46 J. S. HALDANE. and the surroundings of an organism have their being and motion in a life that is yet one and indivisible. This life, from its nature, cannot be envisaged as a something that is 'here and now'. It can be thought, but cannot be pictured. In this sense it is ideal. But none the less it is far more real than those abstract aspects of it which are commonly taken for reality. Apart from this general conception, it seems impossible to reconcile the conclusions of philosophy with those of the natural sciences, and especially physiology. For if an or- ganism be looked upon as nothing more in ultimate analysis than a mechanism kept in motion by energy supplied from the surroundings, it is impossible to avoid conceptions which philosophy has shown to be absolutely meaningless such as that of consciousness being produced by, or accompanying, certain physical states of the brain. Unless we get over in some way such contradictions as this, either our philosophy or our science must prevail at the expense of the other. A subject in connexion with which difficulties of this sort are peculiarly liable to arise is that of evolution. It is pretty clear that, if life is more than a mechanical process, evolution cannot be regarded merely as brought about by the action and reaction on one another of organisms and their environ- ment. That it is so brought about is, however, commonly believed. It appears to me to be not difficult to show that this belief is not warranted by the facts. I can only, how- ever, discuss the matter here in a very general way. When it is said that variations in the environment of an organism tend to produce in it corresponding changes, and that these changes are transmitted to its descendants, a covert assumption is made the assumption, namely, that the word ' produce ' correctly renders the actual nature of the facts referred to. The peculiarities in the breed of dray-horses, for instance, have been ' produced ' arti- ficially. But these peculiarities occur in parts of a living organism, which parts have all the characteristics of what is living. For instance, the nerves supplying the muscle and other parts which are so largely developed in the dray- horse, would, if divided, doubtless reproduce themselves in the same way as in other parts. And the behaviour of the parts of the central nervous system which are modified to correspond with the increased development of muscles, ligaments, skin, bone, &c., would give evidence of a partici- pation in the one living whole which we saw that an organism is. As, from its nature, that whole cannot be thought of as acted on from without, it is necessary to re-
