OF ARTS AND SCIENCES. 247 



Intrinsic Brilliancy 861,000 st. 



These measurements were made with the Bunsen photometer, and 

 were all observed from the same side of the disc as the sun. Judging 

 from my measurement of the moon, I had supposed that it would be 

 easier to make my observations from this side, but I was not pre- 

 pared for the great difference exhibited. From the side of the sun 

 the spot disappeared nearly as perfectly as when measuring a gas 

 flame, particularly if the line of sight was nearly perpendicular to the 

 disc, and the eye was thrown out of focus for it. From the side of 

 the gas, the appearance was that of a bright 3'ellow spot ou a bright 

 blue background ; and the comparison was almost impossible. The 

 varying brilliancy of different parts of the sun's disc was very marked. 

 I took, as usual, the brightest portion, namely, the centre. 



In order to determine the amount of light lost by the porte lumiere, 

 a reflecting photometer was planned and constructed. A somewhat 

 lengthy series of observations showed that the light lost with the 

 best plate-glass mirrors, 3 nuns, in thickness, varied from about 17 to 

 2-4 per cent; depending on the angle made by the incident and reflected 

 rays. I believe no wholly satisfactory results have yet been attained, 

 and the measurement has been attempted only once or twice. My 

 results are represented in Fig. 3. The abscissae represent the angle 

 of the incident and reflected rays. The left-hand ordinates give the 

 per cent of light reflected, the right-hand ones the per cent lost. 



On the Measurement of High Temperatures by the Spectroscope and 



Photometer. 



It is a fact of common experience, that as we heat a body to higher 

 and higher temperatures, it becomes brighter and at the same time 

 whiter, — in other words, more violet light is given off. Here, then, 

 we have a means of determining qualitatively the temperature of any 

 source. Now if we only knew by what law, either the intrinsic 

 brilliancy, or the violet rays increased with the temperature, and 

 knew at the same time the melting points of some of the metals, we 

 should be able to form some idea of the temperatures, not only of the 

 lime, electric, and magnesium lights, but also of the sun and fixed 

 stars. 



Three attempts have been made to determine the temperature of 

 the sun ; one by Secchi, supposing the temperature proportional to 

 the radiation of heat ; the second founded on Newton's law of cool- 

 ing; the third dependent upon a numerical exponent, determiued from 



