pon arena of Immersion Lenses. 141 
inch. Oblique solar rays, whether parallel or not, exaggerate both 
the slightest chromatic errors of the glasses, unless mono-chromatic 
rays be employed, as well as diffraction phenomena. 
Spherical Aberration Geometrically Displayed.—The compara- 
tive initial spherical aberrations for pnewmo and hydro objectives or 
immersion lenses as affecting the nascent pencils, may be con- 
veniently drawn by means of the preceding Tables, on a scale of 
enlargement one hundred times the real size. 
As, however, the deviation between Canada balsam and glass is 
too small to be delineated (see Table III.), I have neglected it in 
the accompanying drawings. In each case a nascent pencil pro- 
ceeding from a brilliant illuminated particle immersed in the 
balsam, is traced in its deviations for different directions for the 
nascent rays from their origin to their points of immergence into 
the front surface of the objective, which is here supposed to be 
plain whatever form of lenses is used for the construction of the 
front set. 
In the water lens the initial aberrations denoted by intersections 
at a, @', &, @, @; .... are geometrically shown to be very 
much less than when an air film is employed and denoted by 
My, B' Ly, Uy, Ls, %, ... (See Fig. 1 and Fig. 2, Plate LX.) 
The adaptation of a ‘“ water lens” (which is simply a hemi- 
spherical plano-convex lens, of a focal length nearly one-half deeper 
than the proposed immersion objective) to an old-fashioned combi- 
nation made on the Lister principle, produces a very inferior 
objective to the new system adopted by Messrs. Powell and Lealand. 
Optically speaking a great many plans may be employed to produce 
a good result. The performance of some of the American and 
European glasses on the immersion system, as compared with our 
best English glasses, requires a more thorough imvestigation and 
testing than perhaps can be given by Nobert’s lines. 
I have not entered here upon the question of the comparative 
dispersive powers of water and glass.* I have not yet been able to 
ascertain the dispersive power or what is really the kind of glass 
employed for the “cover.” Its influence upon definition cannot be 
too strongly insisted upon. The chromatic errors of definition (it 
may here be cursorily remarked) secure rapid transformations from 
five distinct sources of change. 
(1.) The difference in the index of refraction of the glass of the 
“cover” in different specimens. 
(2.) The difference in the thickness of the cover, which no 
* The subject of chromatic dispersion is reserved for a separate article, if it 
can be made suitable to this Journal. 
