Chopter 7 CONCLUSIONS pictures that would otherwise havebeen useless for measurement purposes. brated cameras ‘have similarly performed well. erent The data obtained for Project 9.1 are by far the best yet, although they extend only over the first 10 minutes; little information was obtained for later times. - Good measurements were made forfive of the six Castle shots. Shot 3 was fired under weather conditions stich that none of the cameras saty the cloud. On the other shots, natural cloud interference turned out to be less serious than had been expected. The utility of high~flying aircraft for these purposes is confirmed, as is also the de~ sirability of using distant ground-level cameras. _ ‘The A-28 stabilized mounts have performed brilliantly and have saved many The cali- Many photographs, whose only refer- ence level is the camera fiducial-mark system, were measured successfully. This gives incréased confidence in the results and in the methods by which they were ob~ tained. 38 SECRET TOT Ne ape ANNEA RRM E ahaee