Megalodon size

[theropod]

Currently I see no logical reason why Carcharocles should have a proportionally bigger (ie. regression would produce overestimate) or smaller (ie. underestimate) mouth. As long as no such reasons turn up, why wonder about them?
Of course such variations are always a problem, and especially here, but if they are not indicated thereÂ’s no need for or sense in presuming them.

Toothrow lenght will produce a conservative estimate in the sense of "reasonable", not in the sense of "low". If variables such as tooth spacing are accounted for wrongly, then it it will produce over- or underestimates (eg as in our great white example, in which the specimen must have had particularly wide spacings due to individual variation).

[Grey]

But what influence can have the fact that meg's toothrow is larger at UA size parity ?

I mean is it really reliable to use the formula, especially adapted to the great white, to megalodon, which have wider teeth at parity, no small third UA, thus a larger, different set ?

[theropod]

No formula is really realiable, but if any is, it´s this one. That larger teeth are not distributed the same way is quite normal.

However as you can see from Kent´s graph, individual teeth sizes are very variable even within Carcharodon carcharias(using single teeth-TL relationships), while its toothrow lenght nevertheless has a much better (and fairly strong) correlation with overall size.

Thus this variability in the distribution of the largest teeth does not mean a method based on jaw perimeter won´t be relatively accurate any more. The problem merely arises if you just compare upper anterior teeth, especially when in terms of lenght.

[Grey]

I'm fairly agreed with all of this.
My only concern is if this method induces conservative (but reasonable) estimates on a general trend.

[theropod]

I meant conservative in the sense of "best estimate", neither particularly low (which is the meaning of conservative usually implied, even in scientific works) nor particularly high. It apparently gives very reliable results for extant shark species.

The problem is that variations in interdental spacing can potentially create such over- or underestimates.

btw regarding the latter we could try and get a rough figure by simply measuring some pictures of jaws. It wonÂ’t be very exact, but would help to get a rough idea of the typical impact this has.

[Grey]

I'm skeptical that using great whites toothrow gives overestimates in megalodon using the same interdental spacing, as megalodon toothrow appears larger. Now, the apparent vast variations in great whites teeth disposition may complicate this.

[creature386]

He said both is possible, under- and overestimates. Variations are always possible in both directions.

[Grey]

But basically, as meg toothrow is longer when using the same interdental spacing, I'm not surprised that we get an underestimate while using great white parameters in which the decreasing tooth width is quite greater than in meg (at least based on the two compared sets earlier). It does not tend toward an overestimate on the same basis for meg. That's why I suspect the mean point of 17.5 m TL for a 14 cm wide tooth in the regression is more likely to be conservative. Now even there, variations may complicate this. I'm positive with that as I favor conservative means. I'm just not sure if Brett Kent kept the (apparent) slower tooth width decreasing in Carcharocles in mind in his regression based on great white bite perimeter. If you think that's worth to ask him theropod, don't hesitate (as you're launched in that discussion with him).

Now, I wonder how Mike Siversson estimated his >19 m max TL.

[theropod]

IÂ’m afraid I still cannot follow the logic. I may be misunderstanding you, so that it makes no sense to me.

Of course megalodonÂ’s toothrow is larger, because the animal is larger.
With the great white regression, we estimate its size based on the size of its toothrow. The larger it is, the larger the estimate will also be. We are hypothetising tooth spacings of the same proportion as in C. carcharias, whatever they are exactly, so at the same spacing they will have the same toothrow lenght.

Assuming a similar relationship between mouth size, and TL, it will estimate it just fine (the problem of course lies in the size difference, but we cannot tell in which direction and to what extend it influences its proportions relative to the regression).

ArenÂ’t you confusing this with KentÂ’s regression for root width, which is an entirely different thing? Using an UA this is indeed impacted by the variations in relative tooth size and may yield rather low figures in this case, when using upper anteriors. Using other teeth, eg. those that are the relatively widest ones in meg (the second lateral in the case of HubbelÂ’s set), it will behave differently.

You could test that by adding up tooth widths from the boneclone list and simply checking in what ratio toothposition-X-width/total-toothrow-lenght they each are, or I can do it tomorrow if you want.

[Grey]

Try it if you have interest thanks.

