These remarks, I hope, may have sufficiently cleared the ground to admit of our starting afresh on the consideration of such motor automatisms as are at any rate not morbid in their effect on the organism, and which I now have to show to be _evolutive_ in character. I maintain that we have no valid ground for assuming that the movements which are _not_ due to our conscious will must be less important, and less significant, than those that _are_. We observe, of course, that in the organic region the movements which are _not_ due to conscious will are really the most important of all, though the voluntary movements by which a man seeks food and protects himself against enemies are also of great practical importance--he must first live and multiply if he is to learn and know. But we must guard against confusing importance for immediate practical life with importance for science--on which even practical life ultimately depends. As soon as the task of living and multiplying is no longer all-engrossing, we begin to change our relative estimate of values, and to find that it is not the broad and obvious phenomena, but the residual and elusive phenomena, which are oftenest likely to introduce us to new avenues of knowledge. I wish to persuade my readers that this is quite as truly the case in psychology as in physics. As a first step in our analysis, we may point out certain main characters which unite in a true class all the automatisms which we are here considering--greatly though these may differ among themselves in external form.