170 COUNT EUMFOKD. 



lights. When the lights were equally intense, the 

 shadows were equally dark, but when one of the lights 

 was more powerful than the other, the shadow corre- 

 sponding to that other was rendered pale, because the 

 light from the most intense source fell upon it. Ee- 

 moving the more intense light farther from the screen, 

 until a point was reached when the shadows appeared 

 equal, Rumford obtained all the elements necessary for 

 the computation of the relative intensities of the lights. 

 He had only to apply the law of inverse squares, which 

 makes a double distance correspond to a fourfold in- 

 tensity, a treble distance to a ninefold intensity, and so 

 on. In connection with these experiments he dwells 

 repeatedly upon a defect which harasses the official gas- 

 examiners of the present day, and that is, the fluctua- 

 tions of the candles used as standards of measurement. 

 These photometric measurements are succeeded by a 

 brief but beautiful essay on * Coloured Shadows,' which, 

 in connection with another short essay on the < Harmony 

 of Colours,' strikingly illustrates Rumford's penetration 

 and experimental skill. He produced two shadows, one 

 from day light, the other from candle-light. The daylight 

 shadow being shone upon by the candle, was, as might 

 be expected, yellow, because the candle sheds a yellow 

 light. But the other shadow, instead of being colour- 

 less, was ' the most beautiful blue that it was possible 

 to imagine.' He states clearly that the colour of one 

 shadow is real, while that of the other is imaginary. 

 He finds it * impossible to produce two shadows at the 

 same time from the same body, the one answering to a 

 beam of daylight, and the other to the light of a candle 

 or lamp, without these shadows being coloured, the one 

 yellow, and the other blue.' He obtained shadows 

 from a light coloured by means of interposed glasses, 

 and compared them with shadows obtained from un- 



