Summary We share the interest in and enthusiasm for the study of the representation and integration of knowledge that marks the present set of papers. A few of them express a common theme regarding the form of this representation, statable as a belief in a common conceptual code for knowledge. We do not find clear empirical support for this view and find, furthermore, that some of the a priori assumptions underlying the view are questionable. We especially question the ready adoption of a notion of equivalence in picture—word testing, and argue that some of the tests are based on a faulty description of their stimuli; and we show further that some of the claims for bilingual equivalence may be accommodated as well by notions of strategy, set, and skill. In the end we suggest that means-specific representations may provide as plausible an account of knowledge as appeals to a common code do. These many conjectures do seem amenable to empirical test and the present set of papers could well provide a stimulus for further study of these fundamental issues regarding the description of the mind's operations.