Joty—A Theory of Sun-Spots. 699 
of the formation of a vast area of liquid matter. It is to be 
expected that the depth in or beneath the photosphere at which 
this phenomenon will occur will vary, according to the circum- 
stances of cooling. There is every evidence to show that matter 
thus separated out into the liquid state, and exhibiting the 
phenomenon of surface tension and a true liquid surface, will 
differ but little in density from the surrounding gas from which 
it is derived. Nor is there any physical difficulty in supposing 
the precipitated liquid to float upon denser gas beneath. 
The formation of even large spots is sometimes a matter of a 
score of hours only, and their disappearance is often as sudden : 
facts consistent with a simultaneous change of state over a large 
area. 
They are less luminous than the surrounding photosphere. If 
the liquid matter were opaque—a character we may ascribe with 
safety to liquid carbon, to the liquefied metallic elements, or to 
mixtures of these—a great reduction in emissivity would certainly 
result. 
Violently agitated, probably, this ocean would reflect near its 
margin the surrounding banks of photospheric matter, and, further- 
more, the inrush over the cooler area of liquid would tear down 
these gaseous and cloudy masses on to the surface of the ocean, 
phenomena producing the intermediate brightness as well as the 
jagged and often cyclonic appearance of the penumbra. The 
facts that the limits of penumbra and umbra often follow with 
rude parallelism the meeting of penumbra and photosphere is 
consistent with this explanation. 
The presence of in-falling cooler gas, as determined by the 
spectroscope, into the Sun-Spot would similarly follow as a conse- 
quence of the fall of temperature over the surface of these solar. 
oceans. 
The duration of the spot will depend on the rate at which heat 
is transferred from beneath into the liquid, as well as upon the 
rate at which the liquid is fed from above by in-falling rain. In 
considering the rate at which heat will be transferred into the liquid, 
it must be remembered that the liquid is under physical conditions 
resembling those which preserve the drop of water on the red-hot 
silver plate, in fact isin the spheroidal state. And, again, the rate 
