Mr. J. J. Waterston on Solar Radiation, 505 



t constant and a variable the proportionate absorption represented 



Br 

 by -k— is a constant quantity ; but so far as r depends on t, 



the value -5- 1 - increases witli t. and the causal relation may be 

 rom J 



expressed as follows : — 



The heat-pulse travels, carrying with it an intensity that it 

 borrows from the temperature of its source, and encounters a 

 deflecting or absorbing power in passing through a constant 

 element of the atmospheric medium that is exactly proportional 

 to that intensity. 



It would be simpler if the resistance was uniform — if the 

 proportion of force absorbed was constant ; but the observations 

 do not admit of the possibility of this. The curve traced out 

 by the coordinates, r and secant zenith distance, would in that 

 case be no longer the conic hyperbola, but the logarithmic 

 curve*. 



§ 12. At 6 o'clock in the evening of the 31st of July, while 

 making an observation, an extensive shower of thin rain took 

 place overhead and westward towards the sun, without sensibly 

 obscuring its light or affecting its image when examined through 

 a telescope: the value of r descended immediately from 15° to 

 13°. The single observation I took in India, compared with 

 those taken at the same altitude in this country, indicates 

 that the value of r is there fully double what it is here, while 

 the quantity of vapour held in suspension estimated from the 

 dew-point is certainly greater. It would seem probable, there- 

 fore, that the absorbing power of the atmosphere depends on the 

 watery particles contained in it, not upon the aqueous vapour 

 dissolved in it. 



§ 13. Referring to the method of computing the sun's po- 

 tential temperature, described in the l Proceedings * of the Society 

 for March 1860, and employing the same rule with 11, the extra 

 atmospheric value of r equal to 70° at earth's mean distance, 

 we arrive at 12,880,000° as the potential temperature of its 

 radiating surface f. 



If we expose the flame of a bat's-wing jet to one ball of a 

 differential thermometer, the effect is the same whether the 

 broad side or the narrow side of the flame is presented, as I 

 have found on trial. Now the potential temperature being 

 equal to the product of r by the reciprocal of the angular space 

 occupied by the flame, it is in the one case about five times 



* And the points on the Chart ranging in a curve convex towards the 

 axis, and leading by its trend to an inadmissibly low value of R. 



t By a typographical error, x, the potential temperature of the radiating 

 surface of the sun, was represented to be 918,000° instead of 9,180,000°. 



