Richard Wollheim, however, despite his occasional remarks to the contrary, is best seen as borrowing not from Wittgenstein in formulating his theory of pictorial depiction, but, at least implicitly and loosely, from Kant. By emphasizing its Kantian affinities, then, what follows is an attempt to extricate Wollheim's basic position on pictorial depiction from certain recent challenges. It raises initial doubts, suggesting that the matter is messier than is required for Wollheim's reasoning, but then argue that the essentials of his account of pictorial representation do not require seeing-in to the exclusion of seeing-as. Wollheim generally agrees with view about seeing-as but argues that seeing-in stands apart. He first considers dreams, daydreams and hallucinations, which are the 'most primitive experiences' with which seeing-in is 'plausibly connected'. Central difference between seeing-in and seeing-as, from which their various characteristics follow, lies in the different ways in which they are related to 'straightforward perception'.