The roughshark, Oxynotus, enters the LRT. The anal fin exits 13 times. Plus the many origins and ends of spines on fins.

The traditional family, Oxynotidae,
includes only one genus and 6 species. In the large reptile tree (LRT, 1936+ taxa, subset Figs. 3,4) Oxynotus (Fig. 1) nests basal to two LRT taxa, Isistius, the cookiecutter shark, and Dalatias, the kitefin shark, both with smaller fins. So… by the rules these two must also be members of the Oxynotidae. Shark experts (Kriwet and Klug 2009) don’t see it that way. More taxonomic issues below.

Figure 1. Oxynotus centrina in several views. Skull elements estimated. Here the dorsal spines start in the middle, not on the leading edge, distinct from most other spine fin taxa.

Oxynotus centrina
(Linneaus 1758; up to 1.5m; Fig. 1) is the extant angular roughshark. Ridges appear over the eyes and between the pectoral and pelvic fins. The anal fin is absent, as in other clade members. A dorsal spine is present on both dorsals, but not at the leading edge. These chimaera-like sharks feed on small bottom-dwelling prey, like worms, crustaceans and molluscs. The ventral nares are enormous.

Figure 2. Gregorius descends from Hybodus, the shark and is ancestral to Prohalecites at the base of the ray-fin bony fish. Gregorius is also ancestral to Homalacanthus at the base of the spiny sharks leading to lobefins, placoderms, catfish and a variety of other taxa.

The origin and evolution of spines on fins
Academic fish workers think spiny fins on spiny sharks (= acanthodians, Fig. 2) have no clear ancestors and no clear descendants. The LRT (Figs. 3,4) recovers a long set of spiny shark ancestors (like (Oxynotus, Fig. 1) and descendants (like like lobe fins and vertebrates).

Figure 3. Subset of the LRT focusing on basal vertebrates. Colors indicate spiny fins, lobe fins, etc. Note the basal nesting of Squalus, likely an Early Silurian taxon based on chronological bracketing. The catfish, Hoplosternum, has plats and spines.

The origin and evolution of the anal fin
The absence of the anal fin on Oxynotus brings up the subject of anal fin disappearance in vertebrates. In LRT (Fig. 4), the anal fin appears early (Fig. 4 lavender) and then disappears 13x (Fig. 4 amber), ultimately lost forever in Panderichthys + tetrapods.

Figure 4. Appearance (lavender) and disappearance (amber) of the anal fin in vertetbrates.

Traditionally (= in textbooks)
Oxynotus (Fig. 1) is considered a member of the Squaliformes (dogfish sharks). The
last common ancestor in the LRT (subset Fig. 4) is the extant dogfish, Squalus (Fig. 5).

Kriwet and Klug 2009 wrote,
“Dogfish sharks (the Squaliformes) are a highly diverse group of neoselachian sharks occurring in coastal and oceanic, cool temperate, and deep tropical waters worldwide.”

There’s that phrase, “highly diverse” again. This time, however, if Squaliformes are indeed monophyletic, Kriwet and Klug grossly underestimate the diversity of this clade, which includes the authors themsevles!

Figure 5. The spiny dogfish, Squalus acanthi as, in vivo. Note the anal fin is absent in this taxon and all traditional squaliform sharks. The anal fin disappears 13x in the LRT.

LRT taxonomy
In the LRT (Fig. 2) Squalus (Fig. 5) is the last common ancestor of a small clade that includes chimaera (ratfish).

In the LRT (Fig. 2) Squalus is close to the last common ancestor of a much larger clade that includes ratfish, all more derived sharks, all bony fish (including tetrapods).

In the LRT (Fig. 2) Oxynotus and other sharks that look like Squalus are currently separated from Squalus by traditional non-squaliform taxa, like Isurus, a traditional lamniform alongside the CMNH 9280 specimen from the Late Devonian.

More taxa in the LRT will probably solve this issue. Either Lamniformes arise from squaliformes or some phylogenetic definitions need to change.

Figure 6. Dalatias, the extant kitefin shark, has relatively large labial cartilages.

According to Kriwet and Klug 2009
the fossil record of Squaliformes consists predominantly of isolated teeth that date back to the Barremian (Early Cretaceous), ca. 125 million years ago. Full body fossils are found in Late Cretaceous strata.

In the LRT chronological bracketing hypothetically moves that record back to the Silurian.

According to Kriwet and Klug 2009
“The supertree resulting from combining published partial phylogenetic hypotheses the most inclusive estimate of squaliform interrelationships to date showing 23 out of about 35 known fossil and extant genera. The results presented here are in general agreement with most published phylogenetic hypotheses and differ only slightly from others. According to our hypothesis, Squaliformes is composed of six monophyletic groupings of family rank; Squalidae (Squalus and †Protosqualus) is separated at the deepest branching level from all other squaliforms according to this reconstruction.”

Unfortunately,
Kriwet and Klug did what most shark experts did and do: they cherry-picked only those taxa that look like Squalus. Their cladogram includes no rays, ratfish, higher sharks, bony fish and tetrapod descendants of Squalus. In the LRT (Figs. 3, 4) taxon inclusion sheds new light on hypothetical interrelationships.

Let your software tell you which taxa to include and exclude.
Don’t cherry-pick taxa based on obvious traits. Monophyly is a powerful concept that needs to be utilized at all nodes.

References
Kriwet J and Klug 2009. Fossil record and origin of squaliform sharks. (Chondrichthyes, Neoselachii. Pp. xx –xx. (Gallucci, McFarlane and Bargmann eds.) In Biology and Management of dogfish sharks. Chapter: Fossil record and origin of squaliform sharks (Chondrichthyes, Neoselachii). American Fisheries Society, Maryland.
Linnaeus C 1758. Systema naturæ per regna tria naturæ, secundum classes, ordines, genera, species, cum characteribus, differentiis, synonymis, locis. Tomus I. Editio decima, reformata

wiki/Angular_roughshark – Oxynotus
wiki/Acanthodii
wiki/Squaliformes