0·83 of a dioptre, and for this experiment I have chosen one of the lenses which is very nearly the same strength. Of course, the effect is magnified, as the camera is larger than the eye. But the eye can recognize differences infinitely more minute than those which it is possible to reproduce here, and the physiological distortion is even greater than that which it is possible to represent on the printed page. A picture taken with such a combination of lenses is shown in Fig. 2. In this it will be noticed that while the vertical lines are all clearly marked the horizontal ones are blurred and indistinct. If, now, for any reason the globe of the eye has the same distortion from side to side instead of from above downward, this can be imitated by placing the cylindrical lens before our camera with the axis vertical instead of horizontal. The result is shown in Fig. 3. The effect in this case is to blur the lines which before were clear, and make clear those which were blurred. For example, the horizontal line extending
Fig. 2.
along the top of the wall of the terrace, and even some of the strata of the stones, can be distinguished, and the horizontal lines in the building itself are also well defined. On the contrary, the vertical lines are blurred. The pillars at the top of the tower and on the dome itself are all indistinct, while, as a whole, the build-
