gradually disappears, or dissolves, and the second picture gradually appears till the first vanishes and the second occupies its place. A great deal of ingenuity is displayed by the Parisian artists in the composition of these pictures, and the exhibition of them, either in small portable instruments held in the hand, or placed on the table, or on a great scale, to an audience, by means of the oxygen and hydrogen light, never fails to excite admiration.

The pictures thus exhibited, though finely executed, have only that degree of relief which I have called monocular, and which depends on correct shading and perspective; but when the dissolving views are obtained from binocular pictures, and have all the high relief given them by their stereoscopic combination, the effect must be singularly fine.

Very interesting and amusing effects are produced by interchanging the right and the left eye pictures in the stereoscope. In general, what was formerly convex is now concave, what was round is hollow, and what was near is distant. The effect of this interchange is finely seen in the symmetrical diagrams, consisting of white lines upon black ground, such as Nos. 1, 5, 9, 12, 18 and 27 of the Parisian set; but when the diagrams are not symmetrical, that is, when the one half is not the reflected image of the other, such as Nos. 26, &c., which are transparent polygonal solids, formed as it were by white threads or wires, no effect, beyond a slight fluttering, is perceived. As the right and left eye pictures are inseparable when on glass or silver plate, the experiments must be made by cutting in two the slides on Bristol board. This, however, is unnecessary when we have the power of uniting the two pictures by the

convergency of the optic axes to a nearer point, as we obtain, in this case, the same effect as if we had interchanged the pictures. The following are some of the results obtained in this manner from well-known slides :

In single portraits no effect is produced by the interchange of the right and left eye pictures. If any loose part of the dress is in the foreground it may be carried into the distance, and vice versa. In one portrait, the end of the hat-band, which hung down loosely behind the party, was made to hang in front of it.

In pictures of streets or valleys, and other objects in which the foreground is connected with the middle-ground, and the middle-ground with the distance, without any break, no effect is produced by the interchange. Sometimes there is a little bulging out of the middle distance, injurious to the monocular effect.

In the binocular picture of the Bridge of Handeck, the Chalet in the foreground retires, and the middle distance above it advances.

In the picture of the sacristy of Notre Dame, the sacristy retires within the cathedral.

In the Maison des Chapiteaux at Pompeii, the picture is completely inverted, the objects in the distance coming into the foreground.

In the Daguerreotype of the Crystal Palace, the water in the foreground, with the floating plants, retires and takes an inclined position below a horizontal plane.

In the binocular picture of the lower glacier of Rosenlaui, the roof of the ice-cave becomes hollow, and the whole foreground is thrown into a disordered perspective.

In Copeland's Venus, the arm holding the bunch of

grapes is curiously bent and thrown behind the head, while the left arm advances before the child.

In the picture of the Greek Court in the Crystal Palace, the wall behind the statues and columns advances in front of them.

The singular fallacy in vision which thus takes place is best seen in a picture where a number of separate articles are placed upon a table, and in other cases where the judgment of the spectator is not called upon to resist the optical effect. Although the nose of the human face should retire behind the ears yet no such effect is produced, as all the features of the face are connected with each other, but if the nose and ears had been represented separately in the position which they occupy in the human head, the nearer features would have retired behind the more remote ones, like the separate articles on a table.

We shall have occasion to resume this subject in our concluding chapter on the fallacies which take place in viewing solids, whether raised or hollow, and whether seen by direct or inverted vision.

CHAPTER XV.

ON THE PRODUCTION OF STEREOSCOPIC PICTURES FROM A SINGLE PICTURE.

THOSE who are desirous of having stereoscopic relievos of absent or deceased friends, and who possess single photographic portraits of them, or even oil paintings or miniatures, will be anxious to know whether or not it is possible to obtain from one plane picture another which could be combined with it in the stereoscope; that is, if we consider the picture as one seen by either eye alone, can we by any process obtain a second picture as seen by the other eye? We have no hesitation in saying that it is impossible to do this by any direct process.

Every picture, whether taken photographically or by the eye, is necessarily a picture seen by one eye, or from one point of sight; and, therefore, a skilful artist, who fully understands the principle of the stereoscope, might make a copy of any picture as seen by the other eye, so approximately correct as to appear in relief when united with the original in the stereoscope; but the task would be a very difficult one, and if well executed, so as to give a relievo without distortion, the fortune of the painter would be made.

When the artist executes a portrait, he does it from one point of sight, which we may suppose fixed, and corresponding with that which is seen with his left eye. If he takes another portrait of the same person, occupying exactly the same position, from another point of sight, two and a half inches to the right of himself, as seen with his right eye, the two pictures will differ only in this, that each point in the head, and bust, and drapery, will, in the second picture, be carried farther to the left of the artist on the plane of representation. The points which project most, or are most distant from that plane, will be carried farther to the left than those which project less, the extent to which they are carried being proportional to the amount of their projection, or their distance from the plane. But since the painter cannot discover from the original or lefteye plane picture the degree of prominence of the leading points of the head, the bust, and the drapery, he must work by guess, and submit his empirical touches, step by step, to the judgment of the stereoscope. In devoting himself to this branch of the art he will doubtless acquire much knowledge and dexterity from experience, and may succeed to a very considerable extent in obtaining pictures in relief, if he follows certain rules, which we shall endeavour to explain.

If the given portrait, or picture of any kind, is not of the proper size for the stereoscope, it must be reduced to that size, by taking a photographic copy of it, from which the right-eye picture is to be drawn.

In order to diminish the size of the diagram, let us suppose that the plane on which the portrait is taken touches the back of the head, and is represented in section