T HE excellent paper on the plotting of the vertical photographs of the Everest Flight by Lieut. Salt in the February Journal leaves nothing to be added; the subsequent discussion raised two points, one with regard to the use of oblique photographs, and the other concerning the use of ground photogrammetry in air-photo mapping, which were perforce rather summarily dealt with. A certain amount of experience in these two directions has been gained of recent years on the North-West Frontier, so that a summary based on this experience may be of interest. The Wild photo-theodolite has been extensively used for the provision of points from which maps from vertical air photographs can be scaled and fixed in position. The final plot as it comes from the Autograph represents practically the ideal form of for air-photo maps. The plot shows the conformation of a hill-top or the relative detail round a tree and thus prevents any of the doubt which assails the compiler when faced with a trig. point labelled simple Peak or tallest tree, and a photograph showing two or perhaps even three high points or trees within the circle of probable error. The difficulties of the method are mainly in obtaining the photographic material to provide an accurate plot. In the mountainous country of the Frontier it is not always easy to find pairs of peaks or suitable ridges or spurs having sufficient command and the correct orientation and distance apart to provide suitable camera stations. When they have been found, fixing their geographical position may present difficulties. A deliberately planned triangulation is the best method of fixing the camera stations, but this is not always possible when paying a flying visit in tribal territory. Trigonometrical interpolation, though very popular, should be treated with suspicion unless it is carried out from well-proved points derived from a single accurate triangulation. When a site has had to be used which does not altogether fit the ideal conditions for the method or when a slightly doubtful interpolation has been employed to fix the geographical position of one end of the base, the difficulty is to know how far the plot is reliable, in particular in the areas to the flanks or in the distance away from fixed points where inaccuracies in the setting data cause rapidly increasing errors in the plot. On the whole, the photogrammetric method is very suitable, and as the extra photographic equipment is very light, the observer going out to fix for air surveys who takes the phototheodolite is none the worse off if he cannot find suitable bases for photogrammetry, for the theodolite is a normal one and can be used for fixing triangulated control in the normal way. Oblique air photographs are employed on the Frontier for mapping areas of which it is not possible or convenient to take vertical photographs. The inability to take vertical photographs may be due to political considerations rendering flying over the area undesirable or to the ground being higher than can be photographed from the flying height for aircraft with photographic equipment, which seldom exceeds i8,ooo feet above sea-level. Where possible,