ZOOLOGY AND BOTANY, MICEOSCOPY, ETC. 121 



if intended to indicate microscopical dimensions, should be accurately 

 ascertained, care being taken that the minuteness of dimensions and 

 general delicacy and perfection of the test object should be adapted to 

 the power of the lens. A fairly correct estimate of the relative per- 

 formance of lenses of moderate magnifying power may doubtless be 

 thus made by a competent observer, but it is not possible from any 

 comparisons of this kind to determine what may or ought to be the 

 ultimate limit of optical performance, or whether any particular lens 

 under examination has actually reached this limit. 



Assuming the manipulation of the instrument and the illumination 

 of the object to be as perfect as possible, and, further, that the test- 

 object has been selected with due appreciation of the requirements of 

 perfect optical delineation, a fair comparison can only be drawoi 

 between objectives of the same magnifying power and aperture. 

 Which of two or more objectives gives the better image may be 

 readily enough ascertained by such comparison, but the values thus 

 ascertained hold good only for the particular class of objects examined. 

 The best performance realized with a given magnifying power may 

 possibly exceed expectation, yet still be below what might, and there- 

 fore ought to be obtained. On the other hand, extravagant expecta- 

 tions may induce a belief in performances which cannot be realized. 

 The employment of the test objects most in use is, moreover, calcu- 

 lated to lead to an entirely one-sided estimation of the actual working 

 power of an objective, as, for example, when " resolving power " is 

 estimated by its extreme limits rather than by its general efficiency ; or 

 " defining power," by extent of amplification rather than by clearness 

 of outline. So that an observer is tempted to af&rm that he can 

 discern through his pet lens what no eye can see or lens show. This 

 happens chiefly with the inexperienced beginner, but not unfrequently 

 also with the advocate of extremely high powers, in whose mind 

 separation of detail means analysis of structure, and optically void 

 interspaces prove the non-existence of anything which he does not see. 



As much time is often lost by frequent repetition of these com- 

 petitive examinations (which after all lead to no better result than 

 that the observer finds or fancies that one lens performs in his hands 

 more or less satisfactorily than some other lens), it seems worth while 

 to invite attention to a mode of testing which can be readily practised 

 by any person, with a fair certainty of being able to form a really 

 correct estimate of the working capacity of his instrument, measuring 

 this by a standard of strict optical requirements. The advantage of 

 substituting some such proceeding for the comparative trial of lens 

 against lens, so long in vogue, can scarcely be disputed. For, 

 although the best warrant of a well-constructed lens is the fair repu- 

 tation of its maker, and the choice of an objective resolves itself for 

 the most part into the selection of the particular make of one or other 

 of the best accredited opticians, still, when the instrument is pur- 

 chased, its possessor frequently becomes haunted by the desire to pit 

 its performance against that of some neighbour's instrument, or to 

 match the performances traditionally accepted in our handbooks. A 

 short and easy method of testing an objective, not by comparison with 



