Mursula et al. [2011] (MTL11) suggest that there is a 22-year variation in solar wind activity that coupled with the variation in heliographic latitude of the Earth during the year, gives rise to an apparent semiannual variation of geomagnetic activity in averages obtained over several solar cycles. They conclude that the observed semiannual variation is seriously overestimated and is largely an artifact of this inferred 22-year variation. We show: (1) that there is no systematically alternating annual variation of geomagnetic activity or of the solar driver, changing with the polarity of the solar polar fields, (2) that the universal time variation of geomagnetic activity at all times have the characteristic imprint of the equinoctial hypothesis rather than that of the axial hypothesis required by the suggestion of MTL11, and (3) that the semiannual variation is not an artifact, is not overestimated, and does not need revision.