18 Mayer — Phenomena of /Simultaneous Contrast-Color 



<4t- 



\^7 



Rg.15. 



of the disk. Yarious methods were tried to render reliable 

 the comparison of the illumination of the two sides. 



(1) I used the eye alone, regarding only the portions of the 

 ring on the border near the mirror, as shown in fig. 14. 



(2) Two prisms reflected the images of 

 portions of the sides of the ring nearest the 

 mirror to juxtaposition, as shown in fig. 15. 



(3) A tube blackened on the inside, with a 

 black screen on the end next the disk, as 

 shown in fig. 16, was used. The eye was 

 thus shielded from extraneous light and the 

 comparisons of illuminations was made, as in 

 experiments 1 and 2, on the portions of the 

 rings which were nearest the mirror as shown 

 in fig. 16. Yision through this screened tube 

 gave the best results. 



After practice in such comparisons, made during several 

 hours on different days, I became more and more skillful and 

 the results of measurements become more concordant; but 

 such methods of photometry do not ap- 

 proach the accuracy of those in which 

 two contiguous surfaces of different de- 

 grees of translucency coalesce into one 

 surface of a uniform illumination, as 

 happens on the balance of illuminations 

 on the two sides of a Bunsen photometer 

 disk when these are illuminated by lights 

 of the same intensity and of exactly the 

 same hue. A photometer for this mode 

 of observation is described in the follow- 

 ing section. It has, however, the advan- 

 tage over the Bunsen photometer in that 

 it serves to measure the intensities of differently colored lights. 

 The Rotating-Disk Photometer. — The photometer disk was 

 taken apart and a ring of thin white linen paper* of the 

 diameter of the disk and T 9 Q- cm wide was laid on one of the 

 disks ; this was covered by the circle of thin translucent white 

 paper and on this was laid another ring of the thin linen paper. 

 The disks were now clamped together. The outer portions 

 of the open sectors of the disk were thus closed by two thick- 

 nesses of the thin linen paper with the " alba tracing-paper " 

 between them, a, in fig. 17 ; while the inner portions of the 

 sectors were closed by the tracing-paper alone, b, in H^. 17. 



On rotating the disk, it was not possible to balance the 

 colors of the outer half (a in fig. 17), of the ring formed of 



*The best paper I have experimented with for this purpose is water-marked 

 " Crane & Co., Dalton, Mass., Bond, No. 21." 



Fig. 1 6 . 



