Prof. Potter on the Definition of the Temperature of Bodies, 447 



In examining the preceding observations, it will be seen that 

 many of them indicate the compound nature of the spectrum, as 

 explained in my paper " On a New Analysis of Solar Light"*. 

 One of the most curious is the conversion of the blue space into 

 violet, indicating that the red rays, which I suppose exist to a 

 small extent in the blue space, have been less absorbed than the 

 blue, and consequently that the blue space has thus become 

 violet. 



M. Berard, who does not adopt my views, admits this import- 

 ant fact, but asserts that blue becomes violet by merely dimi- 

 nishing the intensity of its light. To me it appears a most 

 extraordinary opinion that a new colour should thus arise, and 

 that red rays not previously seen should appear, or be created, 

 in the blue space by merely reducing its light ! and I cannot 

 suppress the opinion that M. Berard should have sought for 

 some probable explanation of his fact, if it be one, without 

 regarding it as incompatible with the compound nature of the 

 blue space. I have proved by direct experiment, that in lights of 

 a certain feeble intensity the retina is more sensible to red than to 

 blue rays, and hence we have a reasonable explanation of the 

 violet colour in the blue space at certain feeble intensities of the 

 light. 



I do not avail myself, however, of this explanation. I deny 

 the fact that in my observations the blue became violet by dimi- 

 nishing the intensity of its light, and here is the proof drawn 

 from the observations themselves. In the greater number of 

 them the blue was not converted into violet, which it ought to have 

 been in diminished light. The conversion took place in a much 

 smaller number of observations, not because the light was dimi- 

 nished, but because the coloured medium absorbed the blue rays 

 and left the violet. As all my observations were made in light 

 of feeble intensity, the same result should have been obtained in 

 every case where the blue space was left in the spectrum. 



Allerly, November 15, 1862. 



LXI. On the Definition of the Temperature of Bodies, and on its 

 Measurement by Thermometers. By Professor Potter, A,M.f 



nnHOUGH we continually meet with the popular definition 

 JL of the temperature of bodies, or their sensible heat, as 

 indicated by instruments called thermometers, yet its scientific 

 definition is not readily found. M. Fourier, as far as I am 

 aware, is the only philosopher who has attempted to give a 



* Edinburgh Transactions, 1831, vol. xii. p. 123. 

 f Communicated by the Author. 



