[ 123 ] 



XIX. On a Means to determine the Pressure at the Surface of 

 the Sun and Stars, and some Spectroscopic Remarks. By 

 Eilhaed Wiedemann*. 



IN a former paper I tried to show that we can calculate, at 

 least approximately, the time elapsing between two en- 

 counters of the molecules of a gas by measuring the greatest 

 difference in the length of path at which interference of two 

 rays of light coming from the same source of light is possible. 

 I may be allowed here to quote a passage out of that paper f : — 



" Two rays of light can only interfere if they emanate from 

 the same source of light, and if there is no sudden change of 

 phase in the source of light during the time elapsing between 

 the two instants at which the first and second ray leave the 

 source. That difference in phase at which interference is still 

 possible is therefore a measure of the time during which no 

 sudden change in phase has taken place at the source of light. 

 A luminous body sends out light coming from a great number 

 of atoms or molecules. Each molecule will only vibrate regu- 

 larly as long as it does not come within the sphere of action 

 of another molecule ; that is, it will only vibrate regularly in 

 the time elapsing between two encounters. For the different 

 molecules, the time elapsing between two encounters may, 

 according to the molecular theory of gases, be either zero or 

 infinitely large. But during a very short fixed time only a 

 small number of molecules will have suffered encounters; only 

 that small number will produce an even illumination of the 

 field of view : the greater number will have vibrated regularly 

 during the whole time; and for all of these interference is pos- 

 sible. The interference -bands in that case are sharp. If, 

 however, the time which we consider to elapse between the 

 emanation of the two interfering rays increases, a greater 

 number of molecules will have suffered encounters, and the 

 bands will therefore be less distinct. It follows that the higher 

 the order of interference-bands, the more diffused and indi- 

 stinct the bands will be. If the difference of phase corresponds 

 to a difference in time greater than that necessary for the 

 completion of the mean free path, the bands will rapidly dis- 

 appear, as in that case the greater number of molecules have 

 suffered encounters during the time considered." 



Fizeau and Foucault, and, more recently, J. J. Miiller and 

 Mascart, have determined the greatest difference of path at 



* Communicated by the Physical Society, having been read at the 

 Meeting- on June 12th. 



f Wied. Ann, v. p. 503 (1878) ; Phil. Mag. [5] vii. pp. 79, 80. 



K2 



