Dvinosaurus primus (Amalitzky 1921; Late Permian; PIN2005/35) Dvinosauria traditionally include Neldasaurus among tested taxa. Here Dvinosaurus nests basal to plagiosaurs like Batrachosuchus and Gerrothorax and was derived from a sister to Laidleria. Dvinosaurus is a late survivor of a Devonian genesis and radiation. Note the large fish-like branchials.

Acroplous vorax (Hotton 1959, Foreman 1990; Early Permian) was similar to the above taxa, but had more laterally placed orbits. Note the appearance of the palatine where the lacrimal typically is found. On sister taxa the quadrate extends beyond the quadratojugal. The intertemporal was overlooked previously. The preopercular may have been present.

Perryella olsoni (Carlson 1987; Ruta and Bolt 2006; Early Permian). The classification of Perryella was originally uncertain because it shared features with two groups, Trimerorhachidae and Dissorophoidea, which were thought to be distantly related. In Ruta and Bolt 2006, a phylogenetic analysis placed Perryella within Dvinosauria as an intermediate form between trimerorhachids and other dvinosaurs, like Gerrothorax. Here flat-headed Perryella nests with Acroplous (above) within Dvinosauria.

Isodectes obtusus (originally Dendrerpeton obtusus, Cope 1896; Sequeira 1998; AMNH 4569; Early Permian) is a sister to Acroplous in the LRT.

Bermanerpeton kinneyi (Werneburg, Schneider and Lucas 2021; Late Pennsylvanian) was correctly nested within the Dvinosauria 'with branchiosaurid' characters. Here this tiny taxon nests with slightly larger Isodectes. The ectopterygoid does not contact the cheek. Importantly it preserves fore and hind digits in this late survivor of an earlier radiation of very basal tetrapods, so close to fish that preserves fish-like scales.

Batrachosuchus browni (Broom 1903; Early Triassic, 250 mya) nests with Gerrothorax, but does not have quite so wide a skull. A much smaller and more derived species is B. concordi (below) and it preserves a mandible.

Siderops kehli (Warren and Hutchinson 1983; Early Jurassic, 180mya; skull 50cm long, overall 2.5m long) was traditionally considered a chigutisaurid temnospondyl or a brachyopoid. Here Siderops nests with the much smalller Gerrothorax. No branchials and scales were reported. The back of the skull and the extremities are unknown, so modifications were made to reflect that lack of data here.

Koolasuchus cleelandi was a late surviving Early Cretaceous giant from this clade, presently known from just a few bones, like the mandible.

Gerrothorax pulcherrimus (Nilsson 1934, Jenkins et al. 2008; Late Triassic) was originally considered a plagiosaurine temnospondyl. Here it nests with the Trimerorhachisclade some of which share a lack of a supratemporal-tabular rim, straight lateral ribs and other traits. Jenkins et al. showed the skull raised like a toilet seat cover during feeding, rather than opening the mandible on this flattened bottom feeder. Distal limb elements are unknown.

Plagiosuchus pustuliferus (von Huene 1922; Middle Triassic, 240mya; SMNS 57921) nests with the smaller, Gerrothorax in the large reptile tree (LRT, 1444 taxa). Plagiosuchus has a narrower skull and confluent orbits with upper temporal fenestrae, partly due to the loss of the prefrontal, postfrontal and postorbital bones.

Both belong to the first clade to split from the  remainder of basal tetrapods following the transition from fins to fingers and toes. Both retained gills. The wide skull and laterally oriented dorsal ribs indicate Plagiosuchus was a full-time aquatic predator, a bottom dweller with little use of its limbs other than to paddle them around like fins.