Estimates of palaeoflow conditions based on antidune geometries generally are obtained using equations derived from potential flow analysis. Unfortunately, in several previous studies and review articles, misunderstanding and improper substitution of density terms has resulted in two misconceptions (a third is discussed in detail elsewhere). Both concern the commonly derived ratio of antidune wavelength to critical flow depth (L/D). Previously published results which suggest that, for a given set of flow conditions in a two-layer system, L/D increases with increasing density contrast (Δρ) are wrong. Correct substitutions indicate the opposite—there is a slightly inverse relationship between the value of L/D and Δρ. The second misconception concerns the two values reported for critical flows: L/D≥ 6·3 (or ≥ 2γ) and L/D≥ 12·6 (or ≥ 4γ). Despite insinuations that the latter is wrong by a factor of two, both are correct. In this case, what has been overlooked is that the former value is derived for antidunes on free surface flows (a ‘two-layer’ system in which ρ2 << ρ1) and the latter for antidunes at the interface of a two-layer system in which Δp→0 (i.e. ρ2SIMEρ1).