412 MANIPULATION AND PRESERVATION OF THE MICROSCOPE 



which constitiite our daylight fall from every point of the visible 

 heavens (though with greatly diminished intensity). That is to say, 

 we have at disposal a light source extending over 180, 'while the sun 

 itself extends over a visual angle of bat half a degree. Being thus 

 surrounded by an illimitable and self-luminous expanse of ether un- 

 dulations, the question is no longer of parallel rays only, but of light 

 emanating from an outer circle above the earth upon every point of 

 the earth's surface ; and a mirror exposed to such a luminous atmo- 

 sphere must both receive and reflect from all sides and upon all 

 sides. If, however, it be placed under the stage of a microscope, 

 all vertical light is intercepted, and there remains nothing but the 

 oblique incidence as the starting-point of the theory of illumination 

 by converging light ; for it scarcely needs repetition that obliquity 

 of incidence gives inevitable rise to obliquity of reflexion ; and it 



S' 



FIG. 353. Light from the open sky falls upon the mirror in all directions. 



becomes equally clear that in order to strike the object the 

 must al/waysfall obliquely on the 'mirror. 



Then it follows from what has been said that the light falling 

 from the open sky upon a mirror falls in all conceivable directions. 

 Tims fig. 353 shows the lines 1 to 7, including an angle of 30. If 

 nothing intervene, the light of that sky surface must fall upon the 

 mirror, a b, and be reflected on O. The intermediate rays, 2, 3, 4, 

 5, 6, form the converging Ultnn'nndnnj /><'//cil, with of course an in- 

 finity of others filling up the spares between. 



In other words, every point of a mirror is a radiant of a whole 

 hemisphere, and this is equollt/ l/'/f K hctlicr the. mirror be. plmn', 

 concave, or convex, so long as it is exposed to a boundless sky. 

 Therefore a plane, concave, or convex mirror will give a cone of 



