332 
On the Fundamental Conceptions of Biology 
significant change, has in both instances taken place between the earlier and later 
series. There is nothing here of the nature of " discontinuous variation." You 
may go through hundreds of skulls and find occasionally discontinuity in one of 
Mr Bateson's senses, interparietals, wormian bones, and cases of fused atlas, but it is 
not in the direction of these things (even if they be truly discontinuous, which I 
doubt) that evolution has taken place. It is rather in the non-exceptional 
characters which vary with what Mr Bateson would call normal variation. Of 
course Mr Bateson may say that there is really differentiation there ; it is he, 
however, who identifies " differentiant " diversity and " discontinuous variation." 
If then we have no reason to suppose that any of these marked cases of dis- 
continuity have been sufficiently numerous and sufficiently profitable to lead to 
survival (without artificial protection), why should we suppose that those he 
merely asserts exist, but which he says cannot be distinguished*, have been what 
the marked cases have not been, i.e. the material for evolution ? Fix for a moment 
our attention on man ; his races are distinct, and their distinguishing characteristics 
are in large part, at the very least, those which we know to give continuous 
frequency distributions. Take the case of the skull ; as soon as 20 to 40 measure- 
ments are taken on a population we see at once the special features which 
separate and connect that population with other local races. We see at once 
broad relations connecting ancient and modern Egypt, mediseval and modern 
English, Aino and Japanese; we see also the well-marked differences of such 
groups. In no case, however, is it what Mr Bate.son terms a "discontinuous 
variation," still less a " meristic variation," which differentiates the skulls of local 
races. Mr Bateson cites with approval Virchow's statement that " every deviation 
from the type of the parent animal must have its foundation on a pathological 
accidentf." Well, the markedly "discontinuous" variations of the skull, which 
I personally should describe as due to pathological accidents, are precisely those 
which, whether they are more or less frequent in one or another race, do not form 
the distinguishing racial characters of the races. Mr Bateson tells us that a study 
of the continuous variations such as I have made in my memoirs goes " wide of 
its mark, if that aim is the elucidation of evolution." I believe, if we can once 
grasp how the local races of man, even in one organ like the skull, have become 
differentiated from one or more common stocks, we shall have reached the first 
definite stage in the solution of the problem of evolution. But the worker who 
endeavours to solve this question of the local races in man by tabling either 
" discontinuous " or " meristic " variations will make small progress. And if the 
continuous variations can be shown to be a sufficient source of the divergent 
characters in the local races of man, and the so-called discontinuous variations to 
have no importance, we have at least a probable basis for attacking the problem 
* "The attempt to exclude differentiation by definition must constantly fail in practice" (p. 205). 
That is the "issue in a word " according to Mr Bateson. 
t Materials for the Study of Variation, p. 74. Of course every individual deviates from the type of 
its parent, as everyone who has measured a population of parents and children must recognise. But as 
usual the word "type" is here being used biologically or without quantitative definition. 
