If the acquired and inherited variations of the Lamarckian theory drop out as a contribution to the explanation of evolution, we are reduced to two forces only—_innate_, or germinal, variability of offspring, and natural selection. Indeed it might be said that we are reduced to variability alone, since natural selection can do nothing until suitable variations are presented to it. The suitable variations do, however, turn up, and the question is, what causes them? The real difficulty for the school of biologists who, like Weismann, “assume the mechanical theory of the world to be correct,” is how to reconcile the aptness and apparent purposefulness of these variations with any mechanical theory. “We are justified in inquiring,” writes Weismann, “whether the assumption of ‘chance’ germinal variations, which we have hitherto made with Darwin and Wallace, affords a sufficient basis for selection. Osborn says very neatly in this connection, ‘We see with Weismann and Galton the element of chance; but the dice appear to be loaded, and in the long run turn “sixes” up. Here arises the question, What loads the dice?’”[67] What loads the dice? There is the great question in which the realms of biology and of philosophy meet each other! Through that borderland no definite frontier has ever been traced, for in thought as in matter the saying is true that natural groupings have nuclei, but no boundaries. It is all the more essential that men of science should understand philosophy and its methods, and that philosophers should understand science. It is to be feared that at present the second of these desiderata is much more fully realized than the first. However, we have to see now what Weismann, protagonist among contemporary biologists of the mechanical theory of the world, has to answer to the crucial question which he has allowed Osborn to set him. The problem is to discover how innate, germinal variations can come about, of such a nature as to adapt an organism with striking accuracy to its surroundings and way of life, without our assuming either (1) that the exercise of function had any influence in causing heritable variations, or (2) that they were caused by any non-mechanical power, which, so to speak, had in view the objects which they fulfil. For the variations are to be regarded, on Weismann’s theory of life, as completely _fortuitous_ in respect of the objects they serve. How, then, do they come to serve them, in most cases, so admirably well?