OF TOTAL DAYLIGHT DURING A SOLAR ECLIPSE. 205 



The observations unfortunately could not be carried on^ 

 owing to cloud, beyond 9^ 23^, at which time the sun's 

 disk was still more than half eclipsed, so that the rise of 

 the curve beyond this point cannot be given. 



If we compare the curve of chemical intensity (which is 

 nearly symmetrical on both sides of the totality) with the 

 curve of solar obscuration, it will be seen that the rate of 

 diminution of the chemical intensity of total daylight du- 

 ring the first portion of the eclipse up to the point at which 

 the disk is half obscured is greater than corresponds to the 

 area of darkened solar disk, whilst from this point up to 

 totality the rate of diminution of chemical action is much 

 less than that of the exposed portion of the disk. 



Determinations were also made from time to time with 

 the arrangement for shading off all the direct sunlight from 

 the sensitive paper; and these, combined with alternate 

 observations of the intensity of total daylight, give the 

 separate intensities of diffused and direct sunlight. Owing 

 to the very rapid changes which occurred constantly in the 

 condition of the solar disk from passing clouds, only a few 

 isolated sets of these double observations, made when the 

 disk was free from cloud, are of value. 



These observations, however, are sufficient to indicate 

 that the rate of diminution and increase of intensity of the 

 chemically active rays in the direct sunlight is proportional 

 to the changes of area in the exposed portion of the sun's 

 surface. 



Chemical Intensities. 



Apparent Fraction of 



solar time. Total. Diffuse. Direct, disk visible. 



7*^57™ 0*948 0788 o-i6o 0-94 



8 29 o'i83 0*146 0-037 o"43 



8 45 o*ii3 0*090 0*023 o'l? 



9 o 0*000 O'OOO O'OOO O'OO 



9 16 0'152 0-124 0*028 0-2I 



From these numbers it appears that the differences ob- 



