t i6 Count Rumford's Account of some 
with respect to the fallacy of many of the appearances in the 
foregoing experiments. So far from being able to observe any 
change in the shadow upon which my eye was fixed, I was not 
able even to tell when the yellow glass was before the lamp, 
and when it was not; and though the assistant often exclaimed 
at the striking brilliancy and beauty of the blue colour of the 
very shadow I was observing, I could not discover in it the 
least appearance of any colour at all. But as soon as I removed 
my eye from the tube, and contemplated the shadow with all 
its neighbouring accompaniments, the other shadows ren- 
dered really yellow by the effect of the yellow glass, and the 
white paper which had likewise from the same cause acquired 
a yellowish hue, the shadow in question appeared to me, as it 
did to my assistant, of a beautiful blue colour. I afterwards 
repeated the same experiment with the apparently blue sha- 
dow produced in the experiment with daylight and candle- 
light, and with exactly the same result. 
How far these experiments may enable us to account for the 
apparent blue colour of the sky, and the great variety of co- 
lours which frequently adorn the clouds, as also what other 
useful observations may be drawn from them, I leave to phi- 
losophers, opticians, and painters to determine. In the mean 
time, I believe it is a new discovery, at least it is undoubtedly 
a very extraordinary fact, that the eyes are not always to 
be believed, even with respect to the presence or absence of 
colours. 
I cannot finish this letter without mentioning one circum- 
stance, which struck me very forcibly in all these experiments 
upon coloured shadows, and that is, the most perfect harmony 
