328 ANNUAL EEPORT SMITHSONIAN INSTITUTION, 1910. 



nights; made bolometric measurements of the water- vapor bands in 

 the infra-red spectrum on one day ; and measured with the bolometer 

 on two days the relative brightness of the sun and many different 

 parts of the sky. My expedition lacked the picturesqueness and eclat 

 which distinguished Langley's, with its private car, its guard of 

 cavalry, and a budding astronomer of the first rank, the renowned 

 Keeler, as assistant. However, I rode from Mojave to Lone Pine 

 (about 150 miles) in an automobile trying all the while desperately 

 to keep my pyrheliometer from being broken, and was consequently 

 jounced myself, once to the roof of the automobile, and barely escaped 

 a broken nose. My treasured pyrheliometer afterwards rolled down 

 the Mount Whitney trail twice with the pack mule, and the second 

 time the mule was killed, but the instrument reached the top in 

 safety. My measurements of 1910 are, of course, not yet reduced. 



Considering that practically identical results have been obtained by 

 simultaneous " solar constant " measurements at sea level (Washing- 

 ton) and 1,800 meters (Mount Wilson), and again at 1,800 meters and 

 4,420 meters (Mount Whitney), observing by the spectro-bolometric 

 method of homogeneous rays in each case, I think we must admit that 

 Langley's second difficulty was a bugbear and not an insuperable 

 obstacle. I therefore venture to announce that I believe the true 

 average value of the " solar constant " of radiation is for the years 

 1905 to 1909, 1.92 calories per square centimeter per minute. We 

 know that the earth's temperature is higher at sun-spot minimum 

 that at maximum. Hence I suppose that the values so far observed 

 are a little below the mean for a term of years, and I propose as the 

 most probable mean value of the " solar constant " 1.95 calories per 

 square centimeter per minute. 



Our results at Mount Wilson have strongly confirmed the impres- 

 sion gained in 1903 that the " solar constant " is really a variable of 

 short and irregular periodicity. We have tested this conclusion by 

 all means in our power. But the one obvious and necessary test, that 

 of establishing a second far-distant cloudless station and carrying on 

 there with equal facilities and experience a series of " solar constant " 

 measurements simultaneous with those on Mount Wilson, we have 

 not been able to make for lack of funds. 



