proceedings: philosophical society of WASHINGTON 141 



lines in the spectra of twenty of the chemical elements. Nearly 80 

 per cent of the total number of lines remained unidentified, and many 

 of these, in addition to the well-known absorption bands A and B, due 

 to oxygen in the earth's atmosphere, were suspected to be of terrestrial 

 origin. An opportunity to separate the solar from the telluric lines 

 by aid of the Doppler-Fizeau displacement suffered by lines of solar 

 origin came with an invitation to use the Porter spectrograph at Alle- 

 ghany Observatory. Light from the eastern and western limbs of the 

 sun was photographed simultaneously in two parallel spectra in which 

 all lines of solar origin appear displaced in opposite directions, while 

 those due to absorption in the earth's atmosphere have the same wave- 

 lengths in both spectra. The spectrograms show that about 75 per 

 cent of the absorption lines with wave-lengths between 6400 A and 9400 A 

 arise in the atmosphere of our earth. 



A line (wave-length 7664 A) characteristic of potassium is under 

 ordinary circumstances exactly coincident with one of the lines in the 

 A band due to terrestrial oxygen, but was compelled to betray its solar 

 origin by means of the Doppler-Fizeau effect. The presence of free 

 oxygen in the solar atmosphere was demonstrated by the coincidence 

 of six solar lines with lines in the emission spectrum of oxygen. All 

 of the evidence for solar oxygen lies in its infra-red spectrum. 



Discussion: This paper was discussed by Messrs. Sosman and 

 Humphreys. 



The second paper on The hot spell of August, 1918, was presented by 

 Mr. A. J. Henry. 



The hot spell of the first decade of August, 191 8, had its origin over 

 Montana and the central and northern Plains States. It was prac- 

 tically stationar}^ over eastern Kansas and western Montana from the 

 ist to the 3d, inclusive, and on the 4th spread eastward by way of the 

 Ohio valley to western Pennsylvania and also northeastward into the 

 southern part of the Lake region. On the 5th it reached its greatest 

 geographical extension; the area affected on the evening of that date 

 was a little more than a million square miles, or just about one-third 

 of the area of the LTnited States, excluding Alaska and outlying pos- 

 sessions. The increase in the area of the heated territory from the 4th 

 to the 5th was in round numbers a quarter of a million square miles. 

 The peak of the high temperature in the west was reached on August 

 3-5 and in the east on August 6-7. In the east the daily maxima declined 

 irregularly until the 14th, when normal conditions were reached. The 

 period of extraordinarily high temperature was short, not exceeding 

 three days at any one place. The high temperatures were associated 

 with a dry atmosphere and there was therefore not much bodily dis- 

 comfort and but few heat prostrations were reported in the daily 

 press. 



The abnormally high temperature was due to a combination of favor- 

 able circumstances. These in the order of their importance are: (i) 

 A pressure distribution that inaugurated and maintained a system of 

 southerly winds over the great interior valleys and the middle Atlantic 



