292 TECHNICAL MICROSCOPY. 



ployed, liable to error, because the image of the object suffers a 

 greater or less distortion through the eye-piece, in consequence 

 of which the amplification appears higher or lower according to 

 the distance from the centre ; whence, of course, as Harting has 

 already pointed out, the relative value of the micrometer-scale 

 undergoes an alteration. In order to avoid this defect of the 

 eye-lens, Mohl does not measure the size of the objective-image 

 under the fixed eye-piece by moving the spider-web, but moves 

 the eye-piece with the crossed-threads right across the image by 

 means of the micrometer-screw, so that on each adjustment only 

 its centre is optically effective. 



It appears to us that the asserted defect of the eye-lens, which 

 is said to be eliminated by Molil's suggestion, does not in reality 

 exist. For it is perfectly immaterial whether the eye-lens 

 magnifies uniformly or not, provided the image which it gives 

 possesses the distinctness necessary for focusing. The plane of 

 the objective-image coincides of course with the plane of the 

 micrometer-scale, and the space between the two margins, through 

 which the micrometer-screw has to pass, is a quantity entirely 

 independent of the distortion of the image by the eye-lens ; both 

 image and scale suffer the same changes through the eye-lens. If 

 the margins of the objective-image are sharp and clearly defined 

 so that the threads in the eye-piece can be adjusted exactly to 

 them, then, even if the eye-lens were to magnify twice as strongly 

 at the edge as at the middle, the result of the measurement would 

 remain unaffected. The determinations of size made by Harting, 1 

 according to which ten divisions of a glass micrometer, measured 

 separately in succession and in the centre of the field of view, 

 corresponded in section to 75 -8 degrees of the index, while the 

 total extent of the ten divisions, measured directly, amounted to 

 73-5 degrees (instead of 75*8), prove only that the objective-image 

 measured was not magnified uniformly but less towards the edge. 2 

 This is a defect which is met with to a greater or less degree in all 

 Microscopes, and the injurious influence of which can only be 

 eliminated, in cases where extreme accuracy is required, by 

 determining the relative value of the micrometer-divisions or 



1 " Das Mikroskop," 2nd ed. ii. p. 240. 



2 "We cannot understand how Harting concludes that these differences are 

 to be explained by higher magnifications of the marginal parts. On the 

 causes of lower magnification towards the margin, vide p. 64 et seq. 



