Recent Observations upon Solar Physics. 453 



range of cool atmosphere which extends outwards, in which the 

 protuberances manifest themselves, and which are visible by bor- 

 rowed light, three only of the known constituents of the sun 

 (hydrogen, sodium, and magnesium) appear to be able to main- 

 tain a permanent footing in sensible quantities, although traces 

 of the more volatile of the others* must no doubt also spread to 

 nearly their full natural height, which is perhaps not far outside 

 the gauze-like envelope of cloud. 



106. If, now, a tempest sweeping through the sun's atmosphere 

 hurry a portion of the vapours bounded by the shell of clouds 

 into a much higher and cooler situation, they will there act the 

 part described in § 68, and the cyclone beneath and the flaming 

 column above will come into existence. 



107. The discovery of bright lines in the spectra of solar 

 flames is valuable, too, from its suggesting to us a possible, and, 

 indeed, what is now the probable, cause of the four bright hy- 

 drogen lines in 7 Cassiopeia?, and the several other stars of which 

 it is a type. If there be in these stars a shell of attenuated 



high temperature absorbs oxygen, which is disengaged again as the tempe- 

 rature falls. Can it be that the hydrogen and sodium which are constantly 

 streaming downwards unite in some such way with other metallic vapours, 

 and that the compounds as they diffuse upwards are decomposed in equal 

 quantitities — the other metals to be precipitated as a mist of solid or liquid 

 particles, the hydrogen and sodium to be set free as gases? 



This hypothetical account of the limited breadth of the sodium and hy- 

 drogen lines is perhaps too much of a guess to have much prospect of being 

 true ; but I mention it here chiefly because it appears in some degree sus- 

 ceptible of verification. For if the atmospheres of hydrogen and sodium 

 come thus to an abrupt end in a situation at which the temperature sud- 

 denly increases, we should expect to find them emit visible bright lines 

 where they are heated by contact with the metallic atmospheres, though 

 possibly much fainter lines than those of protuberances. Captain Haig's 

 observation, that the lines of the prominences were lost when the spectrum 

 widened out, is, so far as it goes, an answer in the negative ; so are also the 

 observations of spots in which no bright lines have yet been seen ; but these 

 answers do not seem decisive, and it would perhaps be worth endeavouring 

 to ascertain at the eclipse of next year whether faint lines may not be seen 

 in a slit of light from the shell of clouds kept separate from the light of 

 regions further down. 



* Such as copper and zinc. Dr. Miller in his ' Chemistry ' records a 

 curious instance of the volatility of copper at low temperatures. A beam 

 which for many years was suspended over a furnace in a copper-smelting 

 house in Norway contained minute beads of metallic copper studded through 

 its texture : the copper must have been raised in vapour and so deposited 

 within its fibres. Gold has been found similarly studding the beams of 

 refineries. (Miller's ' Chemistry,' third edition, § 523). May we not fur- 

 ther presume that those metals which, when rubbed, emit a smell are in 

 some slight degree volatile in cold hydrogen? Rubbing the metal seems to 

 remove a film of oxide and allow the moisture of the air to oxidize the fresh, 

 surface. The hydrogen which is then set free is probably what carries the 

 taint of metal with it. 



