A Short Piece on Piscivores – Not All The Same


So you think you know a piscivore if you saw one? Not so fast. Take a look:

An array of extinct and extant specialist on fish-eating.

An array of extinct and extant specialist on fish-eating.

Piscivores come in a large array of sizes and morphologies. Not all have teeth. Some are slender-snouted, others broad. What mostly defines a piscivore is that it consumes fish, but not all do it the same way, and for some, it’s purely behavioral, sometimes seasonal, and sometimes based on relation to plentiful food.

Some piscivores, including many piscine ones, acquire prey through engulfing their food whole. Indeed, some fish may be classed base on how large a gape they can manage, when it comes to eating the next fish down on the food chain. But this doesn’t preclude piscivory in smaller-gaped fish, merely that larger prey cannot be taken. Other piscivores trap their prey in their jaws, using serrated teeth, needle-like teeth, or serrated margins of their jaw, to create high-pressure points along the jaw margin that prevent the slippery prey from wriggling free.

And both of these methods are important, because most piscivores, being largely aquatic by rule, tend not to use any other limb to help restrain prey. From penguins to fish to whales to crocodilians, the jaws are pretty much all you got. This tends to cause piscivores to develop similar adaptations to fish-catching. For one, there’s gape, but this more depends on the relationship of prey-size to predator. Another have been the longness or slenderness of the teeth, which may involve either an array of long, piercing teeth or shorter, recurved teeth.

We see the latter in Hydrophis, the fish-specialist ocean-going snake, which uses the typical serpentine oral-transport mechanism of “walking prey into the mouth” through the use of flexible joints in the skull. Gharials have long, interlocking teeth, but they mostly orient in the same direction: slightly procumbent, but curved backwards, and when closing the jaws come together so the tips of the teeth are at the same angle. This makes them very efficient “fish-catchers.” In contrast, presumed piscivores like the extinct elasmosaurs had teeth that broadly splay out of the jaw and had far less curvature. The jaws of the elasmosaur Aristonectes involves much smaller, finer, and less splayed teeth, which would make them comparable to the gharials, despite have a shorter, rounder, and triangular mouth.

One thing that is often brought up in comparison to piscivores is the presence of a radiating row of procumbent teeth at the front of the jaws. However, this, too, varies among aquatic piscivores. The precise mechanism by which procumbent teeth would preclude piscivory is often never explained, though. Instead, it is presumed through analogy, that forward-sloping teeth occur in piscivores, therefore piscivores should have them. This bit of backwards thinking has led to arguments that, for example, some dinosaurs (spinsoaurs, the abelisaur Masiakasaurus, Microraptor, etc.), pterosaurs (e.g., Rhamphorhynchus), and even birds (e.g., Yanornis) were piscivores. This “fact” — incidence of forward-sloping teeth relative to diet — has never been qualified across vertebrate taxa. And in fish, it’s generally not true: most piscivorous fish either lack teeth (as in marlins) or bear large teeth which are arrayed roughly perpendicular to the jaw margin. Especially in smaller fish, the teeth may even curve backwards, as it does in watersnakes, due to the behavior of oral prehension. In those animals in which food is engaged by the jaws, forward sloping teeth make more inhibitors of prey escape. Instead, rearward sloping or curving teeth, as in most carnivorous terrestrial vertebrates, serve to inhibit, as it were, egress from a gaping maw. Even the teeth of fine-food consumers, such as whale and basking sharks, point rearward.

Most of the issues with piscivores in aquatic environments is the segregation of them into two morphotypes: The prehension type, and the engulfment type. Prehensors attempt to restrain prey through use of their jaw shape and/or teeth, whereas engulfers instead attempt to swallow their prey whole. Prehension isn’t even restricted to the jaws; fish-eating bulldog bats, ospreys, sea ernes, and flatheaded cats all use claws to catch prey. These predators, however, possess additional limbs; comparison of aquatic oral-only prehensors with terrestrial oral- or limb-based prehensors has yet to receive much scrutiny.

So it’s no surprise that for piscivores that have the use of their limbs for catching, subduing, or even processing the carcasses of fish (say, brown bears in spring) they tend not to follow these rules of order for the morphology of the jaws. Why, then, should we assume they do? If teeth that do not work to impale evenly along the jaw, that do not recurve towards the oral margin, that work when biting, make poor teeth for catching prey upon.

Gape is certainly an issue in fish-catching, but it has more to do with relative size of the bits of food going down the gullet. The jaw should not be more capable of holding prey than the throat can swallow; it’s sorta the first stop on the way to the stomach. More than one fish has been found having choked to death while trying to swallow prey it barely managed to pass through the mouth, and this is true for snakes as well. If the animal doesn’t regurgitate its prey, it will die. Terrestrial animals tend not to involve these factors because many of them develop additional methods of processing prey than mouthing at it with rows of teeth (like many sharks do). They tear at it, pick it apart, then swallow it piece by piece. So while the piscivorous osprey has very recurved talons, it still rips its prey up. Less so the egret, with its relatively small and straight claws, so it swallows prey, and gape is less of an issue that the width of its mandibles, providing space into the gullet. That is also true of the pelican, which will swallow whole other birds that can barely fit into its pouch. At that point, the actual limiting factor is the size of the crop and the width between the shoulders including furculae, passing food from the crop into the stomach.

