The Galileo spacecraft encountered the Earth once on December 8, 1990 (Earth I), and again on December 8, 1992 (Earth II). These flybys provided excellent opportunities to evaluate the performance of the energetic particle detector (EPD) and establish analysis procedures in a relatively well-known environment. Further, because Galileo's Earth flyby trajectories were very rapid and nearly radial, the radiation belt measurements provided an excellent snapshot of trapped radiation. The EPD data agree with and extend the familiar radiation belt empirical models established by the National Space Science Data Center. Because of the rapid flyby and the 20-s spin period of Galileo, great care had to be taken to remove time aliasing from the pitch angle distributions. Large anisotropies were also present due to spatial density gradients. Spherical harmonics were fitted to the pitch and phase distributions in order to remove the effects of time aliasing and density gradient anisotropies and obtain fluxes from which phase space densities could be computed. The phase space density was calculated at pitch angles and energies that conserve the first and second adiabatic invariants. The extracted phase space density was compared with that computed from the AP8 model of the stably trapped proton fluxes. Earth I and Earth II observations and the AP8 model were compared for values of the first adiabatic invariant at 10.0, 15.0, 20.0, 25.0 30.0, 35.0, 40.0, 45.0, and 50.0 MeV/G and of the second adiabatic invariant at 0.10, 0.15, 0.20, 0.25, and 0.30 G0.5 RE. Results show that for lower values of the first and second adiabatic invariants the AP8 values slightly exceed the observations, and for higher values of the first two adiabatic invariants the AP8 values are slightly lower than the observations; and in the middle range of invariants there is relatively good agreement between AP8 and observations. ¿ 1997 American Geophysical Union