The Gallium Anomaly, Recounted: The BEST Two-Zone Result, Its Sterile-Neutrino Price, and the Cross-Section Correction That Would Be Required

Gallium radiochemical detectors measure the capture of electron neutrinos on 71Ga through the reaction 71Ga(nu_e, e-)71Ge. Their solar-neutrino rates are now understood: the weighted average of the SAGE, GALLEX and GNO results, 66.1 +- 3.1 SNU, agrees with the oscillation-corrected standard solar model prediction of about 66 SNU. A separate discrepancy persists for artificial neutrino sources of known strength. The BEST experiment placed a 3.414 MCi 51Cr source at the centre of two nested gallium volumes and measured ratios of observed to predicted 71Ge production rates of R_in = 0.791 +- 0.05 and R_out = 0.766 +- 0.05. Combining the two zones as independent measurements gives R = 0.779 +- 0.035, a deficit of 22.2 percent lying 6.3 standard deviations from unity; a conservative combination of all six gallium source calibrations that accounts for the correlated cross-section uncertainty gives R = 0.82 +- 0.05, or 3.6 standard deviations. Interpreted as averaged oscillations to a sterile neutrino in a simple 3+1 scheme, the combined BEST ratio requires sin^2 2theta = 0.44 +- 0.07, a very large mixing, with Delta m^2 at or above about 1 eV^2 so that the oscillation length is comparable to or shorter than the metre scale of the target. Interpreted instead as a cross-section error, the anomaly requires the 51Cr capture cross section to be reduced by 22.2 percent, from the Bahcall value of 5.81 x 10^-45 cm^2 to 4.52 x 10^-45 cm^2. That correction is 6.1 times the published upper uncertainty of the Bahcall cross section (3.6 percent) and 4.5 times the upper uncertainty of the 2023 revision by Elliott and collaborators (4.9 percent). Even setting the entire excited-state contribution to zero raises the combined ratio only to 0.82. The anomaly is therefore neither a small effect nor one that a cross-section adjustment within its stated errors can remove.