4i2 EXPERIMENTAL INVESTIGATIONS. 



though it was in a room with a northern aspect, on which the 

 sun never shone. In twelve days the water under the glass 

 vessel had lost 6*3 grains, that under the porcelain 5*4, show- 

 ing a difference of nearly a sixth part in the evaporation in 

 favour of the tube exposed to light. I mention this experi- 

 ment here as showing a probability that the liberation of 

 vapour or gas may be accelerated by light, as M. Donny's 

 remarkable experiments seem to show that evaporation is a 

 surface action ; and the effect of light on polarised plates may 

 be somewhat of the same nature. I will not, however, enter 

 into theories, which must be necessarily vague, on such a novel 

 subject, but conclude by giving a table of some of the most 

 trustworthy results which I have obtained when the sunlight 

 has been most steady, in order to show the extent of deflec- 

 tions and differences with coloured light : they were all made 

 between the hours of 10 a.m. and 12, and on the finest days I 

 could select. My first experiments were made in London, in 

 the months of June and July. I was then absent for some 

 time on circuit ; and I have resumed and continued them 

 during my vacation in the months of August and September. 



June 21, apparatus having been arranged on June 12. 

 Solution, water with a few drops of sulphuric acid. 



Deflection by sunlight 8, i.e. from 12 the deflection by 

 polarisation, to 20. 



Exposed platinum positive. 



July 4. Same solution, platinum having been changed on 

 the 24th. 



Deflection by sunlight of 9, i.e. from i to 10. 

 Exposed platinum positive. 



August 24. Solution, water with sulphuric acid. 



Deflection by polarisation . i 



Sunlight .*..... 5 



Blue glass 3 



Yellow glass ...... 2 



Red glass 2 



Exposed platinum positive. 



