Abstract
It is shown that a wave function which is a simple product of two-particle correlation functions gives a smaller short-range three-body correlation energy (calculated as the ground-state expectation value of the kinetic energy and short-range part of the interactions) than the approximate Faddeev wave function used in the recent work of Bethe. Thus for a plausible nucleon-nucleon interaction, the short-range three-body correlations contribute less than 1 MeV per particle to the energy of nuclear matter.