There is, however, a use of the term concerning which it seems necessarily to enter a word of protest. Impulse is by some regarded as the underlying cause of the conative tendency. Now science, as such, has nothing whatever to do with underlying causes. If, as a matter of observation and inference, we have reason to believe that there is such a tendency, science simply accepts the fact, and endeavours to formulate the conditions under which it arises, and to trace its observed or inferred antecedents. No doubt many of us find it difficult or impossible to rest content with the strictly scientific position, that of unquestioning acceptance of the facts of nature as we find them given in experience. We say: Here is an observed tendency the conditions and antecedents of which are described by science. But what causes the tendency; what is the impelling force? Now to such questions science can give no answer. Science deals with phenomena, and tries to tell us all about their conditions and their antecedents. But whenever Science is asked: “What is the underlying cause of the phenomena,--that which calls them into being?” Science should always give one answer and one only: “Frankly, I do not know; that lies outside my province; ask my sister Metaphysics.” Science ought to have nothing whatever to do with force as the underlying cause of anything in this universe of phenomena. And impulse, as the impelling force which calls a conative tendency into being, is a metaphysical, not a scientific conception. We need not further discuss the psychological nature of impulse. Indeed, the little that has been said would not have been necessary to our inquiry were it not that we frequently have occasion to speak of animals as “creatures of impulse,” and to refer to their behaviour as due to impulse. What do we mean by such expressions? If we regard conative tendency as a fact (whatever may be said for or against its being also a specific experience), and if this fact is the tendency of the conscious situation to develope in certain definite ways, then we may define _impulse_ with sufficient clearness by saying, with Dr. Stout, that it is characterized by being unreflective. Conative tendency thus comprises two categories--impulse and volition; the one unreflective, the other involving deliberation. Before passing on to consider how impulse is partly determined by the feeling-tone and the emotional attributes of the conscious situation, we may first draw attention to the important way in which the results of conative tendency afford the data through which consciousness attains its unity in the midst of diversity of experience.