Management of Light in Illumination. 107 



the immediate advantage, for the preservation of the 

 eyes and for facilitating vision, which must necessarily 

 be derived from the protection of the eyes from the too 

 powerful action of the direct rays which proceed from 

 the flames of lamps and candles, the only objection 

 that can be made to the proposal for masking those 

 flames by screens must be founded on a supposition 

 that those screens must necessarily destroy a great 

 deal of the light. Now that this is not the case in fact 

 I learned more than twenty years ago, from the result 

 of the following experiment. 



Two wax candles, of the same size, and burning with 

 the same degree of vivacity, were placed on two tables, 

 at the distance of about 8 feet from each other, in two 

 tall cylindrical glass jars, about 6 inches in diameter, 

 made of fine transparent glass ; the polish of the sur- 

 face of one of them having been taken off by grinding 

 it with emery. At the distance of about 1 6 feet from 

 these lights, a sheet of white paper was presented to 

 them, in a vertical position ; and a small cylinder of 

 wood, about a quarter of an inch in diameter, held in a 

 vertical position, was placed before the paper, at the 

 distance of about 2 inches. 



This cylinder caused two shadows to be cast on the 

 paper; and as these shadows were reciprocally illumi- 

 nated by the two burning wax candles, if that placed 

 in the transparent glass jar had emitted considerably 

 more light than that placed in the jar of ground glass, 

 the two shadows could not have been of the same 

 density. They were, however, very nearly of the same 

 density; which, as it proved evidently that there was 

 little or no loss of light in its passage through ground 

 glass, as this was contrary to my expectation, it sur- 



