Buzzsaw!: An ancient spiral-toothed shark

February 27, 2013 • 6:43 am

Imagine an ancient shark with a single spiral tooth, shaped like a buzzsaw, in its lower jaw.

That’s what’s reported in a new paper in Biology Letters by Leif Tapanila et al. (free download).  The spiral-like structure of this fossil, Helicoprion, had been known for some time, but it was curious: what seemed to be a single serrated tooth in the shape of a logarithmic spiral. Biologists had wondered how it was placed in the jaw, how it was used, and even if it was a tooth rather than some other part of the body.

This is what the structure, now known from the paper to indeed be a tooth, looks like (all photos and drawings from the paper):

Helicoprion specimen IMNH 37899, preserving cartilages of the mandibular arch and tooth whorl. (a) Photograph and (b) surface scan of fossil, positionedanterior to the right, imbedded in limestone slab.
Helicoprion specimen IMNH 37899, preserving cartilages of the mandibular arch and tooth whorl. (a) Photograph and (b) surface scan of fossil, positioned
anterior to the right, imbedded in limestone slab.

The authors did CT scanning of a fossil found in 1950 in Idaho, dated to the early Permian—about 270 million years ago.  The tooth, fortuitously, was embedded in the remains of the skull, something that’s rare because sharks have cartilage instead of bones in their skeleton. (That’s why shark teeth are so much more common in the fossil record than sharks themselves.) Using scans, they were able to show the placement of the teeth in the jaw.

Here’s a model, based on the CT scan, of the tooth placed in the mouth, taken from the side (lateral position):

Picture 3

And an oblique view from the side with the position of tooth interpolated from the scan:

Picture 4

(For you readers who know anatomy, here’s the key: bp, basal process; c, cup-shaped portion oflabial cartilage; ep, ethmoid process; lj, labial joint with base of root; pf, lateral palatine fossa; pp, process limiting jaw closure; qf, lateral quadrate fossa;qmf, quadratomandibular fossa; qp, quadrate process.)

The tooth apparently grew continuously, and, as the shark’s mouth closed on its prey, could be rotated up and back, cutting the prey and forcing it into the back of the mouth.  BBC Nature interviewed the first author and gives more information:

Using the computer images, the team could build a 3D model of the jaw, to reveal how the tooth spiral worked.

“As the mouth closes, the teeth spin backwards… so they slash through the meat that they are biting into,” Dr Tapanila told BBC Nature.

“The teeth themselves are very narrow: nice long, pointy, triangular teeth with serrations like a steak knife.

“As the jaw is closing and the teeth are spinning past whatever it’s eating, it’s making a very nice clean cut.”

What could those teeth be used to eat? The answer is rich (P. Z. take note):

Dr Tapanila said that this evidence, combined with the “rolling and slicing” mechanism, provided clues to what the ancient fish ate.

“If this animal were eating other animals that were very hard or [had] hard armour plating or dense shells, you would expect more damage to their teeth.

“This leads us to believe that our animal was probably eating soft, squishy things like calamari. It was probably eating squid or its relatives that were swimming in the ocean at the time.”

As the paper reports, the spiral teeth are involved with a suite of other skull adaptations. Anatomy buffs take note:

Retention of teeth in a continuously growing whorl necessitates specialized morphologies, including the buttressing labial cartilages to maintain rigidity and alignment of the whorl, as it occludes between the upper jaws. With the jaw articulation next to the whorl, closure of the lower jaw rotates the teeth dorsoposteriorly, providing an effective slicing mechanism for the blade-like serrated teeth and forcing food to the back of the oral cavity.

Accommodating the continuous growth of the logarithmic whorl required commensurate anterior and dorsal expansion of the mandibular arch to house the symphyseal structure. Based on the largest diameter whorls in the IMNH collections, Helicoprion jaw length and height could exceed 50 cm, nearly double the size of IMNH 37899. Pre-mortal tooth wear or breakage is rare in Helicoprion [5,6]. This may be a result of rapid tooth production—some whorls exceed 150 [cm.!]—along with prey selection of soft-bodied animals, such as cephalopods [6] or poorly armoured fish.

