In his earlier writings Driesch rejected this idea, because it did not seem to satisfy our etiological need, and also because he thought that he could reach his goal from the standpoint of initiating stimuli (_Auslösungen_). Driesch now assumes that the stem of tubularia and the archenteron of the starfish, for example, have a polar structure. Bilateral forms, as the whole larva of the starfish, have a coördinated system of two axes with unlike poles and one axis with like poles, each of a given length or proportion. The ends of the axes are characteristic points of the system. If, in such a system, a typical act of differentiation appears, to which we can assign a cause, so far as the location is concerned, a change will occur as follows: To take the simplest case, that of a system with only one axis having unlike poles, as the archenteron of the starfish, in which differentiation has not begun, we can picture to ourselves the formation of the divisions of the archenteron in a causal way by supposing the end of the axis, or pole, to be the location (_Sitz_) of an initiative “action at a distance” (_auslösende Fernkraft_). This locality, just because it is the end of a system, is something special; and it acts in such a way that wherever an effect is produced, it is the cause of that effect. This very way of looking at the problem postulates a sort of causal harmony. But how, it may be asked, can a special point or pole of an axis bring about an action in the system? This can be shown by means of a simple case, viz. the dividing up of the archenteron of the starfish into its characteristic parts. There are two effects produced, viz. the formation of the two constrictions of the wall. We need not consider the fact that the constrictions are formed, for this is established in the potence of the system, and is awakened by the initiating cause, but the place at which the constrictions are produced is that for which we should account. We must think of this cause as “action at a distance,” and indeed as an “action at a distance” that works at a determinate, typical distance. This inherent measure of distance of the action is not one of absolutely fixed size, for a gastrula made shorter by an operation also subdivides into proportionate parts. The action starts from the poles of the system, and acts, not at an absolute, but at a relative distance, since it is dependent upon the length of the axis of the whole differentiating system. “The localization of ontogenetic processes is a problem _sui generis_. The phenomenon can always be expressed on the scheme of cause and effect, if we assume the ‘action at a distance’ to start from fixed points of a differentiating system.”