Equatorial observations by Explorer 45 of ring current ions during a storm recovery phase show pitch angle distributions and decay rates inconsistent with proton charge exchange with neutral hydrogen. This inconsistency, together with the assumption that a continual strong proton source to L=3 throughout the recovery phase is physically unreasonable, implies that the recovery phase ring current ions at L≲4 and energies of ≲50 keV are not dominated by protons. He+ is an attractive alternative. Sharp et al. (1976a, b, 1977) presented mass spectrometer measurements of precipitating ions obtained at low altitudes during the same period. Unfortunately, fluxes at L≲4 were generally below the level of detectability. However, during a 5-hour period of enhanced magnetic activity ~16 hours into the storm recovery, barely detectable H+ and O+, but no He+, were observed in a small region near L=3. If it is assumed that the ion precipitation mechanism does not favor H+ and O+ over He+, the lack of detectable He+ implies that during the enhanced activity the ions were not dominated by He+ on those flux tubes from which the H+ and O+ were precipitating. Sharp et al. suggested a source of H+ and O+ to L=3 during this period as an explantion for thier observations. The equatorial observations suggest to me an alternative explanation to that of Sharp et al.: near local midnight during the storm recovery phase and during the period of enhanced activity the magnetic field lines labeled L=3 at low altitudes by a magnetic field model are extended so as to map to field lines labeled L?3.7 at the equator. This explanation implies a possible equatorial source of H+ and O+ to L=3.7 but not to L=3.0 and does not preclude the recovery phase ≲50-keV ions from being dominated by He+ for L≲3.7, even during the period of enhanced activity. It is supported by (1) perceptible equatorial flux increases at L=4 but not L=3.5 and L=3.0, (2) the consistency of the precipitating ion energies with the energies expected near L=3.7 from wave-ion interactions, and (3) the expansion of the equatorial plasmapause to L?3.8 during this time.