experienced during the trajectory. All  information was then fed into the control computer—attitude information from the ST-90, angle-of-attack signals from the angle-of-attack meter, and  guidance signals from the guidance computer—which assimilated the information and signaled the hydraulic actuators for such actions as swiveling the  main and vernier engines or operation of the jet nozzles 71 . The target date to accomplish firm G&C plans was November 1956, and this was met. Specifications were furnished to the Ford Instrument Company for the  delivery of the ST-90 in January 1957. A testing program was then under way until October 1957, when JUPITER Missile 3 was used to flight test the ST-90 and related components. It worked 72 . Propulsion System Perhaps one of the most difficult of the development problems to resolve was in the propulsion area, an item over which ABMA had the least control. The heart of  the system was the main engine, a component commercially produced by NAA. As earlier mentioned, the supply source was saturated in satisfying the demands  of four long-range missile systems. Since the Air Force was in charge of the development of three, ABMA's requirements had considerable difficulty in being met.  Also, ABMA thought that the NAA engine was only marginal in satisfying JUPITER needs, yet there was little opportunity for NAA to enter into a large-scale  research program. To ABMA's way of thinking, the over-all engine program should involve one _____________________________

71. Ibid.; JUP Dev Plan, FY 58, 29 Sep 56, Hist Off files. 72. JUP Prog Rpt for Nov 56, 8 Dec 58; JUP Prog Rpt for Oct 57, 8 Nov 57, Hist Off files. 73. Study, Liquid Rkt Engine Dev Prog, 19 Jul 56, Hist Off files.