128 
ASTRONOMY: ADAMS AND BURWELL 
short duration of ecKpses has necessitated the use of spectroscopes of 
moderate power, and as a consequence the determination of the bright 
line wave-lengths has by no means reached a degree of accuracy compar- 
able with that of the dark Knes of the solar spectrum. Accordingly 
such questions as relate to small differences of wave-length between 
solar and flash spectrum lines, as well as some which depend upon the 
structure of the Hnes have of necessity remained unsettled. 
It was shown by Young, Lockyer and others that the flash spec- 
trum might be observed without an eclipse imder conditions of good 
definition and high transparency, and Young was enabled to observe a 
considerable number of bright lines visually. Through the use of photo- 
graphic methods this work was extended greatly by Hale at the Kenwood 
Observatory and later at the Yerkes Observatory with the 40-inch re- 
fracting telescope. With this instrument he observed visually the 
green carbon fluting in the flash spectrum which lies in a very thin stra- 
tum close to the sun's visible edge. 
The photography of the flash spectrum was one of the investiga- 
tions undertaken with the 60-foot tower telescope of the Mount Wilson 
Observatory and the 30-foot spectrograph used in conjunction with it. 
A preliminary Account of some of the results obtained was published in 
1909 by Hale and Adams. The results of a more detailed study of the 
photographs obtained at that time are indicated in this communication. 
The total number of Unes measured upon the negatives is slightly 
larger than that obtained from eclipse negatives in the same region 
of the spectrum (X 4800 to X 6600). Thus a comparison with Mitchell's 
recent determinations from plates obtained at the eclipse of 1905 shows 
a total of 901 Hnes for the ecHpse photographs and 1027 for those taken 
without an eclipse. The chief interest in such a comparison, how- 
ever, lies in the difference in the lines observed. The ecKpse photo- 
graphs show as strong bright lines nearly all of the strong dark lines 
of the solar spectrum. On plates taken without an eclipse many 
of these lines remain dark, or at most show faint bright fringes on 
either side. The proportion, however, of faint lines of the solar spec- 
trum represented as bright lines in the flash spectrum is much greater 
than on the eclipse photographs. It seems very probable that this 
difference in behavior is due to the difference in the level of the ob- 
servations. The eclipse plates refer to a high level in the solar at- 
mosphere, and, as St. John has shown, the strong lines of the solar 
spectrum originate at a higher effective level than do the fainter lines. 
A comparatively low level for the observations made without an eclipse 
