This article reports descriptions of the logic of constructing the Munsell color solid as a coherent psychological system, the history of its improvement by a committee of the Optical Society of America, and findings obtained by the method of multidimensional scaling (MDS) in a series of studies. When multiattribute color differences that subjects can naturally perceive are matched with lightness differences between two grays, Munsell colors can be embedded in a three-dimensional Euclidean space with the following reproducibility. Interpoint distances of the constructed configuration are proportional to scaled color differences, and the root mean squares of discrepancies between these were of the order of .23 in the Munsell V unit, (i.e., only about three times of just-noticeable difference of lightness). Three major deviations of the result from the Munsell notation, embedding colors in some non-Euclidean spaces, and psychological implications of spatial representation of color are discussed.

Visual space (VS) is a coherent self-organized dynamic complex that is structured into objects, backgrounds, and the self. As a concrete example of geometrical properties in VS, experimental results on parallel and (equi) distance alleys in a frameless VS were reviewed, and Luneburg's interpretation on the discrepancy between these 2 alleys was sketched with emphasis on the 2 hypotheses involved: VS is a Riemannian space of constant curvature (RCC) and the a priori assumed correspondence between VS and the physical space in which stimulus points are presented. Dissociating these 2 assumptions, the author tried to see to what extent the global structure of VS under natural conditions is in accordance with the hypothesis of RCC and to make explicit the logic underlying RCC. Several open questions about the geometry of VS per se have been enumerated.

When a subject (S) was asked to list all the items belonging to a specified category, cumulative totals of items produced exhibit a smooth curve as a function of time. The equation proposed by Bousfield and Sedgewick in 1944 was fitted to single sequences of individual 5s, and it was suggested that either the S scans at a constant rate and scans exhaustively or that the density of relevant items under scanning decreases linearly. When the retrieval was repeated in immediate succession, the functional form of the sequences changed and two models were proposed: one assuming operation of two parallel scanning channels and the other assuming the change of retrieving rate to an exponential form.

Subjective impression of numerosity of dots randomly scattered in an area of 19.5 X 24.5 em, 36"'1,010 in number (+), was scaled by four different methods with five subjects. The absolute identification showed that the number of dots is underestimated if +>10. Each of two scales based upon ratio judgments is a power function of +with the exponent of 0.87, and each of two scales based upon difference judgments is also a power function of +when the origin of the scale is appropriately defined. The relationship between the four scales and comparison with scales in other sense modalities were discussed from the viewpoint of scaling methodology.