622 



SCIENCE. 



[N. S. Vol. XVI. No. 407. 



gase§, and as these are slightly soluble they 

 dissolve to some extent in the fluid. The 

 gases in solution can be taken out by dis- 

 tillation or by exhausting the water, and 

 if we compare their volume with the vol- 

 ume of the water as steam, we should find 

 about one volume of air in 60,000 volumes 

 of steam. This wo^^ld then be about the 

 rough proportion of the relatively perma- 

 nent gas to condensable gas which existed 

 in the ease of the vaporized ocean. Now let 

 us assume the surface of the earth gradu- 

 ally cooled to some 200 degrees below the 

 freezing-point; then, after all the present 

 ocean was frozen, and the climate became 

 three times more intense than any arctic 

 frost, a new ocean of liquid air would 

 appear, covering the entire surface of the 

 frozen globe about thirty -five feet deep. 

 "We may now apply the same reasoning to 

 the liquid air ocean that we formerly did 

 to the water one, and this would lead us 

 to anticipate that it might contain in solu- 

 tion some gases that may be far less con- 

 densable than the chief constituents of the 

 fluid. In order to separate them we must 

 imitate the method of taking the gases out 

 of water. Assume a sample of liquid air 

 cooled to the low temperature that can be 

 reached by its own evaporation, connected 

 by a pipe to a condenser cooled in liquid 

 hydrogen; then any volatile gases present 

 in solution will distil over with the first 

 portions of the air, and can be pumped off, 

 being uncondensable at the temperature of 

 the condenser. In this way, a gas mixture, 

 containing, of the known gases, free hydro- 

 gen, helium and neon, has been separated 

 from liquid air. It is interesting to note 

 in passing that the relative volatilities of 

 water and oxygen are in the same ratio as 

 those of liquid air and hydrogen, so that 

 the analogy between the ocean of water and 

 that of liquid air has another, suggestive 

 parallel. The total uncondensable gaa 

 separated in this way amounts to about one 



fifty-thousandth of the volume of the air, 

 which is about the same proportion as the 

 air dissolved in water. That free hydro- 

 gen exists in air in small amount is con- 

 clusively proved, but the actual proportion 

 found by the process is very much smaller 

 than Gautier has estimated by the combus- 

 tion method. The recent experiments of 

 Lord Rayleigh show that Gautier, who es- 

 timated the hydrogen present as one five- 

 thousandth, has in some way produced 

 more hydrogen than he can manage to ex- 

 tract from pure air by a repetition of the 

 same process. The spectroscopic examina- 

 tion of these gases throws new light upon 

 the question of the aurora and the nature 

 of the upper air. On passing electric dis- 

 charges through the tubes containing the 

 most volatile of the atmospheric gases, 

 they glow with a bright orange light, which 

 is especially marked at the negative pole. 

 The spectroscope shows that this light con- 

 sists, in the visible part of the spectrum, 

 chiefly of a succession of strong rays in the 

 red, orange and yellow, attributed to hy- 

 drogen, helium and neon. Besides these, 

 a vast number of rays, generally less 

 brilliant, are distributed through the 

 whole length of the visible spectrum. The 

 greater part of these rays are of, as yet, 

 unknown origin. The violet and ultra- 

 violet part of the spectrum rivals in 

 strength that of the red and yellow rays. 

 As these gases probably include some of 

 the gases that pervade interplanetary 

 space, search was made for the prominent 

 nebular, coronal and auroral lines. No 

 definite lines agreeing with the nebular 

 spectrum could be found, but many lines 

 occurred closely coincident with the cor- 

 onal and auroral spectrum. But before 

 discussing the spectroscopic problem it will 

 be necessary to consider the nature and 

 condition of the upper air. 



According to the old law of Dalton, sup- 

 ported by the modern dynamical theory of 



