CANADIAN FISHERIES EXPEDITION, Idl.'rlS 



117 



samples as is afforded by comparison with samples taken with non-seleetive imple- 

 ments, yet the circumstance in question gives us further grounds for hoping that the 

 fish of medium age at any rate have been taken in something like their natural pro- 

 portion. This presumption is further confirmed on examination and comparison of 

 the curves for the different samples. Here, as will be seen, the different age-groups 

 are represented in by no means equal numbers, and what is more, the samples them- 

 selves agree in this respect to a degree which, considering the small number of si>eci- 

 mens and the largo numbers of age-groups, may be taken as very satisfactory. 



Fig. 28. 



The curves being age-curves, there is, of course, transposition from 1914 to 1915 

 of the two characteristic modes; in 1914: there are comparatively many fish aged 7 

 and 11 years, in 1915 many of 8 and 12 years old, whereas the age-groups 6, 8, 9 and 

 10 in 1914, and 7, 9, 10 and 11 in 1915 contain relatively few specimens. 



In the case of the younger fish, we do not find the same agreement in the different 

 samples. It must in all probability be taken as due to accident that we find, for the.se 

 ages also, an almost perfect agreement when we take the two 1914 samples! together 

 as one. and compare the resulting age-curve for the whole with that for the three 1915 

 samples together. This is shown in fig. 30, where the grouping is arranged according 

 to year-classes, and not in order of age. 



