352 Fart III. — Tiveniieth Annual Feport 



Excluding the figures for 6th November and 18th October, the 

 increments show a fair uniformity, and indicate an apparent increase 

 from the second to the third annual series of 84'8mm., or 3| inches, 

 from the third to the fourth of 83'2mm., and from the fourth to the 

 fifth year of 82'3mm. It would, however, be erroneous to assume that 

 these amounts represent quite accurately the real increase in length of 

 the plaice in a year, because first, as stated, an examination of the 

 curves shows that the groups are unequally represented, and second, 

 the haul of 6th November, comprising 1898 fishes, the sexes of which 

 were distinguished and carefully measured separately, gives by far the 

 most regular curves (PI. XIV). 



The unequal representation of the various groups in the other hauls 

 is well shown by the curves. As a rule, only one group was present in 

 approximately due proportion, the groups of younger fishes being 

 represented principally by the larger individuals of the group, while the 

 larger individuals of the older groups were for the most part absent. 

 The consequence of this is that the maximum ordinate of the curves of 

 the smaller group and of the larger group does not stand over the centre 

 of the base of the group (whose limits have been previously determined), 

 but is approximated to the maximum ordinate of the central group, 

 which is duly represented by both small and large individuals; and of 

 course the arithmetical average derived by computation from the 

 measurements det'iates in like manner. The distance between the 

 maximum ordinate of the perfect group and the ordinate standing over 

 the middle of the base of the imperfect groups measures nine or ten 

 centimetres, as in the case of 6th November. 



This unequal representation of the groups is, as stated, brought 

 about by the depth of water in which the hauls happened to be made, 

 and it will be seen from the depths given in the Table above that the 

 haul in which the shallowest water was entered was that of 6th 

 November. In most of the hauls, as a rule, only one group was well 

 represented, and it was usually the fourth; but in the hauls on 6th or 

 7th June the third group was also present in nearly due proportion, and 

 in this case the apparent annual increment approximates closely to that 

 indicated in the haul of 6th November. The haul of 18th October, 

 which gives anomalous increments, comprised only 142 fishes, from 

 sixteen fathoms, and the curve is very irregular. 



It appears, therefore, that the yearly increment of growth of the 

 plaice in Aberdeen Bay is most accurately shown in the haul of 6th 

 November for the various groups, and for Groups III., IV., and V. 

 the measurements may be taken as indicating pretty nearly the real 

 growth in a year from one to the other. But a careful examination of 

 the curves makes it evident that the first group (II.) and the later 

 groups are not properly represented. The fish, as already stated, were 

 obtained in two hauls, one with the fine-meshed net, chiefly in six 

 and seven fathoms, and one with the ordinary otter-trawl in sweeps from 

 eight to thirteen fathoms, the first haul lasting for thirty-five minutes, and 

 the second for four hours and twenty minutes. Series II., except six fish, 

 were entirely the product of the first haul, while the remaining series 

 were entirely the product of the second haul. On the time basis the fish 

 in the first haul should be multiplied by a little over seven, and this would 

 to some degree extend the range of the group (and the base of the curve), 

 but it ought not materially to alter the maximum ordinate or arith- 

 metical mean. It will be observed from the one-centimetre curve 

 that the interval between the base of the columns representing 

 the second and third series is considerable. This would to some extent 

 be improved by the multiplication of the first series as described, but 



