156 ELEMENTS OF LABORATORY WORK 



slit, we may now use the incandescent vapour of various sub- 

 stances by placing them on platinum foil in the Bunsen flame. 

 Instead of the continuous spectrum we shall observe bright 

 lines varying in colour, number, and relative position with the 

 substances used. By throwing the rays of the sun upon the 

 slit by a reflector we obtain a continuous spectrum containing 

 certain dark lines to be afterwards explained. 



We may reasonably assume from this experiment that 

 variety of colour is caused by the varied action of bodies upon 

 white light. A red object absorbs that portion of white light 

 which is not red, and reflects the red light alone. 



It is important to note that thermal dispersion is observ- 

 able, and also thermal reflexion and refraction; but the 

 thermal effects of radiation are more troublesome to deal with 

 than light. 



100. The Interference of Light. It was first noticed by 

 Young that two rays of light, converging in a dark room from 

 two small holes close together, yield certain alternate light arid 

 dark bands, when a screen is placed where they overlap. The 

 dark bands disappear when one ray is cut off. In other words, 

 the addition of light may, under certain circumstances, produce 

 darkness. We have already learnt from the use of a prism 

 that white light is the joint effect of light of various kinds 

 or colours, and this is again shown by the observation that 

 with white light the dark bands are alternated with coloured 

 ones in this experiment, while with light of one colour, such 

 as the yellow light given by sodium compounds, alternate 

 bands of black and yellow alone are given. 



Fresnel obtained a similar result more conveniently in 

 two ways : one by the aid of reflexion, the other by means of 

 refraction. In the first method two metallic mirrors, a b and b c, 

 are placed side by side, so as to be nearly in the same plane, 

 but not quite. A ray of light from d is caused to fall upon both 

 mirrors. The reflected rays converge to a point /, and appa- 

 rently come from two vertical images, g and h, close together at 

 the back of the mirrors. On placing a screen at /, alternate 

 dark and light bands should be obtained. 



In Fresnel's second experiment a prism a b c of very large 



