This means that many neuroanatomists may balk at the functional interpretation I have given their sort of data, whereas many neuroscien tists in other fields may balk at references to neuroanatomy they cannot confirm from memory and begrudge the time to look up. Members of both groups may consequently decline to read the book on the ground that it is too risky or too difficult to follow, or both. Let me say, in my own defence, that although the argument presented here draws upon data from a number of different fields, they fit together in ways that are often quite striking; that the problem of memory is in itself an important one; and that theory in neuroscience, if it is to fulfill its function of unifying our understanding by way of testable propositions, has got, at some point, to become interdisciplinary. I submit that the present theory meets those requirements and may therefore justify the effort it asks of the reader. To the extent that it passes the critical test of experiment, it will also give research in the fields related to memory a clearer rationale, thereby, perhaps, saving time. Introduction What follows is an outline of the theory of memory functions developed in this monograph. It will not include the supporting data and references used throughout in the text, but will simply present the essentials of the argument.