306 Royal Society :— Mr, G. J. Stoney on the Physical 



sensibly less dark in the blue, and to lines which emit a still more con- 

 siderable amount of light in the green, yellow, orange, and red, 

 those colours in which a body gradually heated begins to glow. 



A detailed scrutiny of the lines emitted by the various gases 

 leads to several interesting results. Hydrogen and iron are the two 

 most abundant constituents of the sun's outer atmosphere, and play 

 in it the same part which nitrogen and oxygen do in the earth's. 

 There is but the merest trace of sodium present. The other gases 

 are met with in intermediate quantities. Again, barium cannot have 

 a vapour-density so high as would appear at first from its atomic 

 weight, and therefore probably belongs to the same class of elements 

 as cadmium and mercury, which have vapour-densities half of what 

 correspond to their atomic weights. To these several results we may 

 attribute the probability 3. 



The photosphere consists of two strata which may be distin- 

 guished. The outer of these is shown to be cloud in the ordinary 

 sense of the word — that is, solid or liquid matter in a state of minute 

 division, and denser than the part of the atmosphere in which it is 

 dispersed (probability 3). This cloud is precipitated from its vapour 

 by the chill produced by its own abundant radiation towards the 

 sky, a chill which constitutes the shell of clouds a surface of mini- 

 mum temperature considerably cooler than either the layer above 

 it or the layer beneath (probability 3). The hotter layer which is 

 outside the luminous clouds seems to have a depth somewhat 

 greater than the length of the earth's radius (probability 2). Just 

 outside it there is a second layer of luminous clouds, but so exces- 

 sively thin that they can be seen only during a total eclipse, on which 

 occasions a portion of them has been seen under the form of two arcs 

 of cloud extending for some distance on either side of the points of 

 first and last contact, where alone a sufficiently low part of the sun's 

 atmosphere was disclosed (probability 3). Above these there soar- 

 other clouds raised by causes which will be referred to further on. 

 ;, About the middle of the hot stratum over the photosphere there is 

 a surface of maximum temperature, outside which the temperature 

 decreases almost continuously to the limit of the iron atmosphere. 

 A little outside this there is a second very feeble maximum, the tem- 

 perature of which falls short of the heat of the flame of a Bunsen's 

 burner ; and outside this, through the immense height which is 

 tenanted by sodium, magnesium, and hydrogen alone, the temperature 

 goes on decreasing till it becomes excessively cold. These results 

 are made out with probabilities 2 and 3. 



Within the luminous clouds the temperature very rapidly waxes, 

 and the density, too, appears to receive a nearly sudden increase. All 

 gases with a vapour-density more than about eighty times that of hy- 

 drogen are imprisoned within the shell of clouds by the comparative 

 chill which there prevails, cooperating with the intensity of the force 

 of gravity exerted by the sun. Between the film of clouds and the 

 stratum immediately beneath there are violent motions of convection, 

 which both carry up fresh vapour to be condensed into cloud, and 

 carry down the cloud into a region where it becomes mist and rain. It 



