238 



Part III. — Eighteenth Annual Report 



It has been shown in these pages how the relative dimensions of various 

 parts of the body — e.g., the skull — may vary with sex, age, and in different 

 regions. The exact amount of the changes observed have been stated, 

 but the conclusions have not been pressed too far. It is believed that the 

 changes in the dimensions are only the superficial evidences of what 

 remains constant throughout. The various structures and organs are so 

 bound up with one another that any change in the one may be accom- 

 panied by changes in others, but these relative changes mean that a 

 certain balance of the organs is constantly maintained. 



How this balance of the various organs and structures is to be expressed 

 according to the mathematical method of treating variations is still an 

 unsolved problem. The correlation of two, and even three, structures can 

 be displayed by Pearson's and Duncker's methods of treating correlation, 

 but this only shows the amount that one part varies in company with 

 another ; or the correlation of the averages may be worked out by 

 Heincke's method, and this is perhaps the nearest approximation to a 

 solution of the problem that has yet been made. This shows how an 

 extreme variation in one part of an individual may be counterbalanced by 

 variations in other directions of other parts. But, apart from the inherent 

 difficulties in connection with the method, this leaves us in the end with 

 facts illustrative of the simple statement that there is a balance of organs. 

 In the present paper it has been shown that the balance is most probably 

 one of mass or w T eight about a certain point in the body, which may be 

 called the centre of mass or gravity. The changes in the relative dimen- 

 sions have been referred to the need of preserving constantly this balance 

 It therefore follows that in order to express the balance of organs they 

 must all be reduced to the same common denominator — that of mass or 

 weight. The practical, as welPas the mathematical, difficulties in the way 

 of expressing this balance are very great, and all that can be done at 

 present is to show that the solution lies in this direction. 



Under the section on race-variability it has been demonstrated that a 

 " species " may be naturally divided into groups, which under normal 

 conditions may have no connection with one another. The practical 

 bearing of this lies in the possibility of telling where any given specimens 

 have come from. The fisherman with long experience is often able to do 

 this almost at a glance, and there is no reason why the naturalist should 

 not be able to do the same in his laboratory. The necessary requirement 

 is that the relative values of the characters at different ages and sizes 

 should be tabulated for the various groups. This should be done for all 

 the common species of food-fishes. The actual need of this has not yet 

 been felt in the case of the marine forms, but cases have arisen amongst 

 the fresh-water forms, such as the trout, where knowledge acquired by 

 the present method would have been invaluable. We cannot tell either 

 when such knowledge of the marine forms, especially of the herring, cod, 

 and plaice, may be required. It simply means that we should be able to 

 distinguish between the different groups of a species just as we are at 

 present able to distinguish the species, although the latter is not at all 

 easy in many cases in the present condition of our knowledge. 



A further important aspect of the practical bearing of this study lies in 

 the ability that arises from it to track a certain group through the sea. 

 If the characters of a certain race — say those of the plaice of the northerly 

 North Sea — have been carefully tabulated and differentiated from those 

 of neighbouring races, we should be able to follow these plaice wherever 

 they went. This has already been illustrated in a previous paper. 

 Heincke has shown it also in the case of the Bohuslan herring, and there 

 seems little doubt that it is only lack of definite knowledge that prevents 

 our doing so in other cases as well. 



