may not have been aware of. Equally important, or perhaps even more so, he needs to learn of the pitfalls of existing methods. One might say, wistfully, that it would be nice to be able to pick up somebody's description of a new piece of apparatus, to go into the laboratory, to build it, and to have it work, the first time! There is, however, a serious problem of the interaction between the experiment per se and the theory for which the experiment is designed. Very often, this interaction renders problematic the interpretation of "direct" observations. An example, from experience of the senior editor of this volume, is the question of contact angle hysteresis. (See Chapters 1 and 2.