Anyone privileged to have known J. J. Gibson, even slightly, will recall the excitement of being with him. Perhaps it came from his delight in odd views you might present to him; or from his zest for discovery; or from his freedom, on reaching new conclusions, to abandon his older ones. All these engaging traits have intellectual equivalents in his theory of perception and receive a final summation in his last book. The Ecological Approach to Visual Perception (1979) [1] opens with an outline of his trajectory from The Perception of the Visual World (1950) [2]-'my explanation of vision was then based on the retinal image' [ 1, p. 1]-through his formulation of the ecological approach 'based on what I call the ambient optic array' [1, p. 1] in The Senses Considered as Perceptual Systems (1966) [3], to its refinement here. The ecological approach presents a world inhabited by observers. It is ordered, hierarchical: a set of. 'nestings', but the units are determined by the needs and interests of the observer, not, as in physical science, objectively established. The life of the observer, his moving and exploring, goes on in a medium that 'affords' this activity, not on a rigid grid of space-time coordinates. The time element is fundamental to this environment and must never be ignored, if we would understand it. 'Actually [he claims], it is a new approach to the whole field of psychology, for it involves rejecting the stimulus-response formula . . . What psychology needs is the kind of thinking that is beginning to be attempted in what is loosely called systems theory' [1, p. 2]. The import of his formulation is vast. It challenges behaviorism and the more sophisticated arguments of cognitive psychology and parallels recent explorations in the philosophy of mind, for instance, in the work of Hilary Putnam. (Gibson asks: 'Why must we seek explanation in either Body or Mind? It is a false dichotomy' [1, p. xiii].) In the last section of his book Gibson considers forms of vision uniquely human: how we see pictures. They are not made to help us move around in the real world. A picture is a record: it 'preserves what its creator has noticed and considers worth noticing' [1, p. 274]. It offers a two-fold apprehension: the perception of a real object (the actual picture surface) and the apprehension of a virtual object (whatever is represented). Motion pictures do this too, but they also come close to natural vision, for they suggest a moving observer-the moving camera. This affinity triggers Gibson's excitement with film as a medium, exceptional among psychologists. 'Filmmakers understand me much better than most people', he would say; and, indeed, those psychologists who have studied film in recent years have been mostly psychoanalysts. Gibson's is an enthusiasm for the sheer visual kinesthesis produced by movies 'the progressive picture', he calls them, protesting against any pat view of film as a sequence of retinal snapshots. In the terms of ecological optics, 'The progressive picture displays transformations and magnifications and nullifications and substitutions of structure along with deletions and accretions and slippage of texture. These are the "motions" of the motion picture' [4].