tracted. If the change in the current's intensity occur gradually,
there will be a moment in which the cylinders will remain in
equilibrium.

[Illustration: FIG. 4.--EDELMANN'S QUADRANT ELECTROMETER.]

Prof. Zenger's differential photometer that we shall finally cite is
an improvement upon Bunsen's. In the latter the position of the
observer's eye not being fixed, the aspect of the spot changes
accordingly, and errors are liable to result therefrom. Besides,
because of the non-parallelism of the luminous rays, each of the two
surfaces is not lighted equally, and hence again there may occur
divergences. In order to avoid such inconveniences, Prof. Zenger gives
his apparatus (Fig. 10) the following form: The screen, D, is
contained in a cubical box capable of receiving, through apertures,
light from sources placed upon the two rules, R and R'. A flaring
tube, P, fixes the position of the eye very definitely. As for the
screen, this is painted with black varnish, and three vertical
windows, about an inch apart, are left in white upon its paper. Over
one of the halves of these parts a solution of stearine is passed. To
operate with the apparatus, in comparing two lights, the central spot
is first brought to invisibility, and the distances of the sources are
measured. A second determination is at once made by causing one of the
two other spots to disappear, and the mean of the two results is then
taken. As, at a maximum, there is a difference corresponding to 3/100
of a candle between the illumination of the two neighboring windows,
in the given conditions of the apparatus, the error is thus limited to
a half of this value, or 2 per cent. of that of one candle.

[Illustration: FIG. 5.--WILD'S APPARATUS FOR STUDYING MAGNETIC
VARIATIONS.]

Among the apparatus designed for demonstration in lecture courses, we
remarked a solenoid of Prof. Von Beetz for demonstrating the
constitution of magnets (Fig. 11), and in which eight magnetized
needles, carrying mica disks painted half white and half black, move
under the influence of the currents that are traversing the solenoid,
or of magnets that are bought near to it externally. Another apparatus
of the same inventor is the lecture-course galvanometer (Fig. 3), in
which the horizontal needle bends back vertically over the external
surface of a cylinder that carries divisions that are plainly visible
to spectators at a distance.

[Illustration: FIG. 6.--ELECTROMETER FOR ATMOSPHERI