One of the standard art-historical exercises is the search for the pictorial sources of individual works of art. The purpose of this article is to suggest that this exercise need not constitute an end in itself, as it usually does, but that it can yield valuable information about the status of an image in a given social context and about the response it evokes. Rubens's modello of The Assumption and Coronation of the Virgin in Leningrad raises a number of iconographic problems; these in turn are largely resolved by the discovery of an important pictorial source for the work. That source and its relation to the work by Rubens may be used as an illustration of some of the ways in which it is possible to determine the associations which the seventeenth-century beholder made when looking at works of art - even when they were not actually recorded by the beholder himself. While it can never be possible to recover the full range of such associations - because many would have been too personal and idiosyncratic - the art historian may regard it as his province to reveal at least some.