H. Draper— Oxygen in the Sun. 269 
bearing cannot at present be distinctly seen. In the case of 
nitrogen the broad bands between G and H exhibit, under the 
most intense incandescence, a tendency to condense into narrow 
bands or lines, and indeed there are some sharp lines of nitro- 
gen in the photographs now presented. 
It does not follow, therefore, that the bright bands of oxygen 
are necessarily the brightest parts of the solar spectrum. Other 
substances may produce lines or bands of greater brilliancy. 
There is also another cause for a difference of appearance in 
a bright-line spectrum produced in a laboratory and bright 
| lines in the Sun. While the edges of a band in the Ho spec- 
trum may be nebulous or shaded off, the corresponding band 
in the solar spectrum may have its edges sharpened by the 
action of adjacent dark lines due to one or another of the metal- 
lic substances in the Sun. 
On the whole, it does not seem improper for me to take the 
ground that, having shown by photographs that the bright 
lines of the oxygen spark spectrum all fall opposite bright por- 
tions of the solar spectrum, I have established the probabilit 
of the existence of oxygen in the Sun. Causes that can mod- 
| ify in some measure the character of the bright bands of the 
\ solar spectrum obviously exist in the Sun, and these, it may be 
inferred, exert influence enough to account for such minor dif- 
| ferences as may be detected. 
In closing, it may be well to give some idea of the amount of 
labor and time this research has already consumed, and this 
cannot be better done than by a statement of the production of 
electrical action that has been necessary. Each photograph 
demands an exposure of 15 minutes, and, with preparation and 
onde teat at least half an hour is needed. The making of 
) 
the Gramme has made 20 millions of revolutions. The petro- 
leum engine only consumes a couple of drops of 
> stroke, and yet it has used up about 150 gallons. Each drop 
. of oil produces two or three 10-inch sparks. It must also be 
borne in mind that comparison spectra can only be made when 
the Sun is shining, and clouds therefore are a fertile source of 
Oss of time. 
APPENDIX. 
[We find in the Astronomical Register a Report. of the Dis- 
cussion which followed the reading of Dr. Draper's paper. 
this is the expression of the opinion of the best English author- 
ities upon the conclusions reached in the memoir and as it will 
be'seen, we suppose, by but few of our readers, we present it 
’ in-pretty full Ren Fh as a matter of general interest.—Eps. } 
