The piculets have long been classified as four genera: Sasia Hodgson, 1837; Verreauxia Hartlaub, 1857; Picumnus Temminck, 1825 (including Vivia Hodgson, 1837); and Nesoctites Hargitt, 1890. They have typically been grouped in the subfamily Picumninae (e.g. Peters 1948, Winkler et al. 1995, Dickinson 2003), although most authors have removed Nesoctites from this subfamily and placed it either in Picinae (Dickinson & Remsen 2013) or in its own subfamily, Nesoctitinae (Wolters 1976, Benz et al. 2006, Gaudin 2022). In contrast, Sasia, Verreauxia and Picumnus have always been placed in Picumninae (e.g. Peters 1948, Wolters 1976, Winkler et al. 1995, Dickinson 2003, Dickinson & Remsen 2013).

In the first multilocus molecular phylogenetic study by Fuchs et al. (2006) Sasia and Verreauxia formed a separate clade from Picumnus and these two groups represented an unresolved trichotomy with Picinae. Nesoctites was not included in this analysis. Benz et al. (2006) found Sasia and Verreauxia sister to Picumnus, albeit with no support, but the two groups were separated by a deep divergence exceeding that found between Nesoctitinae and Picinae. In another multilocus phylogeny (Fuchs et al. 2007), Sasia and Verreauxia were not sister to Picumnus; instead, Picumnus was more closely related to Picinae than to Sasia and Verreauxia. In another study (Fuchs et al. 2013), phylogenies based on mitochondrial DNA, nine autosomal and three Z-linked loci did not place Sasia and Picumnus as sister taxa; rather, analyses of mitochondrial DNA and the Z-linked loci placed Sasia closer to Picinae than to Picumnus, whereas the autosomal loci placed Picumnus closer to Picinae than to Sasia. A further study based on mitochondrial DNA sequences placed Sasia closer to Picinae than to Picumnus (Winkler et al. 2014). A sixth multilocus study placed Sasia and Verreauxia sister to Picumnus, but again separated by a deep divergence (Dufort 2016). More recently, a multilocus study by Shakya et al. (2017) placed Sasia and Verreauxia closer to Picinae than to Picumnus. Analyses of the two mitochondrial rRNA and 13 protein-coding genes again recovered a deep split between Sasia/Verreauxia and Picumnus but failed to unambiguously resolve the relationships between these taxa and Nesoctites/Picinae (JF et al. unpubl. data). All phylogenetic studies that included Nesoctites placed it as the sister of (other) Picinae, consistent with its removal from Picumninae.

The phylogenetic evidence clearly supports the distinctiveness of Sasia/Verreauxia from Picumnus. Importantly, it is unclear if Picumninae (excluding Nesoctites) is a monophyletic group. This suggests that the piculets (excluding Nesoctites) may be best treated as two separate taxonomic groups: Picumninae is restricted to Picumnus (and Vivia, if that genus is recognised for P. innominatus, e.g. Gaudin 2022); and another taxon is recognised for Sasia and Verreauxia. ‘Sasiinae’ was recently used as a group name for the latter two taxa but this represents a nomen nudum, as was indicated by the author by his use of square brackets (Gaudin 2022). A Google search (24 September 2022) showed that the name ‘Sasiidae’ is sometimes used in online publications for a group of moths but this is clearly a misspelling of the name Sesiidae Boisduval, 1828 (Lepidoptera). As it appears that no family-group name is available for Sasia and Verreauxia (Bock 1994), and that the names ‘Sasiidae’ and ‘Sasiinae’ have not been made available for any other taxonomic group in zoological nomenclature, we propose: