A CHAPTER ON LIGHT AND COLOUR. Ill 



from the size of the rings of those colours. It is necessary to add 

 a retardation of half a wave-length at each reflection taking place 

 in glass at a surface separating glass and air, an addition which is 

 in perfect accordance with what would be anticipated by theory^ 

 and is, in fact, necessary for the continuity of the motion. But 

 for this the centre of the reflected system would be bright, not, as 

 is the case in fact, dark. The above conclusion as to the position 

 of a given colour in the series is modified by this fact. 



The iridescent colours given by a drop of oil or turpentine on 

 the surface of water, by the thin scales of the wings of some 

 insects, by thin films of mica and other readily laminated bodies, 

 and by soap-bubbles, are all due to the same cause. Two streams 

 of light, which in their passage to and from the thin film have 

 traversed paths of slightly different lengths, are brought to the 

 same point. At one place they strengthen each other; at a neigh- 

 bouring place the displacements due to them are opposed, and 

 they extinguish each other. For one thickness of the film the 

 difference of path may be an exact multiple of half a wave-length 

 for a particular colour, which will therefore be wanting at that 

 place, and the film will in consequence appear coloured. For 

 different thicknesses the colours will be different. 



2 2.— Though the sensation of light is not produced by radia- 

 tions whose wave-lengths are greater than that of red hght or less 

 than that of violet, the nature of such radiation is not different 

 from that of the intermediate wave-lengths. Beyond the extreme 

 red end of the spectrum we find radiation which possesses the 

 power to heat bodies exposed to its influence, and m this part, as 

 well as in the visible part of the spectrum, there are evidences of 

 lines of absorption in the sun's radiation. 



Beyond the violet end of the spectrum also, there are radia- 

 tions possessed of very strong actinic power, and capable of being 

 rendered obvious by their action on sensitive photographic plates. 

 Probably one reason why photographs of celestial objects some- 

 times shew details not visible to the eye is that they are due to 

 radiations of shorter wave-length than the extreme visible part of 

 the spectrum. 



23.— Fluorescence.— Celestial photographs reveal faint stars 

 owing to a different reason. The eftect of a light sensation on 



