1878.] determining the limits of microscopic vision. 225 



In the case of Nobert's celebrated lines, ruled on glass at the 

 rate of 112,000 per inch, it is impossible to estimate the exact 

 diameter either of the ruled lines or the adjacent separating inter- 

 vals. The photographs examined by a good lens give a most 

 uncertain appearance. The lines appear about as broad as the 

 intervals in the example lent to me. The interval may therefore 

 roughly be taken at half, i.e. 2 -20^000 ^^' '^^^J nearly agreeing with 

 the experiment just detailed. But in this case the miniature was 

 only reduced 38 times. The spider-lines most closely, however, 

 resembled the appearance of Nobert's finest band. In the more 

 recondite effect of the miniature by Beck's -gL, 140 times reduced, 

 the spider-lines were reduced to four times the tenuity of Nobert's 

 lines — Band xix. 



In support also of the precision of the miniature, it must be 

 considered that the miniature is formed by a cone of rays of the 

 same extreme aperture as the objective, and is received by another 

 of similar large aperture. It is produced under precisely the same 

 aperture and transmitted through a similar one. 



Another argument greatly in favour of the miniature result is 

 the following experiment. If two fine wires be placed parallel 

 against the light, they cannot be separated unless the interval is 

 considerably larger than the smallest size visible at a given distance 

 by the unassisted sight. 



Least visible separation. Single pin just visible. 



Fig. 1. Fig. 2. 



In reference to this point Dr Jurin found he could discover a 

 pin stuck in a window 40 feet away from him subtending an angle 

 of two or three seconds : but if he placed two pins together he 

 could not distinguish them separated, except they were so far apart 

 as to make an angle of 40 seconds^ Taking this remarkable fact 

 into consideration, that a bright interval could not be discerned by 

 Dr Jurin unless it were ten or fifteen times larger than the objects 

 forming it, we may congratulate ourselves on the excellence of 

 modern objectives which enable the eye to discern an interval 

 only four times larger than the diameter of the web miniature, 

 reckoned at one-millionth of an inch. 



Referring once more to the formula 



\ 



2 sm a 



1 Quoted by Mr Broun, F.B.S. Proc. R. S. No. 163, p. 525. 



