The rocket experts of the Jet Propulsion Laboratory of the California Institute of Technology began their study of single-stage rockets for the Bureau of Aeronautics in December 1945 and completed it by July 1946.* They wrote six reports, with the earliest appearing on 3 January 1946. The study was based on three assumptions: (1) the orbiting vehicle would be a single-stage liquid propellant rocket, (2) the propellants would be liquid hydrogen and liquid oxygen, and (3) the exhaust velocity of the rocket would be 3240 meters per second at sea level and 4320 at very high altitude. The rocket performance values were furnished by David Young of Aerojet. With these assumptions, the JPL men sought to determine the most suitable trajectory and designs for minimum initial-to-final mass ratio.
The final report, appearing in July, stated that if the single-stage rocket was launched from sea level, the initial-to-final mass ratio must be 8.70; if launched from a high mountain (4300 m), the mass ratio could be decreased slightly.12 These results made clear to the Bureau of Aeronautics what the next steps should be: expand Aerojet's work on the experimental performance of hydrogen and oxygen and get improved weight estimates for rocket engines and vehicle structures. The latter called for the experience of an airframe manufacturer. The JPL-GALCIT study also pointed out that the mass ratio requirements for orbiting a satellite could be greatly reduced if multiple-stage rockets were used.
Apparently as a derivative of these classified military studies, Frank Malina and Martin Surnmerfield reported on the problem of escape from earth by rocket in August 1946, and Malina presented the results at the Sixth International Congress for Applied Mechanics at Paris in September. They made a strong case for using hydrogen-oxygen. A multistage rocket using nitric acid and aniline (a combination in use at that time) was considered too large to be practical even for a 5-kilogram payload. They concluded  that a multistage rocket of reasonable size using liquid hydrogen and liquid oxygen could carry a payload of 45 kilograms and was within engineering feasibility. They assumed an exhaust velocity of 3660 meters per second for hydrogen-oxygen, five stages, and a gross mass of 37 600 kilograms. The authors also pointed out the advantages of using hydrogen as the working fluid with heat supplied by a nuclear reaction. Potential exhaust velocities were as high as 11 400 meters per second-close to the vehicle velocity needed for escape from the earth's gravitational field.13
* Participating in the study were W. Z. Chien, Lt. Comdr. E. C. Sledge, Lt. Comdr. G. G. Halverson, J. V. Charyk, and H. J. Stewart. Stewart wrote the final report.