From the end of the 1950s to the mid-1970s, a huge amount of work and various controversies focused on the isolation and characterization of the "molecules of memory". Did such molecules exist, and if so, what was their nature? Was it possible to transfer a specific acquired behaviour from a trained organism to a naive one by means of pure macromolecules, RNAs or proteins (peptides)? Was there a molecular code of memory, and if so, what was the code? These experiments have been largely forgotten, although they were published in the best journals and in a plethora of books, and were the subject of active debates in many symposia and meetings. In retrospect, this period in the history of neurobiology appears as an aberration. We now have at our disposal good descriptions of these controversies, what we have yet to do though is to position the entire episode in its scientific, historical and epistemological context, to consider what it can teach us about the value of concepts such as 'information' which still have a very important place in contemporary biology. Finally, it forms an interesting case-study of how science reacts to extraordinary scientific claims (Stern 1999).