I’m sure you’re probably wondering what the animal looked life in life. Here’s a reconstruction from the paper:

Picture 1

UPDATE: Reader gb james, in the comments below, points out a longer National Geographic article on the species with more pictures. Here’s one:


h/t: Dom


Tapanila, L., J. Pruitt, A. Pradel, C. D. Wilga, J. B. Ramsay, R. Schlader, and D. A. Didier. 2013. Jaws for a spiral-tooth whorl: CT images reveal novel adaptation and phylogeny in fossil Helicoprion. Biology Letters 9:10.1098/rsbl.2013.0057.

39 thoughts on “Buzzsaw!: An ancient spiral-toothed shark

  1. No god would have thought of something as strange – to our eyes – as this fish, but Natural Selection on the other hand, is bounded only by what material it starts with.

    1. That’s exactly what I was thinking! No evolutionist would have thought of something as strange – to our eyes – as this fish, but God, on the other hand, is bounded by nothing.

  2. Seems like a major tooth production line for an eater of soft tissue. Could the rarity of broken teeth be explained, instead, by rapid replacement?

        1. Umm, no, wrong Bond villain I think.

          Possibly Doctor No or Ernst Stavro Blofeld, and almost certainly Karl Stromberg and Franz Sanchez. But Auric Golfinger’s obessession was almost entirely directed at gold. And golf.

  3. Thank you for posting this. People used to say ‘Now I’ve seen everything!’ when confronted with something like this. It is pleasant to reflect how far I am from having seen everything.

  4. Oh, a quick check of the free paper shows it’s not quite as weird as some (including Nat Geo) are reading it.

    The structure is called a ‘tooth whorl’, i.e. a whorl of teeth, not a single tooth.

    Also, of course, the logarithmic spiral represents successive stages of growth, which proceeds in the usual way from small to large, i.e. ‘NEW TEETH FORMED HERE’ in the Nat Geo figure is exactly wrong; teeth are SHED from the outside extremity of the series. (Otherwise, where would they be shed to?)

    Look into the mouth of a Great White and you see a queue of replacement teeth behind every one of the functional teeth, ready to pop a new one off the stack into biting position. Basically the same thing, except Helicoprion has invested most of its tooth tissue in the median mandibular zahnreihe, and found a way to use them in parallel rather than series.

    1. ‘NEW TEETH FORMED HERE’ in the Nat Geo figure is exactly wrong

      I’m not so sure. The paper says:

      In our model, the posterior region of the lower jaw is the site where ever-larger tooth crowns are produced

      which is where Nat Geo has it.

      Moreover, this is where you’d expect it to be by homology with other sharks, in which teeth are shed at the front edge of the jaw and are replaced by new teeth forming behind them. That’s the same direction of growth as Nat Geo shows for Helicoprion.

      The difference seems to be that Helicoprion never (or rarely) sheds teeth; instead, they’re wrapped in cartilage and continue to rotate down into the jaw as new teeth form behind them. So the oldest teeth are the ones at the center of the spiral, and the reason they’re smaller is because the shark itself was much smaller when those earliest teeth formed.

      1. Modern (adult) sharks continuously grow new teeth (from small to large) before shedding them.

        If Helicoprion reversed this process, and retained all older teeth, how did it avoid ending up with an unusable mouth clogged with ever larger teeth? Did the early teeth (and early part of the whorl) get absorbed?

        1. I suppose the answer would be an ever larger mouth in an ever larger fish… But I too thought it logical to suppose small teeth grew into large ones – otherwise there would have to be some very complicated changes in the ancestral sharks?

          1. I don’t want to beat this to death, but perhaps a fruitful way to think of it is as a kind of serrated spiral tusk rather than a row of individual teeth. On this view, the tusk grows from the big end, where it’s attached to the jaw (as you would expect), and produces new serrations as a side effect of that growth. The small serrations we see farther along the spiral are not new growth; they’re remnants of an earlier stage of growth when the entire structure was smaller.

            It just so happens that this tusk’s curvature brings it back into a pouch in the animal’s jaw, instead of outside the mouth like an elephant’s or narwhal’s tusk.

            I see nothing about this way of looking at it that’s particularly counterintuitive or evolutionarily challenging compared to other, similar growth patterns such as snail shells or ram’s horns.

