94 SUMMARY OF CURRENT RESEARCHES RELATING TO 
form the keynote for artificial illumination in every subsequent text- 
book. They are repeated by Carpenter, who, after giving directions as 
to centering and focusing the image of the lamp-flame on the object, 
says that “ the direction of the mirror should then be sufficiently 
changed to displace the image and to substitute for it the clearest light 
that can be obtained.” Further, he recommends that while with day- 
light the condenser should be used in focus, with lamplight it should 
be somewhat racked down. From this I gather that Dr. Carpenter’s 
best artificial illumination is oblique light out of focus. Of course the 
actual fact is that daylight focus is not nearly so important as lamplight. 
In illustration of another kind of mistake, as late as ten years ago it 
w r as recommended that the diaphragm be placed above the condenser 
as giving a better result than when placed below. 
Of course the optical effect is precisely the same, the only thing is 
that the diaphragm below the condenser is much more readily manipu- 
lated and is much more likely to be accurate in centering, unless the one 
above be of the cap form. To change a cap diaphragm necessitates 
either the removal of the slide or the condenser, and all for no purpose. 
The next idea was worse, viz. the calotte diaphragm. This being 
fixed to the stage and not to the substage, gave as often as not excentric 
pencils. Whatever diaphragm is used it obviously must be centered to 
the condenser and must move with it, otherwise it will be put out of 
centre during the operation of centering the condenser to the optic axis 
of the objective. 
Further, the calotte diaphragm is useless for ordinary illumination 
without a condenser, as the apex is not the proper place to cut the 
illuminating cone. The proper place, therefore, for a diaphragm, when 
no condenser is used, is some distance from the object, and when a 
condenser is used, is at the back of the combination. Further, when a 
diaphragm is above the condenser the apertures become almost microscopic 
in size, and a very small difference between them will make a con- 
siderable difference in their effect ; but when they are placed behind the 
combination they may be larger, and it becomes more easy to graduate 
them in accordance to any desired effect. 
Again, it is a fallacy to suppose that a Kelner eye-piece is superior 
to a condenser as an illuminator for high powers. 
A Kelner eye-piece, if a C, is only 1 in. in power, and has a small 
angular aperture somewhat less than 45°, therefore it cannot possibly 
give a cone at all comparable with that from a most elementary con- 
denser. It might be used as a substage condenser for low powers, but 
from its small aperture it would hardly give a good dark-ground illumi- 
nation for a 1-in. objective. 
With regard to low power condensers, the Webster (as designed by 
Webster) is the proper form. There are many so-called Webster con- 
densers in existence which are on a totally wrong principle. The right 
kind of Webster has a single front lens and a back lens composed of a 
plano-concave flint and a crossed convex crown, the cemented surfaces 
having a deep curve to overcorrect the lens. The other kind, which is 
quite wrong, has an achromatized front and a single back ; it is merely 
done for cheapness, as small achromatic pairs are not so expensive as 
large ones, and the back lens of a condenser is always larger than the front. 
