426 PHILOSOPHICAL PERIODICALS. is more than thrice as great as its diameter at the zenith. Experiments with discs, undertaken with the view of testing theoretical interpretations of this phenomenon, lead to the conclusion that distance is the controlling factor. Of two objects seen under the same visual angle, the more remote appears the greater. And we take the sun at the horizon to be more remote, because we see it projected upon the apparent surface of the sky, and this surface is more remote at the horizon than at the zenith. Ex- periments leave no doubt as to the watch-glass shape of the sky ; height stands to horizontal radius, on the average, as 1 : 3'5. Finally, the ex- planation of this flattening must be sought in the presence of the atmo- sphere : the author works out his theory by comparing the air to fog or to a series of apposed glass plates. The clouded sky follows (or adapts itself to) the contour of the cloudless, not vice versa.'] E. Wiersma. ' Die Ebbinghaussche Combination s-inethode.' [Test of Ebbinghaus' method of mutilated texts, with three groups of school children,- the schools being so chosen that the factors of age, sex, mental endowment, practice, fatigue could in large measure be isolated for separate study. Elaborate report of results : we can here merely note that Ebbinghaus' own results are entirely confirmed. For psychopathological studies, the author emphasises the recommendation that the texts employed be not too easy.] Literaturbericht. Heft 4. F. Schumann. ' Beitrage zur Analyse der Gesichtswahrnehmungen. iii. Der Successivvergleich.' [(1) The first section is occupied with a detailed proof of the fact (already well established by American investigators) that the presence of a memory image is not necessary for a judgment of comparison. (2) What, then, happens in such a judgment ? The author emphasises four points : the ' cutting off ' of a part-magnitude, the expansion and contraction of atten- tion and the effect of absolute impression. He attempts an ontogenetic interpretation of the last-named factor. (3) The residual influence of the first perception cannot be accounted for by ' unconscious ' mental pro- cesses or by cortical physiological processes. It is, perhaps, to be referred to subcortical centres : observations of Helmholtz and Bering bear out this conjecture : at the same time, the theory will not work in all cases. Ebbinghaus' eye-movement theory is also inadequate. On the whole, then, the ' cutting off ' is to be ascribed to attention ; while the effect of absolute impression must be assigned to residual factors identical with those that condition the revival of ideas, and standing in close relation to the concept. The second part of the paper is taken up with an applica- tion of this theory of successive comparison to a large number of the standard optical illusions. The evidence offered is mainly that of intro- spective detail, and cannot well be summarised. The fact that many of the illusions lessen or disappear with continued observation is regarded by the author as in so far a proof of their foundation in judgment rather than in perception. Experiments with rectangles bring out the subjec- tive accentuation of the relatively longer side. Pure contrast illusions are referable to ' Einstellung ' or 'innere Anpassung '. The Mliller-Lyer (arrow-head and feather) illusion is carefully analysed, and "Wundt's interpretation rejected.] H. Ebbinghaus. ' Ein neuer Apparat zur Kontrolle des Chronoskops.' [A fairly cheap instrument, of great accu- racy, embodying essentially the principle of the original Hipp fall- apparatus.] Literaturbericht. Heft 5 und 9. F. Schumann. ' Beitrage zur Analyse der Gesichtswahrnehmungen : Schluss.' [The author con- cludes his discussion with detailed criticisms of Lipps' theory of the play of mechanical forces and of Witasek's theory of perceptual illusion. He polemises successfully against Lipps' doctrine of the memory image (Vorstellunysbild and attacks Meinong's law that "judgments which
