dicular to the surface of the picture S S, as at B, the picture appears a circle. I admit that this defect only occurs when the field of view of the lens is very large, and the balls are situated very near its rim.
A photographer brought to the author the picture of a castle having a row of statues in front of it, which he had taken with a lens having a large field of view. Singularly, the heads of the statues toward the margin became continually broader, and similarly their bodies; and the slim Apollo-Belvedere, who unfortunately stood on the very edge of the margin, had such full-blown cheeks and so protuberant a paunch that he looked like Dr. Luther.
But, quite independently of these considerations, there is another point that must materially affect the accuracy of photographic representation. Photography generally gives the light parts too light, and exaggerates the dark shadows. This is a fundamental error which is associated with their very nature, and which it is very difficult to avoid. It is seen in the most evident manner in taking objects lighted by a brilliant sun; for example, a statue. If the exposure is short, a detailed picture is obtained of the light side, but the shady side is a black daub or blotch. If the exposure is long, the shady side is full of detail, but the light side exposed too much, and so thickly covered that the details are wanting in it. Hence photographers are often driven to subterfuges if they wish to obtain a correct picture; they are obliged to mitigate the contrasts—to make the light more toned down, and the shades lighter than painters are wont to make them. The latter often exclaim when they see the photographic exposure of a model, and wonder if the picture will be correct. And no doubt, in the case of landscapes and architecture, the results are not always satisfactory.
The author once took a photograph of the interior of a laboratory. It presented the appearance of an ordinary vaulted hall. All was quite excellent. The tables, stones, retorts, lamps, etc., were all seen, only the vaulted ceiling was quite dark. New attempts were made, with exposures of twenty, thirty, or forty minutes. At length a trace of the vault appeared; but now the objects in the vicinity of the window were suffering from too much exposure; that is, they had become as white as if they had been snowed over. This circumstance of photography exaggerating the dark parts appears again in very simple matters, such as the reproduction of copper-plates. A photographer once reproduced a painting of Kaulbach's “Battle of the Huns.” He produced a charming photograph, but the city in the background appeared too thick and black, and not sufficiently toned off. The customer refused the photograph and demanded another. The photographer made another attempt, giving a longer exposure, and now the distance appeared softened down; but, unfortunately, the objects close at hand, which had to appear black and clear, turned out gray. In the end, the photographer escaped from the difficulty by negative
