7§ COSMOS. 



of the Sun, but has hkewise added considerable weight to tha 

 conjectures advanced in reference to the whole physical con 

 stitution of the central body of our planetary system. " A 

 ray of light which reaches our eyes, after traversing many 

 millions of raiiles, from the remotest regions of heaven, an- 

 nounces, as it were of itself, in the polariscope, whether it is 

 reflected or refracted, whether it emanates from a solid, oi 

 fluid, or gaseous body, it announces even the degree of its in 

 tensity. {^Cosmos, vol. i., p. 52, and vol. ii., p. 332.) It is 

 essential to distinguish between natural light, as it emanates 

 directly from the Sun, the fixed stars, or flames of gas, and is 

 polarized by reflection from a glass plate at an angle of 35° 

 25', and that polarized light which is radiated as such from 

 certain substances (as ignited bodies, whether of a solid or 

 liquid nature). The polarized light which emanates from 

 the above-named class of bodies very probably proceeds from 

 their interior. As the light thus emanates from a denser body 

 into the surrounding attenuated atmospheric strata, it is re 

 fracted on the surface, and in this process a part of the re- 

 fracted ray is reflected back to the interior, and is converted 

 by reflection into polarized light, while the other portion ex- 

 hibits the properties of light polarized by refraction. The 

 chromatic polariscope distinguishes the two by the opposite 

 position of the colored complementary images. Arago has 

 shown, by careful experiments extending beyond the year 

 1820, that an ignited solid body (for instance, a red-hot iron 

 ball), or a luminous, fused metal, yield only ordinary light, in 

 rays issuing in a perpendicular direction, while the rays which 

 reach our eyes from the margins, under very small angles, are 

 polarized. When this optical instrument, by which the two 

 kinds of light could be distinguished, was applied to gas flames, 

 tliere was no indication of polarization, however small were 

 the angles at which the rays emanated. If even the light be 

 generated in the interior of gaseous bodies, the length of way 

 does not appear to lessen the number and intensity of the very 

 oblique rays in their passage through the rare media of the 

 gas, nor does their emergence at the surface and their transi- 

 tion into a different medium cause polarization by refraction. 

 Now, since the Sun does not cither exhibit any trace of polar- 

 ization when the light is suflered to reach the polariscope in 

 a very oblique direction, and at small angles from the margin, 

 t follows from this important comparison that the light shin- 

 ing in the Sun can not emanate from the solid solar body, noi 

 from any liquid substance, but must be derived from a ga?>e^ 



