We explain the high values of the acceleration (≈ 2° cy−2) found in the longitude of Mimas by Kozai and Dourneau when they fit to observations their current theory of the Mimas' motion. In fact, we have found that very long-period terms are missing in these theories; their expansion in powers of t well agrees with the observed acceleration. Effects of tidal dissipation are far smaller and could be determined only after accounting of these long-period terms.