508 



SCIENCE. 



[N. S. Vol. XXIII. No. 587. 



1849, and subsequently, to cite the best-known 

 researolies, by Oppel (1856), Helmholtz 

 (1866), Dvorak (1870), J. J. Hoppe (18Y9), 

 Thompson (1879), Zehfuss (1880), Bowditch 

 and Hall (1880-2), Budde (1884), Exner 

 (1887, 1888-9, 1899), Heuse (1888), Stern 

 (1894), J. Hoppe (1894), Borschke and Hes- 

 cheles (1902) and Szili (1905). 



The theoretical explanations of these phe- 

 nomena may be roughly divided into those 

 based upon psychical processes and those based 

 upon physiological processes set up in the 

 visual apparatus. Zollner explained the phe- 

 nomena that had been described by Plateau 

 and Oppel, as he did his own well-known illu- 

 sion, as a purely psychical falsification of 

 judgment. Budde, somewhat similarly, at- 

 tributed them to false interpretation of cor- 

 rectly reported sense-impressions. The phys- 

 iological explanations, to follow Szili, may be 

 subdivided into three groups, according as the 

 essential basis of the illusion is found (a) in 

 eye-movement, (b) in the 'tailing-off' of the 

 retinal excitation or (c) in specific retinal 

 after-images of movement. The first position 

 is represented by Purkinje and, moi'e notably, 

 by Helmholtz ; the second by Johannes Miiller, 

 W. Stern and Wundt; the third, with vari- 

 ous modifications, by Plateau, Dvorak, Mach, 

 Zehfuss, J. Hoppe, Exner and Szili. 



With the general conclusions of the last- 

 named group of investigators I do not wish to 

 quarrel, but I have noted indications that, 

 under certain conditions, there may be an 

 illusion of movement of apparently ' central,' 

 if not strictly psychical, origin. As my ob- 

 servations were but incidental to other work, 

 and as I have had no opportunity to make ex- 

 tended experiments, I can but report them, 

 with apologies for their incompleteness, in the 

 hope that some one may be interested to pur- 

 sue the matter further. 



While experimenting in the Cornell Educa- 

 tional Laboratory upon the transference of 

 habits, Mr. Althaus, one of our graduate stu- 

 dents, had occasion to teach the telegraph 

 alphabet to several observers. To insure uni- 

 formity in ' sending ' the dots and dashes, suit- 

 able perforations were made in a sheet of 

 kymograph paper, which was then placed upon 



the revolving drum : contact between the metal 

 drum and a flexible copper brush operated the 

 sounder at each perforation. These and other 

 conditions obliged the experimenter to watch 

 the passage of the perforations past the tip 

 of the brush for. one to two hours daily dur- 

 ing a period of two months — perhaps forty- 

 five hours in all. 



My experimenter was, at first, ignorant of 

 the illusion of reversed motion, as appeared 

 rather amusingly by the fact that for two 

 weeks he had regularly thrown the brush out 

 of contact whenever he stopped the kymo- 

 graph, under the impression that there was an 

 objective backward movement of the drum 

 that miglit damage the contact point or the 

 paper. Now, shortly after this, he reported 

 that the drum appeared to reverse slightly 

 whenever he glanced at it, without its having 

 been seen just previously in motion — as, for 

 instance, when first entering the laboratory 

 for the afternoon's work. This new illusion 

 became stronger as the experimentation went 

 on, and finally became so persistent that, after 

 six weeks' disuse of the kymograph, it was still 

 strong, and, after eight weeks, still definitely 

 present. 



To test this observation, I placed on the 

 drum a sheet ruled with vertical black stripes 

 4 mm. wide and 15 mm. apart. Close to the 

 drum was placed a fixation-point, for which 

 T found most satisfactory a bit of black card- 

 board 9 mm. square, provided with a small 

 white center 3 mm. square and supported on 

 a slender wire. When the drum (rotating 

 about its vertical axis) was driven at moderate 

 speed — say four revolutions per minute — the 

 usual phenomena of illusory reversal were ob- 

 servable. Furthermore, by nearly continuous 

 observation for forty-five minutes, I was able 

 to secure a slight, but definite, reversed mo- 

 tion just at the moment of glancing at the 

 drum, though after an interval of from ten 

 to fifteen minutes of non-stimulation. But I 

 was not able, in the time at my command, to 

 establish the illusion so that it would persist 

 several hours or days, as was Mr. Althaus with 

 his prolonged tests. Recently, Professor Sea- 

 shore has reported to me a somewhat similar 

 indication of ' central ' factors in an illusion 



