Inferences of the interplanetary magnetic field using polar cap magnetograms have been made by L. Svalgaard for the years 1970--1972 and by both G. R. Heckman and S. M. Mansurov for 1972. In view of reports of contamination of earlier such indices by geomagnetic activity we examine the geomagnetic activity on days incorrectly inferred. In 1972, toward days, incorrectly inferred to have away polarity, have an average Ap index approximately 20% less than that on the average toward day. Away days, incorrectly inferred to have toward polarity, have no significant geomagnetic bias. Since the number of incorrect inferences is low in 1972, the Ap index averaged over all inferred toward days is only 2--6% greater, depending on observer, than that on inferred away days. While the success rate for inferring toward days approaches the maximum expected (84--89%, depending on observer), the success rate for away days is 3--7% lower. We note that although all three observers are more successful in identifying toward days, additional observations are required to assess the significance of this effect. We conclude that for many statistical studies the present-day inferences of interplanetary polarity are adequate substitutes for in situ observations.