86 REPOKT OF THE SECRETARY. 



basis logarithmic plots « prepared from the observed results showed great departures 

 from straight lines for all wave lengths, so that either the absorbing layer is not so 

 thin as first assumed, or the phenomenon is not a simple one of absorption. Acting 

 on the former supposition, it was found that when the absorption was assumed to 

 take place in a homogeneous stratum of about 45,000 miles thickness outside the 

 photosphere, or in other words, a rather thick stratum, yet well within the thick- 

 ness of the brighter part of the solar corona, very little curvature was exhibited in 

 the logarithmic plots at any wave lengths, or for any distance less than 99 per cent of 

 a radius from the center of the disc, or on either day of observation. Evidently the 

 assumption of homogeneity is a very strained one, hardly perhaps to be tolerated 

 even in a preliminary computation. I have, however, given the transmission coeffi- 

 cients found in this way in Table 3. Estimates have been prepared of the amount 

 by which the total radiation of the sun would be increased were this envelope 

 wholly removed, but these are so far dependent upon the assumption referred to that 

 I defer their presentation for the present. 



I wish, in closing, to particularly commend the zeal and ability with which Mr. F. 

 E. Fowle, jr., has aided in carrying through all the work above reported. Mr. Fowle 

 has also published '' during the year a valuable study of the absorption of water 

 vapor in the earth's atmosphere, wliich summarizes measurements and computations 

 he has made from holographic data collected here in the last three years. He finds 

 the transmission in each of the water-vapor bands studied to follow Bouguer's expo- 

 nential law, but with coefficients of transmission varying greatly in amount for dif- 

 ferent bands and for different wave lengths in the same band. In passing from band 

 to band the absorption is generally greater the greater the wave length of the band, 

 but taking each band by itself the shorter wave lengths are most absorbed. 



CONCLUSION. 



The work of the year has been distinguished by useful improvements in the appa- 

 ratus of the Observatory, notably in the bolometer, the pyrheliometer, and the great 

 horizontal telescope. But chiefly it is marked by a great advance toward what you 

 have set as the far-off final goal of our efforts, namely, tlie establishment of a sound 

 basis for long range forecasting of climate, in the study of the solar radiation, and its 

 absorption in our atmosphere. The work of the year has made it seem possible, 

 though not yet certain, that fluctuations of tlie amount of solar radiation as great as 

 10 per cent may occur, and that independently of them there are marked changes of 

 the transparency of our atmosphere. Both causes must profoundly influence cli- 

 matic conditions, and both are being studied here with increasing facility, ac(!uracy, 

 and success, by the aid of the spectro-bolometer. In short, it now seems not improb- 

 able that successful general forecasting of climate will be the not too far distant 

 outcome of our work. 



Respectfully submitted. 



C. G. Abbot, Aid in Charye. 



Mr. S. P. Langley, 



Secretary of the Smithsonian Institution. 



« Compare Plate VI of last year's report. 



^Smithsonian Miscellaneous Collections (Quarterly Issue), Vol. \\, Part 1. 



