It is of primary importance that the picture be held in the right position. Not only must the observer imagine himself looking directly down on the scene but he must hold the photograph in the position in which experience has shown that the image appears the most natural. Otherwise a depression will appear as an elevation and an elevation as a hollow. It is a well-known fact that in telescopic photographs of the moon the craters appear like hollows when the print is held in one position and like elevations when the position is reversed. Experience shows that if the print is held _so that the shadows fall toward the observer_ the objects appear natural. The reason is that the observer sees only those shadows that are caused by light falling towards him. Consequently, the only interpretation that the brain can give to shadows on a photograph is that they are cast by an elevation between the eye and the light. In a picture, therefore, in which shadows fall away from the eye instead of towards it valleys are seen as hills and hills as valleys. In the northern hemisphere this prescribed orientation conflicts with the convention of placing the north side of a map at the top of the page and also with the modern shaded map on which the light is represented as coming from the upper left, or northwest, corner of the map. FAILURE OF AIR PHOTOGRAPHS TO SHOW RELIEF, AND MEASURES TO REMEDY THIS DEFECT In photographs taken from the ground the lights and shadows are such that a high degree of naturalness is possible. But objects seen from directly above, and even those viewed obliquely, though to a lesser degree, are illuminated so uniformly that photographs of them are apt to appear flat. To some extent this has been overcome by the use of extra-sensitive emulsions, special ray filters, and printing papers adapted for accentuating contrast. Many of the photographs used in this book did not allow satisfactory reproduction till the contrast of the negatives was greatly increased by the arts of the photographic laboratory. But, even at its best, no photograph taken vertically affords an adequate idea of the height of hills or the depth of hollows. Only shadows that are particularly well defined can be distinguished as shadows, while small elevations and depressions affect the negative no differently than a difference in marking or color. In military defenses, if the mere surface of the camouflage is sufficiently realistic, the ordinary camera is even more easily deceived than the human eye. It is a well-known fact that man and other animals of the higher order see objects in relief, within a certain range of vision, because the eyes convey to their respective retinas slightly different images of the same object which the brain combines into a relief image. The stereoscopic camera has long been used for the same purpose. Its principle, with certain adaptations that need not be discussed here, has been to some