250 



PROCEEDINGS OP THE AMERICAN ACADEMY 



It is perhaps unnecessary to add, that the above-mentioned law 

 of the increase of the violet rays is inapplicable to flames like the 

 blue part of the gas, where no solid matter is introduced. It probably 

 applies in a modified form, to lime flames, as witness the disap- 

 pearance of the blue Hue in the strontium spectrum, at low tem- 

 peratures. 



Second Estimation of the Sun's Temjjerature. 



Below is given a table showing the total and intrinsic brilliancies, 

 as well as the temperatures, of the several sources referred to in 

 this article. 



Let us now construct a curve with the figures of the seventh column 

 as abscissae and the sixth as ordinates. The gas flame, although not 

 properly speaking an incandescent body, may still be used to fix a 

 lower limit to our curve at that point. This curve is represented 

 in Fig. 4. The horizontal line, Su, represents the corrected 

 intrinsic brilliancy of the sun. It will be seen that the curve 

 cannot intersect it to the left of the left-hand dotted line, and 

 is not likely, so far as one can judge from the form of the curve, to 

 cross it to the right of the right-hand one. These would give to the 

 sun limiting temperatures of 7,600° and 50,000° C. The middle 

 dotted line corresponds to the temperature we previously found of 

 22,000° C. 



