I dealt especially with geometrical illusions of direction such as Poggendorff's, Zollner's, Wundt's etc. and developed my theory, referring to experimental data obtained under various conditions and setting it against other theories. Here in this abstract I should like to summarize our theory. In our mental life we can find various tendencies. Those, which are particularly of importance to our problem, are following. They do not exist separately, but are different phases or aspects of our concrete mental phenomena. (A) Our organic and mental tendency for unity. The tendency of something phenomenal to be apprehended “einheitlich” may be thought one aspect-the phenomenal aspect-of it. B) The tendencies to develop into and from stability or equilibrium, which may be regarded as two phases of “einheitliches Umstrukturieren.” (C) The tendency of emphasis, accentuation or reinforcement and that of neglect, unaccentuation or inhibition. In terms of relation or difference the former may be expressed as differentiation, dissimilation or contrast and the latter as assimilation (Angleichung). They may be thought also two phases or aspects of “einheitliches Umstrukturieren.” The concepts, differentiation and assimilation, are regarded by me as implying referring of something whole, central or predominant to its parts or periphery, neither vice versa nor as bit to bit relations. Our theory, assuming differentiation and assimilation in this sense as fundamental conditions of the illusions of direction, opposes to those theories which interpret the illusions mainly either by some peripheral physiological facts (as eye movement theory, Lehmann's irradiation theory etc.) or by some mere part-to-part relations (as association theory), those theories being discarded experimentally. Some peripheral facts (e. g. certain peripheral limit of our discrimination sensibility, retinal irradiation, etc.) may act in the sense of assimilation or neglect of difference and others (e. g. certain discrimination sensibility of our sense organs, retinal contrast etc.) may act for differentiation or emphasis of difference, and some of them seem to play some part in the illusions of direction, but it is shown experimentally that they afford no constant and essential condition for the illusions. They can be just accessory conditions of them. On the mental or central side we have many facts acting for assimilation (e. g. more or less larger value of our differential threshold than its physiological limitation, our usual neglect of known differences, our content with round number etc.) and many other facts for differentiation (e.g. our occasional nervousness to small differences, various kinds of mental contrast etc.). But taken apart from other conditions, they are quite arbitrary. I think that the principle of assimilation and differentiation is one of fundamental conditions, but no differential or special condition of the illusions. (D) The tendency of vertical and horizontal orientation, briefly rectangular co-ordinates tendency. This R. C. tendency is founded deeply both inside and outside of us and both in our nature and in our nurture, just as our symmetry tendency with which it has the inmost relation. The tendency may be regarded primarily as a manifestation of the general tendency of an organism for equilibrium and is conditioned physically by the direction of gravity and the hereditary symmetrical structure of the organism. One seeks for some horizontal level to live on; when obliged to live on the slope, one levels the ground or builds the floor horizontal. If one has to stand or sit on the slope, one keeps both sides of the upper body symmetrical by bending one leg or by other means.