382 Intelligence and Miscellaneous Articles. 



Accordingly, in measuring the reflecting-power of the grey disks 

 by the old method, the standard white and black disks were com- 

 bined with a small amount of an orange-yellow disk, which made 

 their mixture match the tint of the grey disk under experiment. 

 The luminosity of the orange-yellow disk in terms of standard 

 white cardboard being known, the final result was calculated with 

 equal facility and the annoyance removed. It is hardly necessary 

 to add that in all cases the amount of white light reflected from 

 the standard black disk was taken into account. 



An experiment was now made to test the flicker process by using 

 it to determine the reflecting-power of two grey disks which were 

 afterwards measured in the ordinary way, and the following result 

 was obtained : — 



Direct determination 34-1 29*1 



Kicker „ 33*7 28-6 



Difference *4 °5 



These determinations were not elaborate but may suffice for the 

 purpose. 



Afterwards a more careful set of observations was made to 

 ascertain with what minimum difference in luminosity the flicker 

 became insensible ; a series of five light-grey disks was made with 

 gradations so small that these would have required elaborate work 

 for their individual determination, which was not attempted directly, 

 the method of measuring the least and the most luminous disk 

 being employed, and the difference between the figures divided by 

 four. When these disks were tested flicker fashion in pairs, it was 

 found that the intervals between them were not at all identical, 

 one being larger and its neighbour smaller than the average interval, 

 but each pair nevertheless furnished a faint but distinct flicker. 

 If the intervals had been regular, as was intended, the experiments 

 would have proved that with a difference of J^- part of the total 

 light a flicker could be perceived ; as it was it showed that a con- 

 siderably smaller difference suffices, probably about j-^. 



It is evident that for the purpose of measuring the reflecting- 

 power of coloured disks it will not in every case be necessary to 

 undergo the labour of making an elaborate set of grey disks : if 

 the experimenter is furnished with six or eight brilliantly-coloured 

 disks, the reflecting-power of which has been carefully determined, 

 then, with these and black and white disks, he will be able to match 

 new disks and measure their luminosity in the old way. 



There does not seem to be any reason why this process with the 

 aid of well-known optical devices should not be applied to ordinary 

 photometric work, or to such as is carried on in the interior of the 

 observing-telescope of a spectroscope. In simple work with disks 

 like that here presented, the flicker photometer is necessarily on 

 the model of a staircase, and the transitions, though small, are 

 always abrupt ; but when there are other modes of regulating the 

 illumination this would not be the case, and the advantage of mul- 



