Between the external stimulus (the atmospheric undulations impinging on the auditory membranes, or the light radiations impinging on the retinæ) and the behaviour-reaction something intervenes. This is the individual history of the organism, the “associative memory” of Jacques Loeb, the “physiological state” of Jennings, the “historical basis of reacting” (historische Reaktionsbasis) of Driesch, or the “duration” of Bergson. The last concept is the most subtle and adequate one and we shall adopt it. The physical stimulus, then, leads to a state of consciousness, a perception, and this is succeeded by the action. What is the perception? There may be no perception in a reflex action; there is none in a taxis.[23] These kinds of reaction follow inevitably from the nature of the stimulus--depend upon the latter, in fact; but we cannot fail to observe that the intelligent behaviour of the higher animal involves choice between alternative kinds of action. The perception, then, is this choice, or it is intimately associated with it. But it is something more than the choice of one among many kinds of response. The whole past experience of the animal enters into the perception, or at least all that part of the past experience which illuminates, in any way, the present situation. What the intelligent animal does in response to a stimulus depends not only on the stimulus but on all the stimuli that it has received in its past, and on all the effects of all those stimuli. Into the perception that intervenes between the external stimulus, then, and the action by which the animal responds what we usually call its memory enters. Its _duration_ is really the something which is changed by the stimulus, and which then leads to the behaviour-reaction. [23] We do not find this explicitly stated in this way in mechanistic biological writings. None the less it is implied, and is the legitimate conclusion from the arguments used. Duration, then, is memory, but it is more than memory as we usually think of this quality. The past endures in us in the form of “motor habits,” and when we recall it we may act over again those motor events. Careful introspection will readily convince the reader that in recalling a conversation he is really _speaking inaudibly_, setting in motion the nerves and muscles of his vocal mechanisms. Actions that have been learned endure; in some way cerebral and spinal tracts and connections become established and persist: undoubtedly when a cerebral lesion destroys or impairs memory it is these physical nerve tracts and cells that become affected. But in addition to this we have pure memory (Bergson’s “souvenir pur”). What, for instance, is the visual image of some thing seen in the past, which most people can form, but pure recollection?[24] [24] A visual image may, of course, be something that has never been actually seen. But then its elements have had actual perceptual existence in the past.