268 SUPPLEMENT TO THE 



rious to them ; nor can their colour, strictly speaking, ever 

 be totally subdued, owing to what opticians call the irra- 

 tionality of the coloured spaces in the prism producing 

 what is termed the secondary spectrum ; moreover, their 

 component lenses may be imperfectly centered, and their 

 curves not spherical ; and, to crown all, they may be out 

 of adjustment. 



5thly. All spherical metals and common convex glasses^ 

 may of course be roughly considered as very much under 

 corrected; accordingly their outside rays are well known 

 to be much shorter than the inside ones, when acting 

 either with parallel or diverging light ; the colour of the 

 glasses may also be considered as very much under 

 corrected, being, indeed, in its primitive state of dis- 

 persion. Now this simple proposition will, in the sequel, 

 conduct us to the knowledge of the discriminating charac- 

 teristics of object-glasses and metals in all states of 

 correction and aberration; for example, those of an object- 

 glass over corrected both for sphericity and dispersion 

 must inevitably possess exactly opposite characters to one 

 under corrected, while one which is perfect will be neutral 

 in the phenomena which it will exhibit, and resemble 

 neither. In order to discover the state of the rays both 

 within and without the focus, we must of course put the 

 lens, object-glass, or metal, out of focus, and we shall 

 discover the state of the rays both as they advance 

 towards and recede from the focal point. We may 

 confirm this method by using a common lens of large 

 angular aperture, to form an image of the sun, or of a 

 candle, on a piece of rubbed glass, and observing the state 

 of the rays upon it as they approach or pass off from the 

 focal image, when it will be found that the image within 



