184 Witson— On the Temperature of certain Stars. 
paper patch on part of the slit of the spectroscope. We thus get in the photograph 
the band of continuous absorption due to the paper patch; and where the dark lines 
cross it, they are both widened and darkened. This, I think, clearly shows that 
there may be causes which account for the widening of the lines in sun-spot 
spectra other than the ‘cool downrush.” 
The Sun’s photosphere is usually considered to consist of a layer of clouds 
formed of minute, solid particles of carbon,* or, possibly, calcium carbide. We 
know of no other element but carbon that could possibly remain solid at solar tem- 
peratures; and Prof. Hale quite lately has been able to observe the bright fluted 
spectrum of carbon by placing the slit of his spectroscope tangentially as close as 
possible to the solar image formed by the great Yerkes refractor. 
Now, is it not conceivable that the temperature of the photospheric clouds is 
not very far from that at which carbon would be volatilized, and that a sun-spot 
is a local upheaval of the intensely hot gases which we know must exist below the 
photosphere where the clouds were removed? We-should then only be receiving 
radiation from the gaseous layers which lie below the photospheric clouds; and, as 
these would be a much poorer source of radiation than the solid carbon particles, 
the spot would seem dark. Thus, I maintain that a sun-spot is a local region in 
which the temperature is so high that no substance can there remain in the solid 
form. Now, two arguments have been used against this supposition. First, it 
has been urged that if the temperature rose sufficiently to dissolve the clouds, they 
would simply re-form again ata greater altitude where the temperature was suffi- 
ciently low, and that therefore it would be quite impossible for a star to get rid 
of its photosphere by a rise of temperature. If this argument were pushed 
logically, then every element in the Sun should rise in the Sun’s atmosphere until 
it got a temperature sufficiently low, and then form a cloud stratum. . 
That these different cloud-strata do not exist is, I think, sufficient evidence 
that carbon also, with a suitable rise of temperature, would be unable to form a 
cloud-layer. Secondly, it has been urged that, if a sun-spot be a hole in which 
we can see the deeper gaseous layers, the radiation coming from them—as they 
are thick enough to be opaque—ought to be quite as great as that coming from 
the photospheric clouds, and further, that the spot would not seem any darker 
than the photosphere. 
The late Professor FitzGerald suggested, as a way out of this difficulty, that, in 
a layer of heated gases such as we have in the Sun, there must be powerful con- 
vection-currents, and that a quantity of the light coming from the lower layers 
would be reflected from the surfaces of these currents, and be returned back into 
the Sun. So that a layer of heated gas, no matter of what thickness, could never 
* G. Johnstone Stoney, Roy. Soc. Proc., vol. xvii., 1868, pp. 1-57. 
