Lastly, a few words about the very difficult question of _teleology_. In trying to explain Evolution in a mechanical--sometimes called monistic, but in reality natural--way, we exclude anything like a set purpose, a goal, or ideal, a final condition which the organism strives to attain. Unknown, however, to many morphologists, especially embryologists, their writings are full of this teleological notion. Indeed, there are many cases in which an organism becomes changed, and quickly, too, in a way which cannot but be called reasonable. It starts modifications, be they outgrowths, alterations in shape or colour, or the making good of injuries received, which by 'short-cuts' produce the only advantageous result that can reasonably satisfy the new requirement or altered circumstances. Trees growing in precarious positions, after part of the supporting rock has slipped away, throw out new roots, and rearrange some of the old ones in the only way which could save the tree. In animals which have lost part of a limb the wound closes up, and what is left is turned into a serviceable stump--for example, in water-tortoises (creatures in which reproduction of lost limbs does not happen). In frogs and newts the lost part is reproduced, not correctly, but in a good semblance. Tortoises which have had their shell smashed can throw off an astonishingly large portion and renew the bone as well as the overlapping scutes; but this mending is not neatly done. It serves the requirement, but it is patchwork; the new shell is such as no tortoise ever possessed before. Mammals transported into colder countries, or subjected to continued exposure, grow a thicker coat; and the same kind of tree which in a sheltered valley is tall, large-leaved, and soft-wooded, assumes a very different aspect, although perhaps growing into a healthy specimen, when planted on a wind-exposed hill. There is no room, or, rather, no time, to apply to these cases the principle of many variations or the long-continued accumulation of infinitely small changes. The thing is to be done quickly, or not at all. Nor can we explain the mending of a wound, which implies an activity of countless cells, simply as a case of, or similar to, the reproduction of a lost part; against such an explanation militates the almost absolute unlikelihood of that precise injury having happened before to any of the creature's ancestors. Still, I think we are brought near the solution of the mystery by such considerations. We see no difficulty in the regeneration of a few cells, or in the making good of the disturbance suffered by one of the most simple organisms; but we become suspicious when we see that countless cells, not of one kind, but of the most varied tissues and parts of the body, make common cause in remedying a defect in a serviceable way.