they are hemispheres, and the theory is also confirmed by a microscopic photograph from Lackerbauer’s collection, magnified 3,000 diameters, in which a black central point is seen in the centre of each circle, a certain reflection of the luminous source reproduced in the focus of each of the small demi-spheres which constitute the ornament of the algæ. The microscope, which has progressively shown first the streaks, then the hexagons, and then the round points, will surely clear up the point some day or other. [Illustration: Fig. 133.—Another figure of the same kind.] [Illustration: Fig. 134.—Third figure.] Mr. Silvanus P. Thompson, Professor of Physics at University College, Bristol, has recently presented the French Society of Physical Science with a curious example of optical illusion, the true cause of which is not clearly known, but which we may compare with other facts made known some time ago, of which no precise explanation has been given. Let us first consider in what the effect discovered by Mr. S. P. Thompson consists, according to the description that has been given of it by M. C. M. Gariel; the illustrations here given will also allow of our verifying the truth of the statements. [Illustration: Fig. 135. Fig. 136. Mr. Thompson’s optical illusion. Give a circular movement to these figures, and the circles will appear to turn round.] The first illustration consists of a series of concentric circles of about the width of a millimetre, separated by white intervals of the same size (fig. 135). These dimensions are not absolute; they vary with the distance, and may even be a few inches in width if it is desired to show the phenomenon to a rather numerous auditory. If we hold the design in the hand, and give it a twirl by a little movement of the wrist, the circle appears to turn round its centre, and the rotation is in the same direction, and is equally swift; that is to say, the circle appears to accomplish a complete turn, whilst the cardboard really accomplishes one in the same direction. For the second effect we draw a dark circle, in the interior of which are placed a number of indentations at regular intervals (fig. 136). Operating in the same manner as described above, this notched wheel appears to turn round its centre, but this time in a different direction from the real movement. In this, however, as in the other design, the effect is more satisfactory if we do not look directly at it; the movements also are particularly striking in combinations such as that represented in fig. 137, in which the multiplicity of circles does not allow us to fix one specially. We may add that the same effects may be obtained with eccentric wheels, or even with other curves than circles. By means of a photograph on glass, Mr. Thompson has been able to reflect these designs on a screen where they were obtained on a large scale; a circular movement was communicated to the photographic plate, so that