Megamastax, ‘the largest Silurian vertebrate’ is only known from a mandible, but did this genus have limbs?

Today’s hypothesis is going to be interesting… and risky
because so few clues are available and 173 million years separate candidate sister taxa. Today we’re going to take phylogenetic bracketing to the possible Late Silurian genesis of the Tetrapoda.

Choo et al. 2014 described
two 3D mandibles (Figs. 1, 2) from the Late Silurian (423 mya). The authors considered the holotype to be an “Osteichthyan with multiple rows of closely packed conical teeth on the marginal jaw bones and widely spaced pairs of blunt teeth fused to each of the four coronoids. The presence of cosmine, the shape of the maxilla, and the configuration of the prearticular relative to the coronoids indicate sarcopterygian affinities.”

Traditionally the clade Sarcopterygii includes only and all lobe-fin fishes, including Guiyu (Fig. 4), the oldest. After testing in the large reptile tree (LRT, 2011 taxa, subset Fig. 3), fish with lobe fins are not monophyletic. Lobe fins appear several times by convergence in the LRT and some give rise to ray-fin taxa. To make matters worse, the last common ancestor of all taxa with lobe fins in the LRT is Diplacanthus, a spiny shark (= acanthodian).

Fossil fish don’t follow a neat and ordered chronology
in the LRT (subset Fig. 3). Late survivors nest at basal nodes. Late Silurian and Early Devonian taxa, like Brindabellaspis, nest only a few nodes away from the first tetrapods.

Published length estimates for Megamastax of 1 meter or so
depend on the Eusthenopteron-like model proposed by Choo et al. (Fig. 2). Instead, if Laidleria emerges as the new model, the estimated length for Megamastax would be reduced to about 50cm.

No other tested vertebrates
in the LRT have the traits listed by Choo et al. AND have such a narrow snout tip expanding posteriorly like a straight-edged triangle in dorsal view. In Laidleria the tiny double row of dentary teeth are absent, but the widely spaced coronoid fangs remain. This is a small change for 173 million years of evolution. We all look forward to seeing more complete specimens of Megamastax someday.

Two final arguments for considering Late Silurian Megamastax an early tetrapod:
By the Early Devonian (365 mya) we have Acanthostega and Ichthyostega, two tetrapods evolving back toward a more aquatic lifestyle leaving no descendants in the LRT (Fig. 5). More importantly, by the Early Devonian we also have Tulerpeton, a reptilomorph in the LRT. That leaves the entire Devonian free to produce Devonian sisters to all the tetrapod taxa between Laidleria and Tulerpeton (Fig. 5). The current lack of transitional tetrapod taxa in the Devonian await discovery.

Like I said at the top,
on the basis of these very few clues the origin of tetrapods here moves back to the Late Silurian, 30 million years prior to the first evidence of footprints. This falsifiable hypothesis could change when more complete specimens of Megamastax come to light that link it more parsimoniously to other taxa with fins. For now, Laidleria is the best match and Middle Devonian trackmakers with fingers and toes are no longer anachronistic.

Lesson for today (and every day):
A wide-gamut cladogram is a powerful tool for determining clades. Build one yourself and you, too, will have a powerful tool you can use for the rest of your career in paleontology.

References
Choo B et al. (4 co-authors) 2014. The largest Silurian vertebrate and its palaeoecological implications. Scientific Reports 4. doi:10.1038/srep0524
Kitching JW 1957. A new small stereospondylus labyrinthodont from the Triassic beds of South Africa. Palaentologia Africana 5:67–82.
Warren A 1998. Laidleria uncovered: a redescription of Laidleria gracilis Kitching (1975), a temnospondyl from the Cynognathus Zone of South Africa. Zoological Journal of the Linnean Society. 122 (1–2): 167–185.

wiki/Laidleria
wiki/Megamastax
wiki/Sarcopterygii