Researches on Solar Physics. 185 
atmosphere. It is worthy of remark that the temperature of 
this atmosphere must be lower than that of the photosphere; 
otherwise the absorption which it occasions would be counter- 
balanced by its radiation. _ 
(2.) The beautiful discovery of Kirchhoff leads to the same 
conclusion, since, in order to account for the dark lines of the 
solar spectrum, it is necessary to suppose the existence of a solar 
atmosphere of a lower temperature than the source of light. 
3.) The red flames which are visible during a total eclipse, 
and which have been proved to belong to the sun (Art. 9), indi- 
eate the existence of a solar atmosphere extending in some in- 
stances as far as 72,000 miles above the photosphere, This is 
confirmed by the nature of the light which these flames emit. 
r. De la Rue has found that this light is very rich in actinic 
protuberance which was not visible to the eye. Now it is pre- 
cisely this description of light which characterizes the electric 
discharge in which gaseous matter appears in a highly heated 
State 
20. Let us now endeavor to show the nature of those correc- 
tions which are rendered necessary by solar refraction. 
(1.) A solar atmosphere will make the sun’s photosphere to 
appear larger than it really is; but the angular distance between | 
WO points, each near the center of the visible disk, will not be 
appreciably altered. This will introduce a slight error into the 
calculated position of any point, since in such a calculation we 
make use of the sun’s apparent angular diameter, which is» 
greater than his true diameter. : 
(2.) Apart from this, an error will be introduced into the cal- 
culation of the solar latitude and longitude of a point, this error 
epending upon its position in the visible disk, and being greater 
for those points which are at a distance from the center. 
$ VI. Questions to be answered in the present paper. 
21. In the present paper we shall attempt to answer the fol- 
lowing questions :— 
(I.) Is the pmbra of a spot nearer the sun’s center than its 
penumbra? or, in other words, is it at a lower level? 
: (IL) Is the photosphere of our luminary to be viewed as com- 
posed of heavy solid, or heavy liquid matter? or is it rather of 
the nature of a cloud? A short explanation will render evident 
terials from those of the liquid plane, or it may contain some of 
the materials of the plane in a state of vapor. Our own ocean is 
