158 KEPURT— 1892. 



From the fact that Q' undergoes a regular increase from the red end 

 of the spectrum to the blue, and from the approximate equality of the 

 lowest line of numbers, it may be inferred — 



(1) That the optical effect of electric stress in a given medium (CSj) 

 increases as the wave-length diminishes. 



(2) And more definitely, that the optical effect varies exactly or very 

 nearly in the inverse ratio of the square root of wave-length. 



Eighth Report of the Committee, consisting of Sir Gr. G. Stokes 

 (Chairman), Professor A. Schuster, Mr. Gr. Johnstone Stonet,. 

 Sir H. E. KoscoE, Captain W. de W. Abney, Mr. Gr. M. Whipple, 

 Professor McLeod, and Mr. Gr. J. Stmons, appointed for con- 

 sidering the best methods of recording the direct Intensity of 

 Solar Radiation. {Draxvn up by Sir Gr. Gr. Stokes.) 



The work of the Committee during the past year has been confined to an 

 examination, both experimental and theoretical, of Balfour Stewart's 

 second actinometer when used as a dynamical instrument. Actinometers 

 may be divided into two classes, which may conveniently be denominated 

 dynamical and statical respectively. In those of the dynamical class the 

 mercury or other fluid employed is examined while the head of the 

 column is in motion, in consequence of the exposure being varied by 

 suitable manipulation, and readings of the column are taken at chosen 

 times, or else the times are noted when the top of the column reaches 

 chosen readings. In the statical class the instrument is allowed to attain 

 its permanent state, subject of course to a secular change, such as that 

 due to the varying altitude of the sun, and the results are deduced from 

 the stationary readings of two or more thermometers. Herschel's, 

 Hodgkinson's, and Stewart's first actinometers are examples of the 

 dynamical kind ; the black bulb thermometer, Violle's actinometer, and 

 Stewart's second actinometer, when used as he intended, are examples of 

 the statical class. 



Stewart's second actinometer has been already described (see Reports 

 of the Association for 188G, p. 63, and 1887, p. 32), but to save the 

 trouble of reference it may be well briefly to mention that it consists of 

 an envelope of thick copper, closed on all sides except as regards a small 

 hole to allow the sun's rays to enter, and one to allow the stem of the 

 central thermometer to pass through. In the actual instrument the 

 envelope is cubical, and its temperature is determined by three thermo- 

 meters, with their bulbs sunk in hollow chambers in the thick metal, two 

 (a, b) in the front face of the cube, or that turned towards the sun, the 

 third (c) in the back face. The internal thermometer (d) has a lenti- 

 cular bulb, which is mounted so that it lies at the centre of the cube, or 

 nearly so, and has its plane perpendicular to the incident rays. Th& 

 copper cube is surrounded by a thick coating of felt, and this by a cover- 

 ing of thin brass. The object of this arrangement is to make the tem- 

 perature of the copper cube sensibly the same all round, and at the same 

 time to prevent it from changing more than very slowly when the indtru- 

 ment is exposed. The thermometers a, b, c were graduated to degrees, 

 D to half-degrees, all Fahrenheit. 



