298 THE ADDRESSES, LECTURES, ETC., OF 



withdrawn, and its roof be exposed to active radiation into space, 

 what should we observe through a strong pane of glass inserted 

 in the side of the boiler near the liquid surface, lit up by an 

 incandescent electric lamp within ? The loss of heat by radia- 

 tion from the boiler would give rise to convection currents, and 

 partial condensation of the vapour atmosphere ; then, if the 

 motion of the water was made visible by means of colouring 

 matter, we should observe convection currents in the fluid mass 

 separate and distinct from those in the gaseous mass ; but these 

 convection currents would cause no visible disturbance of the 

 liquid surface, which would present itself to the eye with the 

 smoothness of a mirror. It is only in the event of the steam 

 pressure being suddenly relieved at any point on the surface 

 that a portion of the water would flash into steam, causing a 

 violent upheaval of the liquid. 



The dark spots on the sun appear to indicate commotion of 

 this description, but these are evidently not the result of mere 

 convection currents ; if they were, they would occur indis- 

 criminately over the entire surface of the sun, whereas telescopic 

 observation has revealed the fact that they do occur almost 

 exclusively in two belts, between the equator and the polar sur- 

 faces on either side. Their occurrence could be satisfactorily 

 explained if we could suppose the existence of strong lateral 

 currents flowing from the polar surfaces towards the equator, 

 which lateral currents in the solar atmosphere would cause 

 cyclones or vortex action with a lower and denser atmosphere 

 consisting probably of metallic vapours ; this vortex action ex- 

 tending downward, would relieve the fluid ocean locally from 

 pressure, and give rise to explosive outbursts of enormous mag- 

 nitude, projecting the lower atmosphere high above the photo- 

 sphere, with a velocity measured, according to Lockyer, by a 

 thousand miles a second. It will be seen from what follows how, 

 according to my views, such vortex action in those intermediate 

 regions of the sun would necessarily be produced. 



But supposing that, notwithstanding the difficulties just pointed 

 out, convection currents sufficed to effect a transfer of internal 

 heat to the surface with sufficient rapidity to account for the 

 enormous surface-loss by radiation, we should only have the poor 

 satisfaction of knowing that the available store, would last longer 



