[When recording electrical impulses from a frog nerve-muscle preparation seemed to show a tiresomely oscillating electrical artefact—but only when the muscle was hanging unsupported.] The explanation suddenly dawned on me ... a muscle hanging under its own weight ought, if you come to think of it, to be sending sensory impulses up the nerves coming from the muscle spindles ... That particular day’s work, I think, had all the elements that one could wish for. The new apparatus seemed to be misbehaving very badly indeed, and I suddenly found it was behaving so well that it was opening up an entire new range of data ... it didn’t involve any particular hard work, or any particular intelligence on my part. It was just one of those things which sometimes happens in a laboratory if you stick apparatus together and see what results you get.

Unless social sciences can be as creative as natural science, our new tools are not likely to be of much use to us.

We come back then to our records of nervous messages with a reasonable assurance that they do tell us what the message is like. It is a succession of brief waves of surface breakdown, each allowing a momentary leakage of ions from the nerve fibre. The waves can be set up so that they follow one another in rapid or in slow succession, and this is the only form of gradation of which the message is capable. Essentially the same kind of activity is found in all sorts of nerve fibres from all sorts of animals and there is no evidence to suggest that any other kind of nervous transmission is possible. In fact we may conclude that the electrical method can tell us how the nerve fibre carries out its function as the conducting unit of the nervous system, and that it does so by reactions of a fairly simple type.