Dunkleosteus Downsized

[suspsy]

Looks like the mighty mouthed monstrosity may not have been so monstrous after all!

https://www.mdpi.com/1424-2818/15/3/318?mibextid=Zxz2cZ&fbclid=IwAR1UUlbetRWARoV1UHagmeR-UymcBdgEheK9Ct_bU0xLE6sI_sXpvOe4bEA

[Eatmycar]

Welp that's awfully disappointing...

Don't get me wrong, I love that we're learning more, but it's a little sad to see such an absolute unit get shrunk down!

[ceratopsian]

It's still by no means a minnow though.

[SidB]

And, really, I still would choose to NOT get into the swimming pool with it, even after the downsizing,!

[Faelrin]

Wait, it's only got one author and says it is early access? Has this been peer reviewed yet? Just want to err on the side of caution, since this is kind of a big jump down, and the proportions are quite a bit different as well. Also makes the recent PNSO figure even more too long (but perhaps Favorite's model is still within reason?).

Here's a graphic from the article:



I can't help but be reminded of some varieties of goldfish with these new proportions, like fantails, etc.

Gonna tag @Halichoeres our resident fish expert. What's your thoughts on this?

[Stegotyranno420]

Oh dear   :
The big guy still lives in ny heart

On a more serious note, it is not unlikely for the larger vareities to have existed. If i remember modern fish can vary vastly in size.

[Flaffy]

Was just about to tag Tim in this  as well. Wanted to know what he thought about the new pudgy fancy-goldfish like proportions.

[dinofelid]

The abstract says it's about a new method of estimating total length from "orbit-opercular length" which is said to correlate well both in living fishes and in "arthrodires known from complete remains" (I hope the full paper will include something like a scatter plot showing how far individual species can deviate from the prediction).

One thing I wonder about that diagram is why the head itself seems noticeably smaller--are both outlines meant to be different reconstructions of the same fossil CMNH 5768 referenced in the caption?

On the issue of size variation that @Stegotyranno420 mentioned, the abstract does also mention a larger specimen, CMNH 5936 (only known from a lower jaw according to this diagram), which they estimate at 4.1 m long, would be interesting to see an outline of that one (and to see how big the head would be).

[Kapitaenosavrvs]

Was just about to tag Tim in this  as well. Wanted to know what he thought about the new pudgy fancy-goldfish like proportions.

Hell.

[TheCambrianCrusader]

I'm certainly no expert on this, but I'm sure there must be a number of fish that don't correlate well with the orbit-operculum length estimates. Wolf eels come to mind pretty immediately. Doesn't disprove this hypothesis obviously but I do doubt its reliability to some extent. Gonna have to give the paper a read, good thing its open access.

[Ludodactylus]

Who's out here breeding goofy ornamental Dunkleosteus?

[stargatedalek]

Personally I think the included reconstruction takes the idea of the paper and runs with them, perhaps a little bit too far. Something slightly less compact, proportions like a sandtiger shark for example, might be a bit more likely.

[dinofelid]

I came across an October 2022 thread on one paleoartist's recent reconstruction of Dunkleosteus where he says in this tweet "Today, Jobbins et al. published an amazing find from Morocco-another arthrodire placoderm preserving soft tissue. This fish, named Amazichthys, is closer both in phylogeny and ecology to Dunk and while not perfect is easily the best model for its body we have so far". The Amazichthys body proportions look sort of intermediate between the older and newer reconstructions of Dunkleosteus from the paper in the OP, I wonder how well the orbit-opercular length method would predict it.

[Halichoeres]

Aw, thanks for thinking of me, guys.

I think the best way to think of this paper is as a hypothesis. I've only skimmed it so far, but it seems like a well-argued hypothesis based on quite a robust data set. In principle the hypothesis could be tested/refined upon the discovery of the right specimen, as the tail-shape hypothesis was when Amazichthys was described.

This is a reconstruction included in the paper, which looks a little less odd than the one in the graphical abstract.

If the length estimate is correct--I'll get to that--then a rather thick, foreshortened body is implied, because the angle between the thoracic shields only has a little bit of wiggle room. The reconstructions reflect that angle.

It's a little surprising that orbit-opercular length should be such a strong predictor of overall length, but elongation of various body regions is under the control of some of the same transcription factors (genes that regulate the expression of other genes), so maybe it shouldn't be surprising. My first issue with the analysis is that the structure of the pharyngeal arches in arthrodires is different enough from stem-group gnathostomes that the rear margin of the head does not really constitute a homologous landmark when compared with, say, teleosts.

