The phylogeny in this paper is interesting, for a number of reasons. If _Changmiania_ is a fossorial/burrowing ornithopod, then it suggests this behavior is primitive for Ornithopoda; this is because the next outgroup (Orodrominae) also comprises fossorial burrowers (as the paper makes clear). Further, it's been suggested elsewhere that _Drinker_ lived in burrows, although this has yet to be published (and likely never will):
_Drinker_ has since been referred to _Nanosaurus_ (Carpenter and Galton, 2018). According to Yang et al.'s phylogeny, _Nanosaurus_ belongs to a more derived clade, an unnamed sister group to Clypeodonta.
Also noteworthy (and novel, AFAIK) is that _Kulindadromeus_ comes up as the sister taxon to Marginocephalia (Ceratopsia + Pachycephalosauria), outside Ornithopoda.
Finally, I don't understand why Yang et al. use so many subfamily-based names for their ornithopod clades: Orodrominae, Jeholosaurinae, Parksosaurinae. Subfamilies are subdivisions of families - but Yang et al. do not use them this way.Â
Orodrominae should be Orodromidae - the attribution doesn't change (the subfamily and family are both Brown et al., 2013). Sensibly,
Brown et al. (2013) explicitly defined Orodrominae as being a clade of thescelosaurids. Since Yang et al. recovered Orodrominae outside Thescelosauridae, there is no reason to retain
Orodrominae. Â
Jeholosaurinae should be Jeholosauridae, following Han et al. (2012).Â
Parksosaurinae should be Hypsilophodontidae, because it includes _Hypsilophodon_. There is already a family Hypsilophodontidae available, and has been for donkey's years (it goes back to Dollo, 1882).
The use of a stand-alone subfamily also popped up in the recent monograph on pennaraptoran theropods (Bulletin of the AMNH), where Anchiornithinae was used in preference to Anchiornithidae, even though Anchiornithinae didn't belong in any family: it was recovered as the most basal clade of the Avialae. The reason given was that "Anchiornithinae: we use this stem-based taxon containing _Anchiornis_ instead of Anchiornithidae because the former was proposed earlier [Xu et al. 2016])". But if Anchiornithinae becomes Anchiornithidae, Xu et al. (2016) still get credit (despite Foth and Rauhut [2017] later naming Anchiornithidae). It's not clear why Xu et al. (2016) chose to call this clade Anchiornithinae rather than Anchiornithidae; at the time, the authors may have believed _Anchiornis_ was a troodontid.
I think the basic problem is that coordinated family-level taxa are hierarchial
(superfamily, family, subfamily, tribe), so this hierarchy should be taken into account when coming up with phylogenetic definitions for these clades. Often this doesn't happen, so subfamilies can end up outside of families (as happened with Yang et al.); and families can end up inside other families (as has happened with some hadrosaur trees, which have Saurolophidae inside Hadrosauridae).
Â