114 NATURAL SCIENCE. August. 1895. 



(as in a beam of " parallel rays ") becomes strongly concave, that is, 

 the " rays " become convergent. 



The representation in diagrams of the effect of the lens is not 

 difficult if contours of wave-fronts are drawn instead of "rays," and 

 the mathematical difficulties may be avoided in lectures to medical 

 students, by using only those formulae which do not involve either 

 radii of curvature or indices of refraction. The exact mathematical 

 treatment of the problem, from the point of view of the second doctrine, 

 appears to be almost impracticable ; for the measurement of the 

 refractive index at each point in the lens, though theoretically possible, 

 would be both exceedingly laborious and also liable to very great 

 error. A very slight tremor indeed in the razor, at the moment of 

 cutting the section of the lens, would lead to considerable error ; and 

 the use of a cover-glass would bring in new errors which it would be 

 practically impossible to eliminate. Such exact measurements are, 

 however, not necessary for the purposes of the physiologist : all he 

 requires is to show roughly that the refractive index does increase as 

 we pass from periphery to centre of the lens; in fact, the only experi- 

 ment he requires to make is the one I have described. 



C. Herbert Hurst. 



