290 On the Absorption- spectrum of Oxygen. 



B still quite sharp, but A somewhat obscured by a general ab- 

 sorption at the red end. At 90 atmospheres this general 

 absorption at the red end seemed to extend to about one third 

 of the distance between A and B ; but A could still be seen, 

 when the slit was wide, as a still darker band on a dark red 

 background ; B was still sharp, and the other absorptions all 

 strengthened and somewhat expanded. The diffuse edges of 

 the several bands now extended from about — (1) X 6410 to 

 6190, (2) X5865 to 5635, (3) X5350 to 5280, (4) X4820 

 to 4710, (5) X4480to4455. 



Photographs taken when the pressure of the oxygen was 

 90 atmospheres show a faint absorption-band about L of the 

 solar spectrum, a stronger band extending from about X3600 

 to 3640, a broad diffuse band about the place of the solar line 

 O, and complete absorption above P. 



The absorbent column in the tube at the highest pressure 

 used contained a mass of oxygen about equal to that in a 

 vertical column of the earth's atmosphere of the same section as 

 the tube ; but the intensity of the bands produced by the 

 compressed gas was far greater than that of the corresponding 

 bands in the solar spectrum with a low sun. When the arc- 

 light was replaced by a piece of white paper reflecting light 

 from the sky through the tube, it appeared to the naked eye 

 to have a faint blue tint, similar to that of liquid oxygen, 

 which, comparing our observation with Olszewski's, seems to 

 have the same absorptive powers as the dense gas, if we ex- 

 cept A. This exception is probably only apparent, and due 

 to the difficulty of observing A under the circumstances of 

 Olszewski's experiment. 



The greatly increased intensity of the absorption-bands at 

 high pressures bears out Jansen's observation, that in this 

 group the absorption is proportional to the product of the 

 thickness of the absorbent stratum into the square of its den- 

 sity, while the absorptions to which A and B belong vary 

 directly as the density. 



The appearance, on looking through the tube when gas at 

 high pressure is streaming into it, is very much like that of a 

 black and a colourless liquid, which do not mix, being stirred 

 together, and the tube soon ceases to transmit any light. 

 Transparency returns as the density becomes uniform. Cur- 

 rents produced by heating the tube at one or two points 

 produce a similar effect, and show that such currents in the 

 atmosphere of a star may stop all rays coming from its interior. 



We hope before long to get the tube fitted with rock-salt 

 ends and lenses, and to determine the total absorption of ra- 

 diation by similar masses of oxygen, nitrogen, and hydrogen. 



Cambridge, August 18, 1888. 



