SOLAR SURFACE. 



spaces are observed, also covered with strongly defined curved or 

 branching streaks, more intensely luminous than the other parts 

 of the disk, among which spots often break out. These several 

 varieties in the intensity of the brightness of the disk have been 

 differently designated by the terms facules and lucules. These 

 appearances are generally more prevalent and strongly marked 

 near the edges of the disk. 



21. Various attempts have been made to ascertain by the direct 

 test of observation, independently of conjecture or hypothesis, the 

 physical state of the luminous matter which coats the globe of the 

 sun, whether it be solid, liquid, or gaseous. 



That it is not solid is admitted to be proved conclusively by 

 its extraordinary mobility, as indicated by the rapid motion of 

 the edges of the spots in closing ; and it is contended that a fluid 

 capable of moving at the rate of 44 miles per hour cannot be 

 supposed to be liquid, an elastic fluid alone admitting of such a 

 motion. 



Arago has, however, suggested a physical test, by which it 

 appears to be proved that this luminous matter must be gaseous ; 

 in short, that the sun must be invested with an ocean of flame, 

 since flame is nothing more than aeriform fluid in a state of incan- 

 descence. This test proposed is based upon, the properties of 

 polarised light. 



It has been proved that the light emitted from an incandescent 

 body in the liquid or solid state, issuing in directions very oblique 

 to the surface, even when the body emitting it is not smooth or 

 polished, presents evident marks of polarisation, so that such a 

 body, when viewed through a polariscopic telescope, will present 

 two images in complementary colours. But, on the other hand, 

 no signs of polarisation are discoverable, however oblique may be 

 the direction in which the rays are emitted, if the luminous matter 

 be flame. 



The light proceeding from the disk of the sun has been accord- 

 ingly submitted to this test. The rays proceeding from its borders 

 evidently issue in a direction as oblique as possible to the surface, 

 and therefore, under the condition most favourable to polarisation, 

 if the luminous matter were liquid. Nevertheless, the borders of 

 the double image produced by the polariscope show no signs 

 whatever of complementary colours, both being equally white even, 

 at the very edges. 



This test is only applicable to the luminous matter at or near the 

 edge of the disk, because it is from this only that the rays issue 

 with the necessary obliquity. But since the sun revolves on its 

 axis, every part of its surface comes in succession to the edge of 

 the disk ; and thus it follows that the light emanating from every 



109 



