Abstract
In 1970, Hyland, Rowlands, and Cummings suggested that one could obtain the condensate fraction n0(T) from careful measurements of the static pair-correlation function g(r, T) in the superfluid and normal phases of He4. Their arguments are critically analyzed by going back to the general structure of S(Q) and S(Q, ω) predicted by the field-theoretic analysis of a Bose-condensed liquid. It is shown that at low temperatures, the single-particle correlation functions, which Hyland et al. argue are negligible, in fact make a major contribution to g(r, T) in the region r4 Å.