First description by Blume (1851) as Sciaphila nana, but Giesen (1938), who dissected the type specimen having two female flowers and a bud, concluded it to be an Andruris species (Andruris nana (Blume) Giesen), separated from Sciaphila through a connective appendage in the male flowers (Schlechter 1912). This genus delimitation is accepted by various researchers (see Andruris wariana), and we followed this view for some time for the following reasons.
When checking numerous descriptions of Andruris and Sciaphila spp., I recognized some other features apart from the connective appendages which differ between the genera. Andruris spp. (including A. nana) have unequal tepals either through size and/or a knob-like extension at least on three tepals in the male flowers. If Sciaphila species have unequal tepals (many of them have not) they are either bearded or without anything. Andruris spp. tepals are never bearded. Beyond that, Andruris species have awl-shaped styles without papillae or hairs, whereas all Sciaphila spp. have brush-like styles. The only exception to this is the enigmatic Sciaphila ramosa (suspected to be an Andruris sp. by Giesen 1938), on which Hsieh et al. (2003) reports "style ...awl-shaped ... apex papillate". However, the same authors also state this feature for Sciaphila arfakiana (= Andruris crinita), on which most other taxonomist agree to have no papillae at the style (Giesen 1938, van de Meerendonk 1984, Averyanov 2007). The original description and drawing by Fukuyama and Suzuki (1936) does not show papillae, and none of any Sciaphila figures of awl-shape styles given in the literature show this. Not enough of confusion, Youhao and Cheek (2010) describe S. arfakiana as having a papillate style, but not S. ramosa! It seems as if this feature is not as easy to agree on.
In summary, apart from the connective appendages Andruris spp. could possibly be characterized by combination of the following features: unequal tepals and at least three knobby tepal apices in the male flower, and an awl-shaped style without hairs or papillae.
We admit, these are rather slight differences, one or the other feature is also present in Sciaphila and Seychellaria species. Van de Meerendonk (1984) also rejects the distinction between Sciaphila and Andruris due to the fragility of the anther appendage and refers to Andruris nana as Sciaphila nana. Under this name he summarizes Andruris gracillima and A. loheri of Giesen (1938). Corroborating, Chantanaorrapint and Thaithong (2004) found a specimen of Sciaphila nana in Thailand, which lacks the connective appendages. It is in most characters smaller than the description by Giesen (1938), who amended the limited description of Blume (1851). Another work by Averyanov (2007) mentions Sciaphila nana for Vietnam, this time having a short connective appendage ("filaments hardly exceeding the anthers"), and van de Meerendonk (1984) states for Sciaphila nana "filaments often exceeding the ... anthers". Hence, Sciaphila nana seem to be variable in this respect, speaking against the appendages as a clearly distinctive feature between the putative genera. This may also hold for the style features (see above) in Sciaphila ramosa and S. arfakiana.
The ultimate point for us to merge Andruris with Sciaphila is the molecular phylogeny of Mennes et al. (2013), who found Andruris sp., Sciaphila nana, and two Seychellaria spp. in one clade embedded in other Sciaphila clades. We, therefore, follow the view of the online checklist by Govaerts, Maas-van de Kamer and Maas (2016, search for Triuridaceae) based on Ohashi et al. (2008) to include Andruris in Sciaphila. The taxonomic inclusion of Seychellaria spp. in Sciaphila is still wanting.