On a new Form of Spectrometer. 75 



quantity of the vapour from the gaseous to the liquid state 

 sufficient to cover the surface a few molecules deep. The mo- 

 lecular attraction seems to be very slight in gases, where the 

 molecules are ten or fifteen molecular diameters apart. To 

 get some idea of the amount of work done in compressing one 

 gramme of oxygen to the liquid state, we may consider that in 

 the union of one gramme of hydrogen with eight grammes of 

 oxygen 34,462 units of heat are produced. It matters not 

 that the condensation is brought about by the energy of che- 

 mical separation rather than by the work done in pressing 

 them together in a cylinder. 



The superficial energy of platinum is 169*4 metre-grammes 

 per square metre, or '01694 per square centimetre, equal to 

 '00004 of a unit of heat. The proportion 



9 : 34,462 = a : '00004 



gives the weight of water condensed on one square centimetre 

 of surface, or the volume in cubic centimetres as '00000001, 

 which is the thickness of the layer, or diameter possibly, of 

 the molecules. 



Physical Laboratory, Harvard College, 

 May 14, 1879. 



IX. On a new Form of Spectrometer, and on the Distribution 

 of the Intensity of Light in the Spectrum. By John Wil- 

 liam Dkapee, M.D.j President of the Facxdty of Science in 

 the University of New York*. 



I HAVE invented a spectrometer which, I think, opens a 

 new and interesting field to those who are engaged in 

 spectrum-analysis. 



The ordinary spectroscope is occupied with the frequency 

 of gether-vibrations or wave-lengths. That which I am about 

 to describe has a different function. It deals with the in- 

 tensity or brilliancy of light. It depends on the well-known 

 optical principle that a light becomes invisible when it is in 

 presence of another light about sixty-four times more brilliant. 



In some researches published by me in 1847, on the pro- 

 duction of light by heat, or the incandescence of bodies, I used 

 this method as a photometer, and became sensible of its value. 

 The memoir in which those experiments are related may be 

 found in my recently published i Scientific Memoirs,' page 

 23. 



Having also published in 1872 a memoir on the distribution 

 of heat in the prismatic spectrum, and shown that the course 



* Coiimiunicated by the Author. 



