LIFE AND MECHANISM. 35 far a world of causes and effects is real. All that has been undertaken is to show that the ordinary conceptions of physical science are insufficient when applied to the pheno- mena of life, and that other conceptions must be substituted. The result of our inquiry up to this point has been to show that biological phenomena are to be interpreted under the category of reciprocity rather than under that of cause and effect. * But we shall see that the category of reciprocity has after all only a relative appropriateness, there being essential characteristics of life which it does not express. In the case of a system whose parts reciprocally determine one another, each part, though determined by and deter- mining the rest of the system, has yet a certain independence of its own. The fact of its taking part in the system is not essential to its own existence, since it has numerous proper- ties which belong to it apart from its relation to the system. Now in the case of the system of life the parts are not thus independent. They are determined, not only as regards their reciprocal action on one another, but also as regards what is inherent in the parts themselves of a system whose parts reciprocally determine one another. Size, shape, consistence, and other properties which we think of as inherent in a thing itself, as well as what relates to its determination of and by other things, are determined, in the case of what is concerned in the life of an organism, with reference to that life as a whole. For instance, it is not enough to say that the action of a limb at the same time determines and is determined by the rest of the organism and by its surroundings. For the size of the limb, its shape, structure, and other properties such as ordinarily appear to us to inhere in a thing itself, apart from its relation to other things, are determined with reference to the function which the limb performs. Let us consider again the phenomena which may be observed in connexion with the amputation of one of the limbs of a newt. From the previous discussion we gathered that the characteristic behaviour of the cells concerned in the reproduction of the limb must remain unintelligible unless the life of the animal be brought under the general conception of reciprocity. But it is not only the movements of the cells relatively to one another that have to be explained, but also the assumption by them of properties which inhere in themselves, such as certain sizes, shapes, and consis- tencies. For the cells which take part in the process assume in each case a certain size, consistence, and shape suited to the functions which they severally have to perform. Certain
