1879.] 



Quantitative Spectroscopic Experiments. 



489 



thirds, in consequence of the ratio of the specific heat at constant pres- 

 sure to that at constant volume being now 1*66. It is not so easy to explain 

 why the bright band should appear while the sodium is evaporating, 

 when a depression of temperature should ensue. The depression of 

 temperature will, however, affect the hot bottom of the vessel, and so 

 it and the lowest layer of sodium vapour may be relatively cooler than 

 the layers above. Professor Stokes has suggested another cause which 

 may intensify the effect, depending on the fact that the radiation from 

 the sides of the vessel is partly reflective, partly emissive, while that 

 from the sodium is almost wholly emissive. The bottom of the vessel 

 will be somewhat diminished in brightness by the opening opposite 

 giving no radiation to be reflected by it, while the temperature of the 

 sodium being well sustained by the radiation of the sides, the want of 

 radiation from the opening will not make any sensible change in its 

 emission. Whatever explanation be given of the cause, the fact of the 

 gradual expansion of the bright band as the sodium evaporates bears 

 out the supposition that an increased thickness of sodium vapour, 

 with little or no change of density, produces a wider line. 



The very minute quantity of sodium required to produce a wide 

 absorption was shown in the experiments with the amalgams. In one 

 of these *042 grm. of 'an amalgam containing only ^ per cent, of 

 sodium was introduced into the platinum tube. That quantity of 

 mercury would at 1,500° C, which was about the temperature of the 

 furnace, fill about 25 cub. centims., occupying a depth of about 20 

 centims. of the tube. In this volume there was only about "00005 grm. 

 of sodium, yet the D absorption band was at least twice as wide as the 

 space between the two D lines. 



The quantity of sodium requisite to give the channelled spectrum 

 also appears to be very small. Between 4 and 5 mgrms. put into the 

 platinum bottle, whether alone or as an amalgam, usually gave the 

 channelled spectrum ; sometimes it was seen with even less than 

 4 mgrms. Now the wide part of the bottle held some 200 cub. 

 centims., so that the quantity of sodium diffused through this space 

 and up into the tube would give a vapour of very small density 

 indeed. The complete disappearance of the channellings as soon as 

 the vapours were a little compressed, so as to be driven down into the 

 hot part of the bottle, seems to indicate that the channellings were 

 due to comparatively cool vapour in the upper part of the vessel. 



The facility with which very high temperatures can be obtained by 

 sudden compression, naturally suggests a repetition of our experi- 

 ments with the electric arc in lime crucibles, under circumstances in 

 which a pressure of a few atmospheres can be suddenly applied ; and 

 this we hope to carry out. 



