of the Fishery Board for Scotland. 17 



caught in the southeru parts referred to, as is explained in my paper 

 dealing with the statistics in connection with the place of capture in the 

 Board's Twentieth Annual Report.* 



For the purpose referred to, certain sizes have been selected as separat- 

 ing the mature from the immature individuals of the various species of 

 fisli dealt with ; in several instances they exceed the sizes assigned in my 

 earlier papers on the subject,t where the limit had reference rather to 

 the smallest mature individuals which were found than to the average 

 size of the group or generation on first attaining maturity. The latter, no 

 doubt, is the preferable course in many cases, but not in all, as is some- 

 times supposed. 



The subject, indeed — the fixing of the line to separate the mature from 

 the immature, so as to include as few of the latter with the former as 

 possible and vice versa — is not by any means as simple as it looks. It 

 is really in some degree a complex problem, and the degree of complexity 

 varies in different cases. If the reproductive generation — that is, the 

 group which first attains maturity — were distinctly separated from the 

 next younger generation or group, then the proper limit would be natur- 

 ally the point between ; on one side all the fishes would be immature and 

 on the other side all would be mature, and in such an example the proper 

 limit would be, not the average size at first maturity, but the size of the 

 smallest mature fish that could be caught. On the other hand, if the 

 first reproductive generation were so fused with the next younger genera- 

 tion — if the over-lapping between the two was such — that it contained, 

 within the range of its sizes, as many immature as the other contained 

 mature, then the proper limit would be the average size at first-maturity. 

 I am not aware of any case in which either of these two conditions occur. 

 In some forms in which reproduction takes place at an early age, as with 

 the whiting and the sprat, the over-lapping of the reproductive generation 

 with the preceding generation is comparatively slight, and in such 

 instances the preferable limit in my opinion is not the average size of the 

 group which is mature — which would exclude a large proportion of the 

 mature fishes and include a very small proportion of the immature in 

 compensation — but a limit placed near the minimum size at first-maturity. 



The approximation to the other extreme is to be found in the larger 

 forms, such as the plaice, cod, &c., where reproduction does not take place 

 at an age so early, and where, consequently, from the variations in the 

 rate or growth of the individuals of the different groups or generations, 

 the first reproducing generation becomes to a certain extent fused with 

 the generation immediately preceding. But I do not know of any case 

 in which the fusion is so complete that half of the fishes comprised within 

 it are mature and the other half immature. With the plaice, for example, 

 a study of the curves appended to my paper dealing with the growth of 

 this fish in the Twentieth Annual ReportJ will show that although 

 a considerable number of the fishes belonging to the younger group 

 next to the reproductive group have fused with the latter, the greater 

 number by far are distinct, and in such instances it appears to me 

 that the proper line of division is not the average size of the repro- 

 ductive group, but the point between the two groups, i.e. where the 

 numbers of immature forms contained within the latter is balanced by 

 the number of mature forms contained within the former. 



The precise differentiation of the mature from the immature is further 

 complicated by the circumstance that the males and females do not in all 



* Part III., p. 80, PI. I. 



t Eif/hth Annual Report, Part III., p. 160 : Tenth, ibid. p. 240, 



X Part III., PI. XIV. 



