The capillary ripples which are produced on the water-surface by allowing water to drop on it from a jet, flit across the surface so rapidly that they cannot be followed by the eye. They may, however, be rendered visible as follows: A zinc disc, having holes in it, is arranged in front of the focussing-lens, and turned by hand or by means of a small electric motor. This disc is called a stroboscopic disc. When turned round it eclipses the light at intervals, so that the image on the screen is intermittent. If, now, one of the water-jets is adjusted so as to originate at the centre of the tank a set of diverging circular ripples, they can be projected as shadows upon the screen. These ripples move at the rate of 1 or 2 feet per second, and their shadows move so rapidly across the field of view that we cannot well observe their behaviour. If, however, the metal disc with holes in it is made to revolve and to intermittently obscure the view, it is possible to adjust its speed so that the interval of time between two eclipses is just equal to that required by the ripples to move forward through one wave-length. When this exact speed is obtained, the image of the ripples on the screen becomes stationary, and we see a series of concentric dark circles with intermediate bright spaces (see Fig. 16), which are the shadows of the ripples. In this manner we can study many of their effects. If, for instance, the jet of water is made to fall, not in the centre of the trough, but nearer one side, we shall notice that there are two sets of ripples which intersect—one of these is the direct or original set, and the other is a set produced by the reflection of the original ripples from the side of the trough. These direct and reflected ripple-shadows intersect and produce a cross-hatched pattern. If a slip of metal or glass is inserted into the trough, it is very easy to show that when a circular ripple meets a plane hard surface it is reflected, and that the reflected ripple is also a circular one which proceeds as if it came from a point, Q, on the opposite side of the boundary, just as far behind that boundary as the real centre of disturbance or origin of the ripple P is in front of it (see Fig. 17). In the diagram the dotted curves represent the reflected ripple-crests. [Illustration: FIG. 16.] If we make two sets of ripples from origins P and Q (see Fig. 18), at different distances from a flat reflecting boundary, it is not difficult to trace out that each set of ripples is reflected independently, and according to the above-mentioned rule. We here obtain a glimpse of a principle which will come before us again in speaking of æther waves, and furnishes an explanation of the familiar optical fact that when we view our own reflection in a looking-glass, the image appears to be as far behind the glass as we are in front of it. [Illustration: FIG. 17.—Reflection of circular ripples.] [Illustration: FIG. 18.]