So what does this tell us? Why the chart above? It suggests that there is no one true piscivore morphology, though it’s a little cherry-picked. It’s largely behavioral. Those animals with piscivorous habits will tend to, but not always, have very curved prehensile appendages (jaws or claws) that close in on one another. Food is restricted to the volume of the throat, not necessarily the wideness of the gape. I did try to look at unconventional piscivores, animals that counter my arguments. I also excluded pterosaurs from the chart because of the wide variety of jaws and the problems of making assumptions for the reasons stated above. The best, overall indicator for piscivory, though? Is the preference of prey in relation to behavior, access to prey, and a morphology consistent with it — and this is without weighting these options. For these reasons, I’d pick Pterodactylus and Rhamphorhynchus as equally indicative of piscivory. If I weighted association higher, the latter would be a piscivore, but not the former; and if I weighted morphology higher, the inverse.

It’s certainly something to consider.

About these ads
This entry was posted in Biological Comparison, Biology, Paleoecology, Paleontology, Terminology and tagged , , , , , , , , , , , , , , , , . Bookmark the permalink.

12 Responses to A Short Piece on Piscivores – Not All The Same

  1. Nice! Richardoestesia was probably a piscivore (not to mention Spinosaurus). :)

  2. OK; how many animals with gharial-like adaptations, do not eat fish?

    • Well, that depends. Fish-only diet, or fish-major diet? If it were elongated, toothy jaws with relatively generalized dentition but not piscivores, per se? Many mammals: sperm whales (theuthophage), shrews and a fair number of small insectivorans (as suggested by their name), and even some birds such as the pelagornithids wouldn’t have been specifically piscivores. When it comes to merely the shape of the jaw, then a larger number of birds, including most storks, fall under this header: not piscivores, per se.

  3. Well, spinosaurids at least seem pretty likely to be fish eaters—Baryonyx definitely ate fish at least sometimes and they may have been semiaquatic. Maybe when Sereno publishes on his articulated noasaurid, people will stop thinking Masiakasaurus was a piscivore.

    I’m a little bit confused about where the “heron unenlagiiine” paleoart meme came from, now that I think about it.

    • Baryonyx, as well as spinosaur teeth from north of the Sahara, fulfill three criteria for an inference of specific diet: Association (fish remains collected with), Access (occurs almost exclusively in shallow-water near-shore or deltaic/estuarine/paludal environments), and to a degree morphology (bearing a series of directly opposing upper and lower dentition acting as a prehensile tool; additionally, isotope analysis of spinosaur remains suggest consuming aquatic prey). Behavior is not something we can predict so far without trace fossils, and these are lacking in this case. The same is true for animals like Tanystropheus (assumed to be a piscivore, but not aquatic). These trickier questions are the reasons some fossil animals assumed to be piscivores are omitted from the montage.

  4. Wonderful post, Jaime. One of the best overview on the subject, I must say: clear and concise, you offer lots of examples and it’s a joy reading your post. Kudos to that.

    However, I’m not a 100% sure about the absence of some species in your diagram.
    You wrote in the comments: “Baryonyx, as well as spinosaur teeth from north of the Sahara, fulfill three criteria for an inference of specific diet: Association (…), Access (…), and to a degree morphology (…) […] Behavior is not something we can predict so far without trace fossils, and these are lacking in this case.”
    According to this rule of yours, neither Cryptocleidus nor Mesosaurus should be included since that their behaviour is unknown (I’m not sure about Prionosuchus and Mastodontosaurus, since their behaviour can be inferred by their morphology or by looking at modern analogues). Am I missing something here?

    Nevertheless, beautiful post as always.

    • The major reason I didn’t include dinosaurs or even pterosaurs is because they should be part of the argument for inference I wanted to approach with this. They are more … interesting … to some. So applying this rule to some taxa should result in some estimate of inference. The other taxa shown are aquatic, and would exhibit a higher likelihood of being piscivores than terrestrial animals. Speculating on the piscivory of more terrestrial organisms is another mess that many aquatic taxa don’t have to deal with.

  5. And regardless I’m not certain if Pterodactylus was piscivorous anyways. Compared to other pterosaurs, there’s a higher degree of likelihood because of it’s classical ctenochasmatoid traits of webbed and large feet, which means it could wade and swim, but many modern wading and swimming birds aren’t specialised piscivores, such as coots, phalaropes, geese, spoonbills, ad infinitum.

    Unlike Rhamphorhynchus, to my knowledge there aren’t specimens of Pterodactylus with fish within, so for all we know it truly wasnt a piscivore.

    • Part of the difficulty in this is that most piscivores are so determined by behavior, not by some strict morphology. Piscivory is tricky because the major morphological indicators for it are actually adaptations for catching fleshy food while swimming. “Piscivory” being a diet (behavioral) consisting of aquatic, slippery prey. How one animal acquires fishy or otherwise large aquatic prey must differ from others, but they’d all be “piscivores.”

      As with the presumption that Rhampho is piscivorous because some specimens have fish associated with them, I am skeptical that Ptero isn’t because it lacks fish association. Taphonomy happens, and not all “proper” prey is preserved. Otherwise we’d tend to assume ichthyosaurs were all or mostly cannibals. We think we know better than that.

      • Which is why I think traits usually argued for Pterodactyclus’ supposed piscivory don’t necessarily build up to specialised fish-eating.

        Sure, it may be that known specimens simply didn’t happen to preserve an individual with prey within. However, when there are several Rhamphorhynchus specimens with fish and possibly even cephalopods inside, it may simply suggest that Pterodactylus ate other things, given how large this dichotomy is.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s