164 Prof. O. Reynolds on the [Jan. 6, 



sions. On balancing this deflection, it was fonnd that, to bring the needle 

 to zero, it was necessary to diminish the slide-resistance by 400 millims. 

 Thus the diminution produced in the resistance by exposure to the light 

 of the paraffin-lamp was 3--J-3- part of the whole resistance of the tellurium. 



On exposing the selenium bar used in my experiments to the direct 

 rays of the same paraffin-lamp at the distance of 1 metre, the resistance 

 of the selenium was diminished by one fifth of its whole resistance. 



Prom the above experiment we see that at the distance of half a metre 

 (that is, with light of four times the intensity) the change of resistance 

 in the tellurium under the same conditions is only -ji/oif P ar ^ °^ ^ s 

 whole resistance. 



On exposing the selenium to a constant source of light at dif- 

 ferent distances, the change in the resistance of the selenium on ex- 

 posure for 10 seconds (as measured by the swing of the galvanometer- 

 needle) is almost exactly inversely as the distance, i. e. directly as the 

 square root of the illuminating power. This law is true whether the 

 source of light be 1 candle or an Argand lamp whose illuminating power 

 is equal to 16 candles. 



Taking the mean of a number of experiments, all of which agreed 

 pretty well together, the deflections at the several distances were : — 



At J metre. At ^ metre. At 1 metre. At 2 metres. 



With Argand lamp 170 83 39 



„ candle 41 18 8 



„ candle 82 39 18 8 



Another series of experiments with a candle and Argand lamp 

 (when the illuminating power of the lamp was equal to 12 candles), both 

 at the distance of 1 metre, gave the following results :— 



With the candle the deflection was 19 in 10 seconds. 

 „ Argand lamp „ 66 „ 



The ratio of the deflections is very nearly 1 to 3 J. 



These experiments clearly show that the change in the resistance of 

 the selenium is directly as the square root of the illuminating power. 



[Correction. — In my former paper on this subject (Proc. Eoy. Soc. vol. sxiii. no. 163), 

 on page 536, line 15, omit the word "opposing," and line 21, for "which opposes" 

 read " in the same direction as ; " also on page 539, line 25, for " which opposes a " 

 read " in the same direction as the"; to the end of line 26 add " but in the opposite 

 direction."] 



III. " On the Refraction of Sound by the Atmosphere." By Prof. 

 Osborne Reynolds, Owens College, Manchester. Communi- 

 cated by Prof. Stokes, Sec. R.S. Received November 22, 1875. 



(Abstract.) 

 This paper maybe said to consist of two divisions. The first contains 

 an account of some experiments and observations undertaken with a view 



