308 On the Existence of Carbon in Suns Atmosphere. [Apr. 11, 



all tangential stresses on the surface without affecting the normal 

 stresses ; and, in the second place to set up currents sweeping over the 

 surfaces of solid bodies, thus completely destroying the simplicity of 

 our first solution of the problem. 



14. When external forces, such as gravity, act on the gas, and when 

 the thermal phenomena produce differences of density in different 

 parts of the vessel, then the well-known convection currents are set 

 up. These also interfere with the simplicity of the problem and intro- 

 duce very complicated effects. All that we know is that the rarer the 

 gas and the smaller the vessel, the less is the velocity of the convection 

 currents ; so that in Mr. Crookes' experiments they play a very small 

 part. 



IV. " Note on the Existence of Carbon in the Coronal Atmo- 

 sphere of the Sim." By J. N. Logkyer, F.R.S. Received 

 March 20, 1878. 



It is now four years since I obtained evidence, which seemed to mo 

 conclusive, as to the existence of carbon in the sun's atmosphere. 

 There were two points, however, which remained to be settled before 

 the matter could be considered to be placed beyond all doubt. 



The first was to establish that the fluted bands generally present in 

 the spectrum of the arc, as photographed, which bands vary very con- 

 siderably in strength according to the volatility of the metal under 



o 



experiment, were really bands of carbon— a point denied by Angstrom 

 and Thalen. 



This point is settled by the photographs submitted to the Society with 

 this communication. In these the carbon bands remain the same, 

 though one spectrum is that of carbon in air, the other of carbon in 

 chlorine dried with great care, and the proof that it cannot be the 

 spectrum of a combination of carbon with oxygen lies in the fact that 

 in the chlorine it is more brilliant than in the oxygen. Now, assum- 

 ing the chlorine to have been but imperfectly dried, this would not 

 have happened if a compound of oxygen had been in question. 



The next point was to obtain evidence that there was absolutely no 

 shift in the carbon bands, which sometimes happens when the part of 

 the arc photographed is not perfectly in the prolongation of the axis of 

 the collimator. 



A photograph has been obtained which supplies such evidence. 

 There are metallic lines close to the carbon bands, which are pro- 

 longations of Fraunhofer lines, while the lines which I have already 

 mapped at W. L., 39*27 and 3.9*295, in the spectrum of iron, are also 

 absolute prolongations. Therefore there is no shift in the carbon 

 flutings. Now the individual lines in the brightest portion of the 



