102 Chj'eivalioiij »» Colturtd Shadnivs. 



firous of comparing the inienfity of the light of a clear Iky by day with that of a common 

 wax-candle, I darkened my room, and letting the day-li^ht from the north, coming through 

 a hole near the top of the window- (liutter, fall at an angle of about 70° upon a (hcet of very 

 fine white paper, I placed a burning wax-candle in fuch a pofition that its rays fell upon the 

 fame paper, and, as near as 1 could guefs, in the line of refleflion of the rays of day-light 

 from without; when, interpofing a cylinder of wood, about half an inch in diameter, before 

 the centre of the paper, and at the didancc of about two inches from its furface, I was 

 much furprifed to find^hat the two fliadows projeded by the cylinder upon the paper, in- 

 llead of being merely fliades without colour, as I expeiflcd, the one of them, that which, 

 carrefponding with the beam of day-liglit, was illuminated by the candle, was yellow; 

 while the other, corrt fponding to the light of the candle, and confequently illuminated by 

 the light of the heavens, was of the mofl beautiful blue that it is poflible to imagine. 

 This appearance, which was not only unexpedled, but was really in itfelf in the higheft 

 degree (Iriking and beautiful, I found upon repeated trials, and after varying the experiment - 

 in every way I could think of, to be fo perfe£^ly permanent, that it is abfolutely impoffible 

 to produce two fhadows at the fame time, from the fame body, the one anfwering to a 

 beam of day-light and the other to the light of a candle or lamp, without thefe fliadows 

 being coloured, the one yellow, and the other blue. 



The experiment may very eafily be made at any time by day, and almoft in any place, 

 and even by a perfon not in the leaft degree verfcd in experimental refearches. Nothing 

 more is neceflTary for that purpofc than to take a burning candle into a darkened room, in 

 the day-time, and open one of the window-fliutters a little, about half or three-quarters of 

 an inch for inftance ; when, the candle being placed upon a table or (land, or given to an 

 alFiftant to hold, in fuch a fituation that the rays from the candle may meet thofe of day- 

 light from without af an angle of about 40° at the furface of a flieet of white paper, held 

 in a proper pofition to receive them, any folid opake body, a cylinder, or even a finger, 

 held before the paper, at the ditlance of two or three inches, will projeft two fliadows upon 

 the paper, the one blue and the other yellow. 



If the candle be brought nearer to the paper, the blue (hadow will become of a deeper 

 hue, and the yellow fliadow will gradually grow fainter ; but if it te removed farther off, 

 the yellow fhadow will become of a deeper colour, and the blue fliadow will become 

 fainter ; and the candle remaining ftationary in the fame place, the fame varieties in the 

 ftrength of the tints of the coloured fliadows may be produced merely by opening the window- 

 fliultcr a little more or lefs, and rendering the illumination of the paper, by the light 

 from without, ftironger or weaker. By either of thefe means, the coloured fliadows may 

 be made to pafs through all the gradations of (hade, from the deepefl to the lighteft, and 

 vice verfa ; and it is not a little amufing to fee fliadows thus glowing with all the brilliancy 

 of the pureft and moft intenfe prifmatic colours, then pafliiig fuddenly through all the 

 varieties of fliade, preferving in all the moft perfe£l purity of tint, growing ftronger and 

 fainter, and vanifiiing and returning, at command. 



With refpedl to the caufcs of the colours of thefe fhadows, there is no doubt but they 

 arife from the different qualities of the light by which they are illuminated ; but how 

 they arc produced, does not appear to me fo evident. That the (hadow correfponding to 

 the beam of day-light which is illuminated by the yellow light of a candle, fhould be of a 

 yeIlowi(h hut, is not furprifing ; but why is the (hadow correfponding to the 'light of the 



2 candle. 