I don't argue about the fact that meg has a larger tooth row than GW (of course it has), but that placing their respective UA (or LA) at the same width and using the same interdental spacing, the slower decreasing width of meg's teeth may make the tooth row longer than in Carcharodon, hence it has a potentially important impact on the total size estimate (or has it ?). Just looking at the teeth sets page 26. I hope that's easier to understand ! The formula you've found is excellent, but has megalodon tooth row appears while placing the widest teeth at parity in both, I'm thinking that this formula, using the parameters of Carcharodon, with its faster decreasing width/short row, implies a predictable underestimate for Carcharocles.

What could be tried is to found a shark species with a decreasing tooth row even faster than in GW, and use on it the GW formula, to see if the results generate overestimates.

Edit : put that comparison for purpose. I've simply reduced that pic of GW set to put the UA at the same width than meg. The interdental spacing on this seems even a bit greater in the white shark, but you can see that meg's slower row makes it longer. Great white tooth row decreases faster in width, added to the third bended smaller UA. Hence, using GW set formula on meg set gives a potential underestimate for the later ?



Note : you can try to scale it yourself if you want to verify but I think that in my scale the UA of the GW are still very sightly wider than meg, still the tooth row is noticeably larger in meg.



And for fun, I've scaled both sets with each other, even leaving a margin to the great white to avoid any exagerration.



Seen like this, I cannot envision this megalodon smaller than twice the length of this 5.2 m great white and likely more than that. Of course, others sets would be usefue but I very much doubt this leads to a meg under 10.5 m. Coud be useful to measure the row lenght of this GW too.

Hubbell's set, without adding anything, gives 9.45 m TL. That is most likely an underestimate by comparing both (even recalling the GW set is a little bit too large for the scale). By guess, this meg was most likely more than twice longer than this great white.

[theropod]

But we are not actually placing the widest teeth at parity in both, let alone the upper anteriors. WeÂ’re placing the total toothrow lenght at parity (or rather, assuming a consistent relationship between Toothrow and TL, but thatÂ’s probably what you meant).

I acknowledge an erraneous assumption about tooth spacing will yield to underestimates (or overestimates if we assume too big a spacing). We should try to find out a typical percentage this adds and use that to derive a moderate estimate. 9.45m, not including any spacing, is of course the lower bound to what you can get with the formula, even though some great whites (and by inference probably some megalodons too, sicne they are driven by the same evolutionary factor affecting that) have that spacing. I think this specimen would more likely be in excess of 10m in life.

Yes, the main problematic you are referring to applies to tooth width, not toothrow lenght.

[Grey]

Hmm the problem I point on is the faster decreasing in GW and its bended UA3, which to me induces underestimates for meg. But I'll try to discuss this with Brett.

If we acknowledge 9.45 m for Hubbell's set, the owner of Hubbell's tooth was 17 m TL which sounds too low as 17 m TL is a figure published for others specimens which likely not had an UA about 15 cm wide (and that Kent says they may not reflect maximum size in his chapter). I'd expect a size estimate somewhat exceptionnal for the owner of a 15 cm wide tooth. 17 m is definitey large but not exceptionnal.

[theropod]

Not based on the summed withs of all tooth, unless you assume Carcharocles had an overally smaller toothrow compared to its body lenght.

That the UA3 and the posteriors are smaller in Carcharodon is of no relevance to this method*.

This is a problem relevant for ShimadaÂ’s method, because it compared individual tooth positions to those of great whites. This means it might create underestimates for anteriors and overestimates for posteriors, because the former are relatively smaller in meg and the latter relatively larger.

Keep in mind other estimates were made with different methodologies. ThatÂ’s simply a lower-bound to the potential range of estimates, based on this one. IÂ’ll wager ~10% higher may be more likely.

Also keep in mind the potential size range youÂ’d get with those other methodologies is even greater, it merely isnÂ’t given in most cases.

btw, based on boneclones, the largest tooth (L2) in this meg dentition is ~5.63% the summed width of all teeth, and the UA1 is 5.2%. The largest (the UA1) in the Carcharodon is 6%.

IÂ’d recommend to use the L2 for the root width, since this tooth corresponds best to the relative size of the one the regression bases on in Carcharodon.

* Unless there turns out to be some correlation this has with relative jaw size (which was never indicated so far and would be pretty strange considering its normal that different teeth in the mouth are differently large, without affecting the proportion between the mouth size and overall size).

[Grey]

Added Hubbell's giant UA.

[creature386]

According to what I understood from the google translated text, there were seven described teeth here:
dspace.uevora.pt/rdpc/handle/10174/9455
Does anyone have more data on this? Can one of these specimens be considered large?