Naturalism takes refuge in the doctrine of association, when it does not attain anything with its first claims, and applies this theory in such a way that it seems possible from this standpoint to interpret mental processes as having an approximate resemblance to mechanically and mathematically calculable phenomena. As in physics the molecules and atoms, so here the smallest mental elements, the simplest units of feeling are sought for, and from their relations of attraction and repulsion, their groupings and movements, it is supposed that the whole mental world may be constructed up to its highest contents, will, ideals, and development of character. But even the analogy, the model which is followed, and the fact that a model is followed at all, show that this method is uncritical and not unprejudiced. What reason is there for regarding occurrences in the realm of physics as a _norm_ for the psychical? Why should one not rather start from the peculiar and very striking differences between the two, from the primary and fundamental fact, not indeed capable of explanation, but all the more worthy of attention on that account, that there is an absolute difference between physical occurrences and mental behaviour, between physical and mental causality? These most primitive and simplest mental elements which are supposed to float and have their being within the mind as in a kind of spiritual ether are not atoms at all, but deeds, actions, performances. The laws of the association of ideas are not the laws of a mental chemistry, but laws of mental behaviour; very fixed and reliable laws, but still having to do with modes of behaviour. Their separating and uniting, their relations to one another, their grouping into unities, their “syntheses,” are not automatic permutations and combinations, but express the _activity_ of a thinking intelligence. Not even the simplest actual synthesis comes about of itself, as psychologists have shown by a neat illustration. [Illustration: Square _a_2, next to smaller square _b_2. Above them are horizontal lines _a_ and _b_, the same lengths as the widths of the squares below them. Caption: _a_ and _b_ only associated. Squares of _a_ and _b_ in juxtaposition.] [Illustration: Square _c_2. Above it is horizontal line _c_, the same length as the width of the square below it. Caption: _a_ and _b_ really synthetised to _c_. Square of _a_ + _b_ as a true unity = _c_2.]