another experiment a similar phenomenon can be shown. If the bifurcated tail of a young scup (_Stenopus chrysops_) is cut off by a cross-cut (Fig. 40, _G_, 1-1), it will be found that at first the new material is produced at an equal rate along the entire cut-edge; but it soon begins to grow faster at two points, one above and the other below, so that the characteristic swallow-tail is formed at a very early stage (Fig. 40, _H_) and before the new material has grown out to the level of the notch of the old tail. If the tail of another individual is cut off by an oblique cut (Fig. 40, _G_, 2-2), we find, as shown in Fig. 40, _I_, that at two points the new tail grows faster, but the lower lobe faster than the upper one. These results show very clearly that in some way the development of the typical form of the tail influences the rate of growth at different points. The more rapid growth takes place in those regions at which the lobes of the tail are developing. In other words, although the physiological conditions would seem to admit of the maximum rate of growth over the entire cut-edge, this only takes place in those parts that give the new tail its characteristic form. The growth in other regions is held in check. The same explanation applies to the more rapid growth at that part of an oblique cut that is nearest the base of the tail, for by this means the tail more nearly assumes its typical form. These results demonstrate some sort of a formative influence in the new part. We can refer this factor at present only to some structural feature that regulates the rate of growth. We find here one of the fundamental phenomena behind which we cannot hope to go at present, although it may not be beyond our reach to determine in what way this influence is carried out in the different parts. This topic will be more fully considered in a later chapter.