28 ROYAL SOCIETY OF CANADA 



67. The relative (initial) periodicities for different colours are 

 'best seen from the curves platted on Plate V, in which assumed wave- 

 Jiemgths (obtained by averaging the data of several authors for colour 

 and corresponding wave-lengths) are the abscissae, and initial period- 

 icities of the respective colours, are ordinates. Curve (1) represents 

 the periodicities obtained from Fig. 2, Plate II, and Figs. 1 and 2, 

 Plate III, and curve (2) the periodicities obtained from the curves on 

 Plate IV. 



68. These two curves are very similar in form and would almost 

 coincide if all the ordinates of one were increased, or all those of the 

 other were decreased, by a certain constant amount. This non- 

 coincidence is probably due either to variations of the chronograph or 

 to physiological changes, or both combined. (One week elapsed from 

 the end of the first series to the beginning of the second). 



69. It should be mentioned here that if the reader wishes to 

 compare results obtained with stop-watch and chronograph, respectively, 

 he should bear in mind that the initial periodicities in the former case 

 are usually larger than in the latter, because, with the stop-watch 10 

 oscillations were counted, with the chronograph only 5, and as the 

 periodicity rapidly increases during the act of " fixing," this will suffi- 

 ciently account for any apparent discrepancies — it must also be remem- 

 bered that physiological changes, also, slightly affect the periodicity. 



70. The chief results of the foregoing experiments (pertaining 

 to periodicity) may be briefly summarized as follows : — 



Periodicity changes almost certainly with intensity of impinging 

 light, the act of " fixing " and the physiological condition of the eyes — 

 and probably with the colour of simple and mixed impinging light (after 

 the manner of curves, Plate V, and curves. Fig. 3, Plate III). The 

 periodicity is large in darkness, much smaller in light, and larger in 

 red light than in blue light. The periodicity changes with shade of 

 coloured light (produced by change of sensation, intensity of light 

 remaining constant). Periodicity unaffected by heart-beat (within 

 limits of 50 to 130). 



71. It will have been noted in the curves on Plates I-IV that, 

 the counting of the more refrangible rays commenced sooner than the 

 counting of the less refrangible rays ; this was due to the fact that the 

 oscillations are more distinct vsdth the I B and G- B glasses than with 

 the Y and E glasses, and can therefore be seen more quickly. To see 

 what influence this had upon the results a series was taken with E, CI. 

 and I B, in which the counting was begun at practically the same 

 time (3-1 sees.) after "fixation," in each case, however, althougli the 

 initial periodicities were somewhat closer together than before, the usual 

 order was preserved (viz.: I B -484; G -492; E -502). 



