Page:Proceedings of the Royal Society of London Vol 69.djvu/216

202 individuals he would have seen this also. For in cases of confraternities he must be familiar with the phenomenon of similar variations occurring simultaneously in separate groups of differentiated members.

But let us now suppose we could define differentiation from variation in general, say, as orderly variation. Even so we could not distinguish it unless its order were conspicuous. In a former paper,* Professor Pearson wrote that " the very nature of the distribution of variation, whether healthy or morbid, seems to indicate that we are dealing with the sphere of indefinitely numerous small causes, which in so many other instances has shown itself only amenable to the calculus of chance, and not to any analysis of the individual instance." As I have on many occasions stated, such a description accords ill with the observed facts (A variation. Illustrations to the contrary are numerous and are now becoming familiar ; and even in Professor Pearson's later works references to them are not wanting.

Does not, then, the presence of orderly differentiation, in various degrees, compel us to an analysis of individual instances 1 In plain language, we shall have to pick and choose our cases, and the value of our coefficient of homotyposis will depend entirely on how we do it. Has not Professor Pearson himself been so compelled in more than one of his examples, notably in that of Nigella ? Has he any certainty that such an analysis oiight not to have been made in other examples also 1

He speaks of the extreme difficiilty of determining whether his material is homogeneous in respect of environment, but I miss from his work any deep appreciation of the subtle and evasive quality of differentiation. If any one would obtain a conception of this difficulty let him go to any tree or large plant and set about pruning it, or better, let him try to choose shoots for propagation. Until he tries, it seems simple enough ; but when he begins he finds the shoots are of many complexly differing kinds, and unless he has experience of pruning or of propagation, he will not know which to choose. If he studies the tree attentively, he will soon see that the kinds of shoots are largely definite and, in fact, differentiated. The differentiation may be irregular or regular. That of the leaves may or may not be correlated with that of the shoots. The differentiations may be correlated with the age of the wood, with the absolute size of the tree, they may be peculiar to the variety, or they may be individual to the specimen and defy analysis.!

I am of course aware that Professor Pearson knows all this, but I


 * 'Phil. Trans.,' A, 1896, vol. 187, p. 255.

There are examples not only of differentiations occurring irregularly in one species and regularly in another, but also of the separation of these very forms of differentiation as characteristics of distinct varieties. See for instance the hetero- phyllous Junipers and Cypresses.