different from the rest subordinates the other impressions to it in such a way that they fall together in groups. If the recurrent differ ence is one of intensity, the strongest impression comes first in the group. . . . If the recurrent difference is one of duration, the longest impression comes last. . . . When the impressions are uni form in length and intensity, the mind enforces a grouping by giving fictitious values 'to the impressions, generally with respect to intensity, but sometimes with respect to duration. At the rate .795 sec. the mind intensifies every other sound, so that the series is grouped by 2. . . . The second sound in the group is subordinated to the first. At the rate .460 sec. the mind finds it easy to group a series of auditory impressions by three, by intensifying the first greatly and the second slightly. . . . More than three degrees of intensity do not appear together in the order of their intensities in a series. In grouping by four, which takes place generally at the rate .307 sec., the mind accents the first strongly and the second slightly." Muscular movements accompanying the series are regarded by the writer as the condition, not the result, of the rhythmical grouping. A general principle of rhythm is stated as follows: "The conception of a rhythm demands a perfectly regular sequence of impressions within the limits of about 1.0 sec. and 0.1 sec. A member of the sequence may contain one or more simple impressions. If there are a number of impressions, they may stand in any order of arrange ment, or even in a state of confusion [?], but each member of the sequence must be exactly the same in the arrangement of its elements."
The various forms of the stroboscope; two types of construction. (1) A projection arrangement, similar to that devised by Uchatius. (2) Color-mixture, with an apparatus resembling Purkinje's Phorolyt. (3) Contrast-effects with the stroboscope. (4) Apperceptive completion of defective stroboscopic presentations.
Lalande's explanation of paramnesia is on a level with explanation by metempsychosis. Experiments show 'recognition,' (1) with partial resemblance; and (2) in the case of entirely 'new' presenta-
