317 



fact each part which so intercepts one and transmits another por- 

 tion of the ray acts as an origin of light; and that from every point 

 of the object this light diverges in every direction as from an 

 original source. Now, the microscope is designed to give vision 

 by light thus divergent; all else is an intruder, and does harm 

 by causing fog and glare, which tend to obscure by overpowering 

 the aerial picture. 



" That this divergence of light from the surface after transmission 

 through the substance does exist, admits, I think, of no doubt. 



" Suppose a pencil of light passing through a dark space, and 

 across the line of vision of a spectator, no impression whatever will 

 be conveyed to his eye, save by reflection from particles floating in 

 the atmosphere : let a translucent object of any kind be so placed 

 that the pencil must pass through it, and immediately, no matter 

 what may be the obliquity of direction from the eye of the observer 

 to the anterior surface of the object, it will become visible. 



"Taking it then, as established that the only rays of light which 

 assist in forming the eerial picture on which microscopic vision de- 

 pends are those which diverge from the object under observation, 

 it follows that all other rays which enter the instrument, if they 

 reach the eye, tend to confusion, and it becomes important to ascer- 

 tain the best method of admitting such useless rays, when they can- 

 not, or it is not desired that they should, be excluded. Without 

 occupying time by discussing this point, I believe, as the result of 

 consideration and experiment, that these rays ought to enter either 

 so obliquely as to pass entirely across the axis, and thus not reach 

 the eye at all, which is the condition of black ground illumination ; 

 or that they should be as nearly as possible parallel, in which case 

 they come to a focus, close behind the object-glass, and, therefore, 

 by their rapid divergence the greater part are thrown against the 

 non-reflecting inside surface of the tube, and are thus absorbed, 

 leaving but a small portion round the axis of the cone to reach the 

 eye in a diffused state, giving a field more or less luminous accord- 

 ing to the distance of their focus from the eye. 



"Having satisfied myself of the truth of these premises, I sought 

 for a construction which would realize them; first addressing 

 myself to parallel light with a luminous field. 



" To obtain a parallel beam of condensed light is very difficult, 



