Measures of the proportion of non-synonymous to synonymous mutations such as dN/dS are used to estimate selection forces acting on a genome, gene, or smaller region of that gene. Here a simple ratio of synonymous to non-synonymous mutations is applied to pairs of sequences ancestral to SARS-CoV-2 and also to non-SARS-CoV-2 related control groups. This analysis reveals between some pairs of close ancestors, and SARS-CoV-2, an unexpectedly high proportion of synonymous mutations, suggesting negative selection – even despite the switch in host species and tissue tropism. This suggests a need for further investigation.