Theory of Mr. Crookes's Radiometer* 395 



There was no perceptible effect that could be attributed to 

 earth-currents. The mean loss of time before the signal on 

 make of current was visible, was 1'34 sec, and that before 

 break of current signal was 1*28 sec, the mean being 1*31 sec, 

 which is, as noticed, remarkably close to Mr. Varley's estimate 

 of 1J sec. The final longitude of Mokattam from Greenwich 

 was 2 h 5 m 6 S, 32 ; the results of the four different nights 

 agreed very closely, the seconds from the observed times and 

 signals uncorrected for personal equation being 6*55, 6*64, 

 6*61, and 6*55, an accordance which exceeded all expec- 

 tation. It would appear from this that the longitude was 

 probably correct within two or three hundredths of a second ; 

 but it is to be borne in mind that, although there is strong 

 evidence of the personal equations of the observers having 

 remained very constant throughout the nights of the experi- 

 ment, the amount of each could only be obtained at Greenwich 

 after the return of the expedition, when slight alterations might 

 have taken place in them. 



From Mokattam the longitudes of the Suez and Thebes 

 stations were easily obtained ; and the work was carried 

 further to the east by Mr. Gill, of Lord Lindsay's party, who, 

 with the Germans, determined the longitude of Suez and 

 Bombay. 



XL VII. A Theory of Mr. Crookes's Radiometer. 

 By Professor Challis, M.A., F.R.S., F.R.A.S* 



I ASSUME that the construction of the radiometer and the 

 phenomena it presents are well known. In the Philoso- 

 phical Magazine for January 1857 I have proposed a " Theory 

 of the Transmutation of Hays of Light," which bears in an 

 essential manner on the explanation of the radiometer I am 

 about to propose ; and I have made additional remarks on the 

 same question in the Philosophical Magazine for May 1865. 

 According to the views adopted in those communications, rays 

 of light incident on opaque substances are partly reflected with- 

 out undergoing transmutation and are partly intromitted. The 

 latter part are transmuted into rays of the kind which by their 

 dynamical action produce heat of temperature as distinct from 

 radiant heat. This is what takes place on the bright side of 

 each revolving vane of the radiometer. We may assume that 

 the same quantity of light is incident on the blackened side as 

 on the other ; but because, as the surface is black, none of the 

 light is regularly reflected, it follows that the whole is trans- 

 muted into the heat of temperature. The accession of tempe- 

 rature is therefore greater on the black than on the bright 

 * Communicated by the Author, 



