Focal Isolation of Long Heat- Waves. 249 



surface of the film the oxygen atoms produce more effect 

 than the hydrogen ones. The relative intensities on the 

 photographic plate do not always seem the same as on the 

 willemite screen. 



I find from the photographs that the slope of the straight 

 part of the curves corresponding to the secondary radiation 

 due to the hydrogen atom, negative as well as positive, for 

 the slopes of the two are the same; does not vary appre- 

 ciably with the potential difference between the electrodes 

 in the discharge-tube. 1 have taken photographs with the 

 tube in states for which the equivalent spark gap in air 

 varied from *7 to 4 cm., and found only very slight alterations 

 in the slopes of the curves. As the velocity of the particles 

 in the rays is proportional to the tangent of the angle of 

 slope, this implies that the velocity of the particles in the 

 secondary rays is almost constant, a result 1 had previously 

 arrived at by the willemite screen. 



I wish to express my thanks to Mr. F. W. Aston, of 

 Trinity College, and Mr. E. Everett, for the assistance they 

 have given me with these experiments. 



XXVII. Focal Isolation of Long Heat-Waves. 

 By H. Rubens and R. W. Wood *. 



THE isolation of very long heat-waves, which is usually 

 accomplished by selective multiple reflexions (Rest- 

 strahlen method) can be accomplished also by selective 

 refraction. It was shown in 1899 by Rubens and Aschkinass 

 that it was possible to separate very long heat-waves from the 

 radiation of an incandescent source by means of quartz prisms 

 of small angle *. This method, involving the use of a spec- 

 trometer, did not however prove to he very efficient, on 

 account of the large loss of energy, and the isolated radiation 

 disappeared almost entirely if a quartz plate of any con- 

 siderable thickness was interposed in the path of the 

 rays. 



The method which will be presently described is free 

 from these objections, and has enabled us to obtain heat- 

 waves of greater wave-length than any hitherto observed 

 and with sufficient intensity to make accurate measurements 

 of their properties possible. Like the other method, it 

 depends upon the selective refraction of quartz, the separa- 

 tion being accomplished by means of lenses however. 



* Communicated by the Authors. 



t H. Rubens and E. Aschliinass, Wied. Ann. lxvii. p. 459, 



