No strong dominance of either sex was evident in 



the material taken in September. The single 

 October collection contained nearly 70 percent 

 males. Only two of seven samples taken during 

 November (November 26, 1929, and November 7, 

 1937) showed sufficiently disproportionate repre- 

 sentation of the sexes to be interpreted as indic- 

 ative of segregation. There was little evidence 

 of segregation on the two dates in December. In 

 general, then, it appears that the sexes were segre- 

 gated during April and May, and probably to 

 some extent during part of June, July, August, 

 October, and November. There is no evidence of 

 any distinct segregation during September and 

 December; however, only two samples were taken 

 in each of these months. 



Table 37. — Percentage of males in the daily samples of 

 Lake Erie yelloiv perch 



