Abstract Extrapolations of line-of-sight photospheric field measurements predict radial interplanetary magnetic field (IMF) strengths that are factors of ∼2–4 too low. To address this open flux problem , we reanalyze the magnetograph measurements from different observatories, with particular focus on those made in the saturation-prone Fe i 525.0 nm line by the Mount Wilson Observatory (MWO) and the Wilcox Solar Observatory (WSO). The total dipole strengths, which determine the total open flux, generally show large variations among observatories, even when their total photospheric fluxes are in agreement. However, the MWO and WSO dipole strengths, as well as their total fluxes, agree remarkably well with each other, suggesting that the two data sets require the same scaling factor. As shown earlier by Ulrich et al., the saturation correction δ −1 derived by comparing MWO measurements in the 525.0 nm line with those in the nonsaturating Fe i 523.3 nm line depends sensitively on where along the irregularly shaped 523.3 nm line wings the exit slits are placed. If the slits are positioned so that the 523.3 and 525.0 nm signals originate from the same height, δ −1 ∼ 4.5 at the disk center, falling to ∼2 near the limb. When this correction is applied to either the MWO or WSO maps, the derived open fluxes are consistent with the observed IMF magnitude. Other investigators obtained scaling factors only one-half as large because they sampled the 523.3 nm line farther out in the wings, where the shift between the right- and left-circularly polarized components is substantially smaller.