230 Absorption-Spectra of Oxygen and its Compounds. [June 6, 



was no difference between the spectra of liquid and vapour except in the 

 red, and even here the difference was less marked than at lower tem- 

 peratures. At 130° no distinction was observable between the spectra 

 of liquid and vapour, there were no lines or bands in either, but a good 

 deal of general absorption. Liquid and vapour were dark, and ap- 

 peared much of a colour, but the meniscus at the junction was quite 

 evident. The tube was further heated to 155°, but no further change 

 was noticed in the spectrum. On gradually cooling the tube, at 112° 

 the least refrangible band in the orange was seen coming in both in 

 vapour and liquid, diffuse in both. At 100° the usual lines were well 

 seen in the orange, yellow, and citron of the vapour, faint lines in the 

 green, and none in the blue ; and subsequently the appearances pre- 

 sented on heating followed in the reverse order. 



A solution of N 2 4 in carbon bisulphide gave, in a thickness of 

 7 or 8 cm., diffuse absorption-bands in the green and citron, ill-defined 

 as in liquid N 2 4 and in corresponding positions. In a thickness of 

 1 cm. these bands were no longer visible. 



These observations bear out the supposition that pure N 2 4 is 

 without selective absorption of the visible rays, and that the absorp- 

 tion observed is due to NO 2 , both in the vapour and liquid, this 

 absorption being modified in the liquid by the state of solution in 

 which the molecules have much less freedom. As the temperature 

 rises the proportion of NO 2 increases, and at the same time the density 

 of the vapour increases and the freedom of motion of the molecules is 

 diminished, they are less able to assume the more rapid vibrations, 

 and those which they do assume become less sharply defined, so that 

 the lines fade into bands and ultimately into a general absorption. 



Taking Willard Gribbs's expression (" Equilibrium of heterogeneous 

 substances," ' Connecticut Acad. Trans.,' vol. 3, p. 239) for the density 

 D, in terms of the pressure in atmospheres p, at temperature t° 



D = 3-178 + - ^{0(3-178 + 0)}, 

 where log lo = 9-47056 - - log 10 p, 



as deduced from Deville and Troost's experiments, we find the density 

 of NO 2 at 140° and 50 atmospheres, equal to 2, i.e., equal to the 

 density of N 2 4 vapour at 60° and 1 atmosphere. 



Dewar and Ansdell found the critical temperature for N 2 4 to be 

 156°. 



