It was with some reluctance that I consented to speak on this subject this morning for two very good reasons, the first being that our experiments on this subject which are now under way at Cambridge (and which some of you at least have seen and know something of) have not yet reached the point when anything can safely be published, and the second that, apart from these experiments, I have very little to say on the subject that is new or that I can support with sufficient exact data which is not already well known to most of you.

In view of this you may very well wonder why my reluctance was overcome. The answer is simple; it was because I believe the subject to be so important, and also because I have had very considerable practical experience of it spread over many years.