318 Dr. Herschel's Experiments on the solar, and 



was two-tenths of an inch from the margin of the semicircular 

 pasteboard cover; then, taking the thermometers, I had as 

 follows. 





No. 2. 



No. 3 



0' 



57 



57 



1 



78 



57 



Here, there was no longer the least tinge of any colour, or 

 vestige of light, to be seen on the ball of the thermometer ; so 

 that, in one minute, it received 21 degrees of heat, from rays 

 that neither were visible before, nor could be rendered so by 

 condensation. 



To account for the colour which may be seen in the focus, 

 when the last visible red colour is less than two-tenths of an 

 inch from the margin of the pasteboard which intercepts the 

 prismatic spectrum, we may suppose, that the imperfect refrac- 

 tion of a burning lens, which from its great diameter cannot 

 bring rays to a geometrical focus, will bring some scattered ones 

 to it, which ought not to come there. We may also admit, that 

 the termination of a prismatic spectrum cannot be accurately 

 ascertained, by looking at it in a room not sufficiently dark to 

 make very faint tinges of colour visible. Arid, to this must be 

 added, that the incipient red rays must actually be scattered 

 over a considerable space, near the confines of the spectrum, on 

 account of the breadth of the prism, the whole of which cannot 

 bring its rays of any one colour properly together ; nor can it 

 separate the invisible rays intirely from the visible ones. For, 

 as the red rays will be but faintly scattered in the beginning of 

 the visible spectrum, so, on the other hand, will the invisible 

 rays, separated by the parts of the prism that come next in 

 succession, be mixed with the former red ones. Sir Isaac 



