152 TESTING THE MICROSCOPE. 



appear to us less suitable for testing the aberration than the above- 

 mentioned fissures in an otherwise black field of view. The latter 

 afford, moreover, the essential advantage that a withdrawal of 

 the diaphragm or of the condensing lens renders the marginal 

 zone of the objective inoperative ; the observer is thus enabled 

 to confine the testing to a gradually diminishing central portion 

 of the refracting surfaces. It is often observed that the strong 

 appearance of fog surrounding the microscopic image disappears 

 immediately if the extreme marginal rays are excluded, whilst 

 in other cases it decreases almost imperceptibly, and is therefore 

 due, chiefly, to the central rays. The determination of such 

 differences is, of course, a part of the task of testing. If mercury- 

 globules, or air-bubbles, are used for this purpose, suitable dia- 

 phragms must be applied to the objective ; and there is the further 

 disadvantage that the window-frames or their surroundings are 

 never sufficiently dark to form a black background in a catoptric 

 or dioptric image, and hence the presence of a slight fog can be 

 less easily recognized. Eegarding this point, we should still give 

 the preference to a large spherical air-bubble, which forms the 

 image of a small diaphragm, rather than to a mercury-globule, 

 because the field of the image appears perfectly black up to the 

 bright inner ring. The amount of light which an image of this 

 kind produces is always somewhat less than that given by a 

 fissure, because the marginal rays are weakened by partial reflexion 

 at the upper surface of the air-bubble. 



The answer to the further question, whether an objective, 

 affected by aberration, is under- or over-corrected, is based upon 

 the phenomena which are produced by the varying distance of 

 the eye-piece, or, what in principle is equivalent, the increase or 

 decrease of the object-distance. If we take the simple case of the 

 aberrations all following the same direction, so that the refraction 

 of the objective gradually increases or decreases from the centre 

 to the periphery, the effect of change in the focal adjustment can 

 readily be understood. If A B (Fig. 90) is the objective, p the 

 focal point, or the small focal space occupied by all the central 

 rays up to a certain inclination, and p p the greatest length of 

 the aberration of the marginal rays, for which, therefore, the 

 objective is over-corrected ; the real image will evidently be most 

 clearly seen on a screen when it coincides with a plane/! drawn 

 through the point p. This image appears, however, bordered by 



