[ 336 ] 



XLT. Investigation of the Alteration produced by Heat in the 

 Velocity of Propagation of Light in Water. By Richard 



RUHLMANN. 



[Concluded from p. 281.] 



IV. The Measurements. 



THE following scheme was followed in making the observa- 

 tions. 



(1) Determination of the angle of the prism at the tempera- 

 ture of observation. 



(2) Fixing the minimum deviation for the sodium-line. 



(3) Maintenance of the temperature constant and reading of 

 the thermometer. 



(4) Measurement of the angle of least deviation. 



(5) Measurement of the distances of the axis of rotation of 

 the universal circle from the prism and from the slit. 



(6) Determination of the temperature of the place where the 

 observations were made. 



The angular measurements (1) were always got by taking, as 

 the most correct value, the mean of a short sequence of observa- 

 tions which were not extended over too great a thermal range. 



In order to maintain the temperature constant during the 

 progress of observation, the lamp, after warming the liquid and 

 during the determination of the lines, was either entirely with- 

 drawn or its flame was made exceedingly small. While the outer 

 liquid, which was the warmer, was establishing thermal equili- 

 brium with the inner one and radiation had already begun, there 

 were always some seconds during which the temperature in 

 the liquid remained tolerably constant. This was shown by 

 the refracted images of the spectral lines appearing sharply 

 defined and straight. The position of the lines was then quickly 

 determined and the temperature read off 1 on the thermometer; the 

 minimum deflection was always taken for the sodium-line only; 

 and this position remained unchanged within several degrees of 

 thermal variation ; for even considerable deviations from the true 

 minimum had no influence upon the position of the refracted 

 image. Strictly, it would be necessary to calculate the final 

 result according to another formula. As a rule, we have omitted 

 to take this into account here. 



If, namely, for a ray D for example, the minimum of the de- 

 flection is determined, then the following formula serves for cal- 

 culation : — 



n sin J (d+a) 



D sin | 



(A) 