Despite this, the model predicts the lengths of placoderms really well:

(@dinofelid I believe this is the graph you specifically requested!)

The outliers way above the regression line are eels and eel-like fishes. This is a distinct elongation phenomenon because it involves duplication of vertebral segments during development. That happens in other fishes, too, but not to the same extent.

Anyway, I think there are ecological and biomechanical reasons to suspect that Dunkleosteus was a bit longer than reconstructed here. The 10 meter estimates have been thought to be way too high for years, but 3 - 4 meters might err a bit too far in the other direction. What I would like to see is somebody use this length estimate to build a model and actually estimate its swimming performance. Maybe it would prove to be a really excellent shape for maneuverability! Either way I'd like to see somebody try it out.

Overall, I think this is an interesting and creative approach to size estimation, and much of its power lies in generating hypotheses. It isn't the final word, and doesn't claim to be. MDPI is a bit of a controversial publisher, so it's possible that it's in this journal because it was rejected elsewhere, but I recognize some of the reviewers and trust their judgment. The paper's full version is online and is open-access, but it's easy to miss the link to download the pdf.

[TheCambrianCrusader]

Aw, thanks for thinking of me, guys.

I think the best way to think of this paper is as a hypothesis. I've only skimmed it so far, but it seems like a well-argued hypothesis based on quite a robust data set. In principle the hypothesis could be tested/refined upon the discovery of the right specimen, as the tail-shape hypothesis was when Amazichthys was described.

This is a reconstruction included in the paper, which looks a little less odd than the one in the graphical abstract.

If the length estimate is correct--I'll get to that--then a rather thick, foreshortened body is implied, because the angle between the thoracic shields only has a little bit of wiggle room. The reconstructions reflect that angle.

It's a little surprising that orbit-opercular length should be such a strong predictor of overall length, but elongation of various body regions is under the control of some of the same transcription factors (genes that regulate the expression of other genes), so maybe it shouldn't be surprising. My first issue with the analysis is that the structure of the pharyngeal arches in arthrodires is different enough from stem-group gnathostomes that the rear margin of the head does not really constitute a homologous landmark when compared with, say, teleosts.

Despite this, the model predicts the lengths of placoderms really well:

(@dinofelid I believe this is the graph you specifically requested!)

The outliers way above the regression line are eels and eel-like fishes. This is a distinct elongation phenomenon because it involves duplication of vertebral segments during development. That happens in other fishes, too, but not to the same extent.

Anyway, I think there are ecological and biomechanical reasons to suspect that Dunkleosteus was a bit longer than reconstructed here. The 10 meter estimates have been thought to be way too high for years, but 3 - 4 meters might err a bit too far in the other direction. What I would like to see is somebody use this length estimate to build a model and actually estimate its swimming performance. Maybe it would prove to be a really excellent shape for maneuverability! Either way I'd like to see somebody try it out.

Overall, I think this is an interesting and creative approach to size estimation, and much of its power lies in generating hypotheses. It isn't the final word, and doesn't claim to be. MDPI is a bit of a controversial publisher, so it's possible that it's in this journal because it was rejected elsewhere, but I recognize some of the reviewers and trust their judgment. The paper's full version is online and is open-access, but it's easy to miss the link to download the pdf.

Thanks for shedding some light for us! Its crazy how there's barely any outliers on that graph, I would've expected a few more

[Flaffy]

Well well well, that model certainly is enlightening. Science marches on! Bring on some new Dunkleosteus figures!

It's amusing how BBC's 2 decade old reconstruction bears a striking resemblance to the 2023 paper in terms of proportions.


I guess Paleozoo's version reigns supreme then based on the new paper. Albeit a bit longer compared to the paper recon.

[Paleo Flo]

Science goes on..n in a dramatically way. I always took the shape & body length of Dunkleosteus for granted.

[andrewsaurus rex]

Perhaps another addition for my "Do you miss the Dinosaurs that Never Existed' thread?

In the before and after size picture above, the head seems smaller on the 'after' illustration.  The head sizes are known, so that seems fishy to me...(pun intended).

[Newt]

There's an easy solution. Let's all go to Ohio (there's a phrase you don't hear too often) with picks and shovels, excavate the entire Ohio Shale, and find us some whole-body specimens. We'll need some people to hit the Antrim and Chattanooga shales too, just in case.

[Stegotyranno420]

There's an easy solution. Let's all go to Ohio (there's a phrase you don't hear too often) with picks and shovels, excavate the entire Ohio Shale, and find us some whole-body specimens. We'll need some people to hit the Antrim and Chattanooga shales too, just in case.

We can get the whole forum together