No more than the nature of the energy-transformations, and the essential morphology, does the behaviour of animals afford us the means of setting up absolute distinctions between group and group. Really tropistic behaviour is exhibited by the movements of the stems, roots, and leaves of green plants, or in the movements of Bacteria, and perhaps some unicellular animals. Typically instinctive behaviour is exhibited by the individuals of societies of Insects and by many solitary-living animals belonging to this class; and typically intelligent behaviour is exhibited by the acting of the higher Mammalia. Yet there is undoubtedly much that is truly instinctive in the behaviour of Man, and something of the same nature as his intelligence seems to inhere in the instinctively-acting mammal or insect: how else could an instinctive action become capable of improvement? We cannot doubt that intelligence is manifested by a dog or by much that we see in the behaviour of ants. No rigid distinctions between tropisms, such as we have mentioned above, and the reflexes that may be taken to constitute instinctive behaviour, can be established. Minute analysis, such as that carried out by Jennings on the swimming movements of the Protozoa, leaves us quite in doubt as to how these modes of behaviour are most properly to be described; and all the controversy as to the nature of tropisms, reflexes, instinct, and intelligence surely indicates that these modes of behaviour have something that is common to all of them, and that no clear and certain distinction can be said to separate one from the other. Even those psychic processes which we call intellectual do not seem to be different in kind from some that we attribute to the lower animals: the Protozoan _Paramœcium_ studied by Jennings, or the crabs, crayfishes, and starfishes studied by Yerkes and others really _learn_ to perform actions, but this learning is said to be the result of a process of “trial and error.” The animal tries one series of movements and finds that it fails, tries another and another with a similar result, and in the end finds one that is effective. This is remembered, and when the same problem again confronts the animal it is solved after fewer trials, and finally, after experience, the end-result is attained at once without previous trials. Now many of what we call truly intellectual processes are certainly processes of precisely this nature. Hypothesis after hypothesis _occurs_ to the scientific man (or to the detective, or to the engineer confronted with some exceptional difficulty), and one after another is tested by actual trial, or by a process of reasoning (which is really the rapid and formal resuming of previous experience), until a hypothesis verifiable, or _a priori_ verifiable, is found. What, for instance, are our mathematical methods of integrating a function, or working a long division sum, but methods of scientific “guessing,” and verification of the hypotheses so made?