        2. Comparison to modern sharks don’t seem to be relevant since these guys are closer to ratfish than to sharks. So it seems this is maybe not a case of the shark-pattern reversing itself.

          1. Not just sharks, though. The suggested growth pattern implies (?) that the whorl remains basically static while a new tooth grows to full size, then I suppose moves one step along to start the process again. Just seems odd. How would precurors function?

        3. No, they didn’t get absorbed. Look at the pictures; there they are at the center of the spiral, like the tiny abandoned chambers at the center of a nautilus shell.

      2. Sorry, my bad; it was pretty late when I commented last night, and I’d only taken a quick look.

        Yes, ever-larger teeth form on the outside of the spiral, and after functioning as they move forward they get buried in cartilage instead of being shed (or even resorbed), ever. That really does seem quite strange among vertebrate dentitions, but Gregory K’s comparison to mollusc shells and vertebrate horns and tusks is apt. Rattlesnake rattles are also loosely analogous, growing as discrete (but flexibly interlinked) segments of ever-larger diameter.

  5. Yah, sorry, but this tooth is *obviously* intelligently designed to slice and eat the pizzas cooked for them by the descendants of Adam who worshipped sharks because they are so sodding cool.

  6. I remember seeing a Shark Week show on the Discovery Channel a long time ago on these critters. The artist who was on the show inexplicably chose to render it as having a long, eel-like body.

  7. I may be missing something here, but I’m left wondering how this fish actually got chunks of food into its maw. That tooth-spiral looks like it’d make a great single cut through prey, but then what? That wouldn’t free any decent-sized pieces of food. Did it start at one end of a cephalopod, or whatever, and slice pieces off crosswise? Did it subsist on the tiny scraps freed in the cutting process? It’s cool, to be sure, but it doesn’t seem like a very efficient feeding apparatus.

  8. First of all, this is not a shark. It’s a (stem-)ratfish. That’s one of the two biggest points of the paper!

    Also, of course, the logarithmic spiral represents successive stages of growth, which proceeds in the usual way from small to large, i.e. ‘NEW TEETH FORMED HERE’ in the Nat Geo figure is exactly wrong; teeth are SHED from the outside extremity of the series. (Otherwise, where would they be shed to?)

    No, the teeth aren’t shed at all. Ever. It’s like a tuatara, except there the toothrow is straight: all teeth stay where they are, new and bigger teeth are added close to the jaw joints where the jaw grows, the baby teeth remain near the tip throughout life (well, in tuatara they’re eventually worn down to the bone).

    How would precurors function?

    As John Scanlon said, tooth whorls – series of teeth where old ones (near the front/outside) are not immediately shed while new ones are added at the other end – are actually normal for toothed vertebrates, the big difference is just that Helicoprion has only one whorl in total. Mobile tooth whorls, like this one, also have a precedent: in the onychodonts, bony fish from the Devonian that were fairly closely related to coelacanths, each side of the lower jaw bore a whorl of three very long teeth on a separate mobile bone that formed a joint with the other jaw bones. Being a cartilaginous fish, Helicoprion has the same without the bones – except it has one whorl in the middle of the lower jaw, not two (one on each side of the midline) like an onychodont.

  9. Thanks. I don’t doubt it – I’m no biologist, far less an evolutionary biologist. It just seems odd, having a single tooth-growing site that holds the tooth until it’s fully grown, then presumably shunts the whole tooth row one space along to start another tooth. You live and learn.

    1. I think you’re making it more complicated than it needs to be. Seems more likely to me that there’s some process that extrudes new root material continuously, and when a big enough stretch of new root has been exposed, a crown spontaneously forms there. There is no “single tooth-growing site”; each new tooth crown is a moving instance of such a site, budding off one after another from the single root-growing site.

      1. I was probably influenced by an inappropriately analogous case – elephant dentition – where teeth yet to grow to functional size are present as buds, in line, which would show in a preserved or fossilised jaw: i.e. multiple (and continuously advancing) growth sites. (It was this lack of immature tooth buds in the published description that led me to the erroneous first impression that the smaller teeth in the whorl were spiralling out and growing. Anyway, log me as surprised – or caught out – as opposed to sceptical.)

Leave a Reply