Inflexion of Light by Incandescent Surfaces. 179 



white paper was arranged directly in the path of the reflected 

 beam ; and the distance of the paper from the platinum was in- 

 definitely varied during the experiment. Having accurately 

 marked the boundaries of the reflected beam and its intensity, as 

 far as the eye could judge, the platinum was made part of the 

 circuit of a voltaic battery, the intensity of which was varied so 

 as to produce efi^ects on the foil varying from a heat scarcely 

 visible in the dark, to incandescence up to the point of fusion, or 

 rather to the pomt at which the foil broke, from its diminished 

 cohesion ; for with a weight suspended, although not more than 

 barely sufficient to keep the foil stretched, it always broke ofi" at 

 a temperature short of its point of fusion. In none of these 

 variations, however, was there the slightest apparent difiference 

 in the reflected light on the paper. Or if, as occasionally hap- 

 pened, the shape a little changed during the progress of the ex- 

 periment, it was fully explained by the elongation dependent 

 upon the heat, or by the consequent removal of slight curvatures. 

 A similar experiment was made with difi"used daylight, and 

 with similar effects; also with the light from an Argand lamp. 

 In the latter case, when the reflected beam was so dim as to be 

 interfered with by the light afforded by the incandescent pla- 

 tinum, the image was proportionately aff"ected. Still it preserved 

 its character, and, as far as could be judged, its intensity ; and 

 it was only by a very high degree of incandescence and a very 

 feeble incident light that the reflected image seemed to merge 

 in the direct light from the incandescent body. 



I now brought my eye into the position where the paper 

 had been placed so as to catch the reflected beam, while my 

 assistant alternately made and broke contact with the battery. 

 When the incident sunlight was sufficiently intense to mask 

 the emitted light of incandescence, I could not in the slightest 

 degree distinguish whether the platinum was ignited or cold. 

 When I first tried it, I two or three times complained wrongly 

 to my assistant that he had not made contact when I told him 

 to do so; when the incident light was very dim, the emitted 

 light was of course also distinguishable. I now caused the 

 spectrum from a flint-glass prism to fall on the platinum, and 

 with similar effect ; i. e. when the reflected spectrum was very 

 intense, no difi"erencc could be detected between the light from 

 the platinum, whether cold or ignited, or whether received upon 

 paper or upon the eye ; when less intense, the red portion of the 

 spectrum was elongated by the light of incandescence, and the 

 other portions partook of the character of the spectrum super- 

 posed upon, or blended with, the light of incandescence. 



The prism was also arranged so as to intercept the reflected 

 instead of the incident beam; the effects were similar. 



