THE PERCEPTION OF TIME. 617 



in both directions, errors increase their size.* This time 

 varies from one observer to another, but its average is re- 

 markably constant, as the following table shows. f 



The times, noted by the ear, and the average indiffer- 

 ence-points (given in seconds) were, for — 



WundtJ 0.72 



Kollert§ 0.75 



Estel (probably) 0.75 



Mehner 0.71 



Stevens « 0.71 



Mach f 0.35 



Buccola (about)** 0.40 



The odd thing about these figures is the recurrence they 

 show in so many men of about three fourths of a second, 



* Curious discrepancies exist between the German and the American ob- 

 servers with respect to the direction of the error below and above the point 

 of indifference— differences perhaps due to the fatigue involved in the 

 American method. The Germans lengthened intervals below it and short- 

 ened those above. With seven Americans experimented on by Stevens 

 this was exactly reversed. Tlie German method was to passively listen to 

 the intervals, then judge ; the American was to reproduce them activel}' 

 by movements of the hand. In Mehner's experiments there was found a 

 second indifference-point at about 5 seconds, beyond which times were 

 judged again too long. Glass, whose work on the subject is the latest 

 (Philos. Studien, iv. 423), found (when corrections were allowed for) that 

 all times except 0.8 sec. were estimated too short. He found a series of 

 points of greatest relative accuracy (viz , at 1.5, 2.5, 3.75, 5, 6.25, etc., 

 seconds respectively, and (thought that his observations roughly corrobo- 

 rated Weber's law. As 'maximum' and 'minimum 'are printed iuter- 

 chnncrpably in Glass's article it is hard to follow. 



t With Vierordt and his pupils the indifference point lay as high as 

 from 1.5 sec to 4.9 sec, according to the observer (cf. Der Zeit.sinn, 1868, 

 p. 112j. In most of these experiments the time heard was actively repro- 

 duced, after a short pause, by movements of the hand, which w^ere re- 

 corded. Wundt gives good reasons (Physiol. P.sych., ii. 289, 290) for re- 

 jecting Vierordt's figures as erroneous. Vierordt's book, it should be said, 

 is full of important matter, nevertheless. 



I Physiol. Psych., ii. 286. 290. 

 § Philosophische Studien, i. 86. 



II Mind, XI. 400. 

 TT Loe. cit. p. 144. 



** Op. cit. p. 376. Mach's and Buccola's figures, it will be observed, 

 are about OTie 7«a^/ of the rest — sub-multiples, therefore. It ought to be 

 observed, however, that Buccola's figure has little value, his observations 

 not being well fitted to .show this particular point. 



