272 THE VAKIATION, llACES AND MIGKATIONS OF THE MACKEREL. 



this point in connection with yearling Plymouth fish will be found 

 below. 



A second conclusion is also forced upon us by a comparison of the 

 figures on the left and right sides of the same Table (F), which deal 

 with the frequency of the various values for males and females 

 respectively. Subject to the same reservation in regard to the extreme 

 size-groups, we may say that, for each size of fish, the table shows a 

 perfect agreement between males and females in regard to the relative 

 frcfiuency of high and loiv values. When high values predominate 

 among the males, they predominate also among the females of the same 

 size ; when the percentage of low values rises in one sex, it rises also 

 in the other. There is accordingly a close agreement between the 

 mean number of finrays in males and females of the same size. 



In the series of size-groups of Irish fish, whether male or female, 

 the first distinct fall in the mean number of finrays occurs in the group 

 of 14-inch fish, in which there is a decrease of 0'25 for the males, 

 and 0'29 for the females, as compared with the mean values for the 

 corresponding groups of 13-inch fish. 



Unfortunately the group of 12-inch Irish fish is inadequately repre- 

 sented, and when we pass to the Channel and North Sea fish, we see 

 that the group of 14-inch fish is there represented in the case of the 

 males by only 26 fish, which is also insufficient to yield a reliable 

 result. This is seen, for example, in the lack of correspondence 

 between the means for the Channel males and females of this size, 

 a feature which is still more noticeable for the 15-inch fish, which are 

 represented by only 6 specimens of each sex. 



In order, therefore, to compare the Irish with the Channel results, it 

 is necessary to enlarge the size-groups, and this I have done by dividing 

 the whole set of fish into three, instead of eight, compartments 

 according to the size of the fish. 



These compartments are as follows : — 



(1) 10, 11, and 12-inch fish. 



(2) 13 and 14-iucli fish. 



(3) 15, 16, and 17-iucli iish. 



As we have already seen that the mean values indicate pretty closely 

 the changes which take place in the percentage distribution of high 

 and low values, it will be unnecessary to recombine the percentage 

 values in a separate table, although anyone desirous of checking these 

 results can readily do so from the data given in Table Y. I will give 

 here merely the mean values for the three compartments already 

 defined. 



