248 PROCEEDINGS OF THE AMERICAN ACADEMY 



the experiments of Dulong and Petit. The first two give a tempera- 

 ture of several million degrees, the third about two thousand. I give 

 below the opinions of four well-known modern astronomers, three of 

 them having made the sun their specialty. 



Pere Secchi says, " As to the absolute value of this temperature, 

 we cannot fix it with certainty Nevertheless, this tempera- 

 ture must be several million degrees of our thermometer, and capable 

 of maintaining all known substances in a state of vapor." 



Prof. Newcomb's views : " For the temperature of the photosphere 

 it seems likely that the lower estimates are more nearly right, but 

 the temperature of the interior must be immensely higher." 



Prof. Young's views : " As to the temperature of the sun's sur- 

 face, I have no settled opinion, except that I think it must be much 



higher than that of the carbon points of the electric light The 



estimates dependent on Newton's law seem to me manifestly wrong 

 and exaggerated; on the other hand, the low estimates of the French 

 physicists seem to me hardly more trustworthy." 



Prof. Langley says, " The temperature of the sun is, in my viewj 

 necessarily much greater than that assigned by the numerous phy- 

 sicists, who maintain it to be comparable with that obtainable in the 

 laboratory furnace ; but we cannot assign any upper limit to it, until 

 physics has advanced beyond its present merely empirical rules con- 

 necting emission and temperature." 



Now we know from the experiments of Prof. Draper and others, 

 that as the temperature rises, the light increases ?nuch more rapidly 

 than the heat ; and let us suppose that this law holds good up to the 

 temperature of the sun. Since we do not know any terrestrial high 

 temperature with certainty, great accuracy is manifestly out of the 

 question. Heated bodies begin to give out light at about 500° C ; 

 silver melts at about 1,000° C. Many determinations of the melting 

 point of platinum have been made, which give it in the neighborhood 

 of 2.000° C. The temperature of the electric arc has been estimated 

 at between 3,000° and 4,000° C, — let us say, 3,500°. The intrinsic 

 brilliancy of the carbons of the electric light we found to be 3,141, 

 that of the sun, 30,100. Tins was determined at an altitude of 

 2o°, — let us suppose our atmosphere removed and double it, obtain- 

 ing 72,000. It has been shown by my brother, Prof. Pickering, that 

 only about one fourth the light from the centre of the sun's disc 

 reached the earth. We will therefore multiply its brilliancy by 4, 

 obtaining 288,000. Divide by the intrinsic brilliancy of the electric 

 light (3,141), and we find the sun to be 90 times as brilliant. Then 



