426 • PROCEEDINGS OF THE AMERICAN ACADEMY 



to measure the intensity of the light at regular intervals of time as 

 the sun receded from the portion of sky from which the light was 

 received. 



By allowing the mirror to remain fixed during an entire series of 

 observations, absolute uniformity in the angle of incidence of the light 

 on the mirror was obtained, and thus the percentage of the light re- 

 flected by the miror rendered constant, and exact measurements of the 

 angular distance from the sun could be easily made by simply noting 

 tlie lajise of time. 



One result of this method is that all the observations are made east 

 of the sun ; the part of the sky from which the light is received being 

 necessarily on the sun's path. A reverse series can be readily ob- 

 tained on the west side of the sun, by adjusting the mirror so as to 

 receive light from a point at a convenient distance west of the sun, on 

 the path of the same, and then making observations at regular inter- 

 vals as the sun approaches the point. 



The light was received from a circular sky area, 2° 20' in diameter, 

 or 4.25 square degrees ; and the proportion of the light lost by reflec- 

 tion from the mirror, and transmission through the lens, was about .40 

 of the whole. The following table embraces the results of four series 

 of observations. Column one gives the angiilar distance from the sun. 

 The second, third, fourth, and fifth columns give the intensity of the 

 light received from one square degree of sky ; the unit of intensity 

 beinsc the liiiht of a standard candle at a distance of one metre. Series 



I. was made Jan. 20, 1875, between the hours of 2.45 and 3.47 p.m.^ 

 bemnnins 15° 45' west of the sun. 1 he declination of the sun at this 

 time was 20° ; and, since the angles given in the table for this series 

 were not measured on a great circle, they should be reduced in the 

 proportion of the radius to the cosine of the declination. Series II., 

 III., and IV. were made east of the sua between 12 M. and 1 p.m. ; the 



II. on March 23, the III. and IV. on March 27. 



On Jan. 20, the sky was hazy, had a whitish -gray aspect, and 

 reflected much light ; on the 23d of March it was clear and blue ; and 

 on the 27th very clear, reflecting but little liglit. The great difference 

 in the intensity of the light on these different days is well shown by 

 comparing columns two, three, four, and five of the table. 



The meteorological importance of such observations as these is 

 suggested by the fact that it is commonly believed that a deep blue 

 sky, reflecting little light, indicates the presence of a large amount of 

 the vapor of water in the atmosphere, and the probable approach of 

 rain, and that a very clear night frequently precedes a rainy day. If 



