154 Transactions of the Royal Microscojpical Society. 
The most peculiar difficulty of all is the existence of two sets of 
false images, viz. the one above and the other below the focal 
reality. 
This is seen beautifully in miniaturing (by the method now 
advocated) two distinct structures placed the one behind the other. 
In the case of the perforated metal the false images or eidola were 
made to exist in front of the true image. If therefore another 
structure were placed behind the first, the eidolic images of the one, 
as it were, will be mixed and confused with the real image of the 
other, and vice versa. 
The earnest microscopist, remembering the results he has 
obtained from miniatures of the sun-lit or lamp-lit globule, that 
the false image lies wholly above or below the best focal point 
according as the object-glass is over or under corrected, can have 
no difficulty in perceiving that also when viewing a duller object, 
a false image will similarly lie wholly above or below the real 
according to the corrections. 
A corollary may be drawn from these principles, viz. that in 
duplex structures the u]()per may he test seen hy throwing! the false 
image wholly helow, for then the eidola of the lower structure does 
not mingle with the true image of the upper. And conversely 
the. lower structure is best seen when the false images are wholly 
above, for then the eidola of the upper does not mingle with the 
true image of the lower. In such cases under-correction best 
displays the upper and over^correction the details of the lower 
structure immediately subjacent. 
I now search among the Quadrata diatoms, and with a half- 
inch by Wray, of three lenses, I just discover a waviness like the 
early stage of Podura definition, of somewhat irregular pattern, 
two of these objects lying exactly over each other at right angles. 
Another half-inch by the same maker, with a 0 eye-piece, sharply 
defines these dark waves, and upon entering with a D the pecu- 
liar structure of these markings becomes delicately visible. A Eoss 
J, 1851, resolves them ; but the 1870 water-lens of Powell and 
Lealand's ith now exhibits a more remarkable pattern than the 
Angulata. On applying the finest powers of definition I possess, 
these markings disappear almost entirely, and the view is wholly 
confined to the upper surface ; then alone clearly beaded. 
I have just seen a beautiful example of illusion. On examining 
this evening a fine specimen of Holler's Angulata jpleurosigma, 
I succeeded in finding* several instances where two diatoms overlay 
each other at a favourable angle. On carefully illuminating with 
a small achromatic pencil of direct rays, with a blue shade, I dis- 
* Wilh I'owcll and LcaUiiKl's dry eij^litli and 0 cye-picce about 800 diameters. 
