174 
image (ascertained by reducing the aperture of the micro- 
scope ') when the objective was undercorrected, and above it 
when overcorrected. Brilliant images of glittering particles 
of mercury scattered on black cloth nearly vertically illumi- 
nated, fine gauze 80 meshes to the inch perforated metal, 
gold-leaf displaying against a brilliant light immeasurably 
small perforations exposed on arich malachite green ground,” 
were submitted to be examined in miniature as test-objects. 
From a variety of experiments of this kind the following data 
were arrived at, to guide preliminary observations :— 
That when any well-defined structure is viewed by the best 
microscopes, there exist eidola® or false images on each side 
of the best focal point. 
That they are placed principally above or principally below 
the focal point of centrical pencils, according as the glasses 
are over or undercorrected ; and that for a semgle stratum 
sufficiently thin, these eidola are nearly symmetrically ex- 
hibited on both sides of the best focal point only when the 
compensations are perfectly balanced. 
It follows from these results that when a structure consists 
of two superimposed strata, in such close contiguity as to 
come within the optical limits of the ezdola, the false images 
of the lower stratum are liable to be confused and commingled 
with the true image of the upper stratum when the objective 
is oOvercorrected, and when it is undercorrected the false 
images of the upper are confused with the true of the lower 
stratum. ‘These coincidences of eidola with true focal images 
may in both cases equally delude the observer. 
The next question was the most favorable distribution of 
the elements of magnifying-power. According to a well- 
known optical principle, it seemed desirable to bend the rays 
by less sudden refractions. It is a peculiar result that when 
the incident and emergent pencils are equally bent so as to be 
equally inclined to the axis of an equi-conyex lens, that then 
only is the aberration a minimum. 
The effects of different distribution of power are well shown 
by the following experiments, in both of which the same 
amplification was employed of 400 diameters. 
1 The true image is at once seen by reducing the aperture; for this pur- 
pose a system of circular stops was applied to the microscope at the part 
where the objective is attached, admitting an instantaneous change in the 
aperture, and showing remarkable effects produced by change in the excentric 
aberration. Its mode of attachment is shown at 8, fig. 1, Plate IX, where 
it is marked aberrameter. 
? The gold-leaf is mounted on a slide in the ordinary way, and exhibits 
interesting and instructive phenomena. 
3 Eidwdov. 
