18 On the Jnjluence of Color on Radiation. 



This very extensive induction I do not pretend to have made, but 

 I have multiplied our experiments so much beyond the number made 

 by Dr. Stark, as to be able to show that the supposed influence of 

 color on the absorption and radiation of heat remains yet to be demon- 

 strated, and thus to prevent the admission as proved of what is more 

 than doubtful. 



The principal object was to select a considerable variety of pigments 

 of the same color differing chemically, and of different colors chemi- 

 cally allied, and, as subsidiary, to ascertain the effect of changes of 

 color produced by chemical means on different substances, and the 

 effect of the material used to apply the pigment to the radiating body. 

 Several tin cylinders were procured, two inches high, and 1| in 

 diameter, closed at the bottom, and having fitted to the top a slightly 

 conical tube, to receive a perforated cork, through which to pass the 

 stem of a thermometer. One of these vessels having been selected 

 was coated in successive layers with a pigment. Water which was 

 boiling in a porcelain capsule was then poured into the cylinder, which 

 was suspended by means of two lateral hooks to cords attached to the 

 canopy covering the lecture table. A thermometer introduced through 

 a cork had its bulb nearly in the middle of the axis of the cylinder, 

 and the thermometer by displacing part of the water proved that the 

 quantity contained was the same in each case. A temperature was se- 

 lected for beginning the experiments, sufficiently below that which 

 the introduction of boiling water produced, to permit the rate of cool- 

 ing to have become uniform, and one for ending which was high 

 enough to prevent uncertainty from the slowness of the fall of tempe- 

 rature. The instant of the arrival of the mercurial column at any de- 

 gree on the scale, and of its leaving the same, was noted, and a mean 

 taken for the time of being at that temperature ; a precaution which 

 though superfluous in such experiments as these, will, I am persua- 

 ded, be found of importance where minute accuracy is desired in in- 

 vestigating the motion of heat. One of us observed the thermometer, 

 the other noted the time by a pocket chronometer. 



The time of cooling of the cylinder coated with coloring matter 

 having been ascertained, an additional layer of the same substance 

 was put upon it and the cooling again observed. The time of cooling 

 diminished, of course, until that thickness was attained beneath which 

 no radiation takes place, the time then slowly increased with each 

 additional coat, the conducting power entering as an appreciable ele- 

 ment into the rate of cooling. To show the decided nature of the 



