1867.] 



Constitution of the Sun and Stars. 



29 



therefore become abnormally thin. If this process is not carried so far as 

 to put a stop to the incessant rain beneath the clouds, their increased 

 transparency will give rise to a facula when the phenomenon takes place 

 on a large scale, and to the coarsely mottled appearance of the photosphere 

 where it presents itself in smaller patches. Hence we see why a facula 

 retains its brightness up to the margin of the sun's disk, a phenomenon 

 which is inconsistent with the usually received hypothesis that the grada- 

 tion of light on the sun's disk is due to the absorption of the outer atmo- 

 sphere. If the rain also cease we have the penumbra of a spot ; if the cloud 

 itself is dissolved away, we have its umbra. 



The dark body which is disclosed in the umbra) and penumbrjje of spots 

 must be either an untarnished ocean of some highly reflecting opake sub- 

 stance, or a cloud of some transparent material which scatters light abun° 

 dantly. Both hypotheses are fully considered. 



To most of the foregoing conclusions relating to the photosphere and 

 the adjoining parts we may safely accord the probabiHty 3, 



S. 



Equatorial zone of calms and descending currents. 



Northern zone of variable winds produced by currents about to ascend. 



Zone of variable winds produced by descending currents. 



We have strong reasons for suspecting that the luminous clouds consist, 

 like nearly all the sources of artificial Hght, of minutely divided carbon ; and 

 that the cloiids themselves lie at a very short distance above the situation 



