CELESTIAL CHEMISTRY. 365 



Sir David Gill's words where he says that the aim of the new- 

 astronomy is not so much to find out where a star is, but 

 what it is, and to ascertain wdiat its chemical constitution is, 

 and the chemical history of its development. 



Bv way of introduction let me now place before you a few 

 fundamental physical facts. First of all, a ray of light cannot 

 of itself turn a corner, or travel in any other direction than a 

 straight line. ( )n this law of light what is called pinhole 

 photography depends : the rays of light from any external 

 object which pass through a tiny hole into a darkened chamber 

 can fall only on points exactly opposite to those where they 

 started from : they must, therefore, cross each other and so 

 the image which is formed is inverted. 



An outside cause may. however, bring about a sudden change 

 of direction : for instance, when the beam of light falls 

 slantingly on a polished surface — like that of a looking glass — 

 its path becomes bent at a sharp angle; it is as we say. reflected. 

 Again, when a ray of light passes in a slanting direction from 

 one transparent substance into another, an alteration of 

 direction also takes place, but of a different kind : we then say 

 that the ray of light has been refracted. We have all observed 

 this phenomenon again and again, when the end of a stick 

 has been dipped under water, and presented the appearance of 

 having been bent just at the water surface. Here, of course, 

 the rays of light from the lower part of the stick become bent 

 at the surface, and the eye. looking along the path of these 

 rays, is deceived into thinking that the stick itself has been bent. 

 The amount of refraction that a beam of light undergoes in pass- 

 ing from one transparent substance to another is always the 

 same for the same substances, and we can measure this by 

 what is called Descartes' apparatus. It consists of a graduated 

 circle, witli a glass vessel at the centre filled with water or 

 any other liquid whose refractive index it is intended to 

 measure. Through a small hole in a brass slide capable of 

 moving along the graduated circumference a beam of light is 

 allowed to fall on the liquid, and. passing through the latter, is 

 refracted, the exact direction in which it is bent being 

 ascertained by moving a ground glass slide until the refracted 

 light falls on it, when, by means of the graduations round the 

 edge of the circle, or by two graduated rules, the angle of 

 refraction can be accurately measured. 



Now suppose that a beam of light passes through water 

 contained in a glass vessel with parallel sides; it will then 

 undergo two refractions; one when it passes from the air into 

 the water, and another when it leaves the water and emerges 

 into the air again. The second refraction will send the refracted 

 light along a path parallel to that which it pursued before 

 entering the water; but observe, please, that this is the case 

 only when the opposite sides of the glass vessel are parallel. 

 When these opposite surfaces are not parallel, but are inclined 

 to one another, as in a prism, the second refraction does not 

 bend the beam of light back into the first direction, but turns 

 it still more out of its former coiu^se : if I may so put it, a 

 prism emphasises refraction. 



