Largest carnivore from the Kem Kem beds Aegisuchus or Spinosaurus

[Vladimír Socha]

Good day!

I was wondering, whether the length estimate of 15 to 22 meters for the aegyptosuchid crocodylomorph Aegisuchus witmeri still holds? Especially when there is also a possibility that the skull was not 2.08 to 2.86 meters long, but rather 62 to 90 cm (which would shrink this crocodylomorph to just 3.9 meters long). Anyway, it would be very interesting to discover how did the interactions between these two river predators look like. With its flattened skull, Aegisuchus was probably an ambush predator and its possible prey include bichirs, coelacanths and/or lungfish, as these have been found in the Kem Kem beds. Aegisuchus was an opportunistic predator and may also have preyed on land vertebrates such as small dinosaurs. According to some researches, this huge crocodylomorph could have potentially fallen prey to large theropod predators such as Spinosaurus or even Carcharodontosaurus. Any thoughts on this? Thank you in advance! VS.

References:

Holliday, C. M.; Gardner, N. M. (2012). A New Eusuchian Crocodyliform with Novel Cranial Integument and Its Significance for the Origin and Evolution of Crocodylia. PLOS ONE. 7 (1): e30471.

Morales-Betancourt, M. A. (2014). Comment on "A New Eusuchian Crocodyliform with Novel Cranial Integument and Its Significance for the Origin and Evolution of Crocodylia". PLOS ONE. 7 (1): e30471.

Ibrahim, N.; et al. (2020). Geology and paleontology of the Upper Cretaceous Kem Kem Group of eastern Morocco. ZooKeys. (928): 1–216.

Gayford, J. H.; et al. (2024). Cautionary tales on the use of proxies to estimate body size and form of extinct animals. Ecology and Evolution. 14 (9): e70218.

[Nick Gardner]

Disclaimer: I am the Gardner in Holliday and Gardner (2012). The information shared below only represents my viewpoints, I am not claiming to speak for both of us. I have not discussed this with Casey to my memory, though I don't think either of us would dispute Gayford et al. (2024). A further disclaimer, I discussed this matter quite a bit with one of the authors on Gayford et al. (2024) prior to their submission of the manuscript for peer review and I don't dispute anything that they concluded there. I am willing to take full responsibility for any mathematical errors in the paper.

I think the supplementary material in Gayford et al. (2024) provides enough argument that the large body size estimates were oversized, but they were based on using the same metrics that Gaylord et al. (2024) also criticize (Sereno et al., 2001; Sereno and Larsson, 2009). Holliday and Gardner (2012) did as best they could working within the framework of crocodilian body size estimation as it was understood at the time. It's clear that the different approaches to skull size/shape, body size/shape estimation in crocodilians are more complicated than conventionally thought or even by the equations presented in Sereno et al. (2001; and publications who subsequently applied their approaches).

To really get at this problem, someone needs to go and remeasure multiple measurements thought to predict skull size, shape, body size, body length, etc. in crocodilians for a very large sample against clade diversity, body size disparity, skull shape disparity, and ecology of pseudosuchians as a whole. At one point, I sat down and checked, it would probably take a visit to at least four major museums in North America to get a large enough sample size alone just to look at Alligator and other living crocodilians to try to grapple with this problem in the crown-clade.

 This would not be enough of a data set on its own, it would be necessary to also get measurements of taxa outside of the crown clade all the way back to the root of Pseudosuchia and probably even measure outgroups to Pseudosuchia as well. There are different things happening with trends in body size/shape against skull size/shape inside the crown-clade and outside of the crown-clade (look at the supplementary material in Gayford et al., 2024 and discussions and citations therein for items #36-42).

I think the clear takeaway from Gayford et al. (2024) is that our field has relied too long on approaches that are too simplistic and fail to take into account what happens in animal growth on an individual level or even body size evolution as a whole. Peer reviewers should probably to replicate analyses where possible, especially since these days, that's as simple as rerunning the readily supplied supplementary data through the same scripts for statistical analyses. Individual students or researchers should consider talking to colleagues who are more math-fluent (possibly even outside their departments/divisions) for their feedback before submitting manuscripts for peer review (as peer reviewers may also be similarly disadvantaged and not see the issues). Several conversations I've had with various people have made clear to me that numeracy and statistical literacy are not as well-developed in paleontology as one might expect and there's a lot of go along to get along following what prior researchers did.

Just my five cents,

Nick

[Russell Engelman]

Something to at least partly mention as well, as someone who was an author on Gayford et al. 2024, is that we weren't the ones to really notice the issue but at least partially became aware of it due to a comment the herpetologist Mónica A Morales-Betancourt made on the original paper, as well as Nick himself bringing up this possibility with one of our co-authors. Just to give proper credit to who made the observation.