12 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 65 



Second objection. — In view of the great importance attached by 

 Mr. Very and others to the observation of solar radiation at great 

 air-masses, we reexamined some of our observations of former years 

 which were made at larger than the usual air-masses. For each of 

 the days we give in the preceding table ratios of atmospheric trans- 

 mission coefficients found for different air-mass ranges at many 

 points in the spectrum, first, as obtained by comparing results 

 found at small air-masses with those found at large ones, and, 

 second, b)' comparing those heretofore published with those now 

 obtained at large air-masses. For the determination of transmission 

 at large air-masses, the observations were replotted, using Bem- 

 porad's air-mass tables instead of the secant of the zenith distance. 

 The new plots did not include the observations at small air-masses, 

 thus avoiding any prejudice of the observer which might have been 

 caused by seeing them. The results of the comparison appear in the 

 preceding table. It cannot be said that this indicates any consider- 

 able fall of transparency as the air-mass decreases. Had this been 

 the case the ratios given would in general have been greater than 

 unity. The slight tendency in that direction is hardly beyond the 

 error of determination, and, besides, is to be attributed to the depart- 

 ure of Bemporad's air-masses from secant Z values used in our 

 publications heretofore. 



OBSERVATIONS OF SEPTEMBER 20 AND SEPTEMBER 21, 1914 



For a more thorough test we selected two of the driest and clearest 

 days on which we have ever observed on Mt. Wilson, namely, 

 September 20 and September 21, 1914, for combined spectro-bolo- 

 metric and pyrheliometric measurements, extending from the moment 

 the sun rose above the horizon 1 until the close of our usual observing 

 period at about 10 o'clock in the forenoon. During this interval we 

 obtained on the first day 11 and on the second day 12 holographs of 

 the spectrum, extending from wave length 0.34/1, to wave length 

 2.44 p., and we made 33 pyrheliometric determinations of the solar 

 radiation on the first day, and 34 such determinations on the second 

 day. We observed the barometric pressure by means of a recording 

 Richard barograph, and we observed the humidity of the air by 

 means of a ventilated Assmann psychrometer. 



The following tables include the barometric, hygrometric, and 

 pyrheliometric data : 



1 We computed the apparent zenith distance of the lower limb of the sun at 

 the instant of the start of the first holograph on September 20 to be 88° 20'. 

 The apparent zenith distance of the mountain horizon at that point is 88° 28'. 



