ption,[2] in which case there is no recovery and rebound. Thus a
drop of milk falling on to smoked glass will, if the height of fall and
size of drop are properly adjusted, give forms very similar to those
presented by a drop of mercury. The whole course of the phenomenon
depends, in fact, mainly on four quantities only: (1) the size of the
drop; (2) the height of fall; (3) the value of the surface tension; (4)
the viscosity of the liquid.

The next series of drawings illustrates the splash of a drop of water
falling into water.

In order the better to distinguish the liquid of the original drop from
that into which it falls, the latter was coloured with ink or with an
aniline dye, and the drop itself was of water rendered turbid with
finely-divided matter in suspension. Finally drops of milk were found to
be very suitable for the purpose, the substitution of milk for water not
producing any observable change in the phenomenon.

In Series II. the drop fell 3 inches, and was 1/5 inch in diameter.

[In most of the figures of this and of succeeding series the central
white patch represents the original drop, and the white parts round it
represent those raised portions of the liquid which catch the light. The
numbers under each figure give the time interval in seconds from the
occurrence of the first figure, or of the figure marked [Tau] = 0.]

SERIES II.

_The Splash of a Drop, followed in detail by Instantaneous
Illumination._

Diameter of Drop, 1/5 inch. Height of Fall, 3-1/5 inches.

[Illustration: 1

[Tau] = 0 sec.]

[Illustration: 2

[Tau] = 0 sec.]

[Illustration: 3

[Tau] = .0097 sec.]

[Illustration: 4

[Tau] = .0392 sec.]

[Illustration: 5

[Tau] = .0392 sec.]

[Illustration: 6]

[Illustration: 7

[Tau] = .0979 sec.]

[Illustration: 8

[Tau] = .1095 sec.]

[Illustration: 9

[Tau] = .167 sec.]

It will be observed that the drop flattens itself out somewhat, and
descends at the bottom of a hollow with a raised beaded edge (Fig. 2).
This edge would be smooth and circular but for the instability which
causes it to topple into drops. As the drop descends the hollow becomes
wider and deeper, and finally closes over the drop (Fig. 3), which,
however, soon again emerges as the hollow flattens out, appearing first
near, but still below the surface (Fig. 4), in a flattened, lobed form,
afterwards rising as a column somewhat mixed with adherent water, in
which traces of the lobes are a