A Blog by Ed Yong

Shark Dads Lose Babies to Unborn Cannibal Siblings

Inside its mother’s womb, an unborn sand tiger shark is busy devouring its brothers and sisters. It’s just 10 centimetres long but it already has well-developed eyes and a set of sharp teeth, which it turns against its smaller siblings. By the time the pregnant female gives birth, it only has two babies left—one from each of its two wombs. These survivors have already eaten all the others. They’re the bloody victors of a pre-birth battle.

The arched back, upturned snout and protruding teeth of a sand tiger shark give it a particularly brutish look. Its reproductive habits don’t help. After sex, any fertilised eggs settle in one of the female’s two uteri. Like a quarter of shark species, the sand tiger never lays these eggs. Instead, they hatch inside her once they reach a certain size.

Timing matters. The first embryo to emerge in each uterus—the ‘hatchling’—always cannibalises its younger siblings. It’s so voracious that at least one scientist has been bitten by a sand tiger pup while unwisely sticking a finger in a pregnant female’s uterus.

The cannibal not only nourishes itself on its siblings’ bodies, but also gains sole access to the nutritious supply of unfertilised eggs that its mother provides. On this rich diet of yolk and flesh, the hatchling grows at a tremendous pace. When it is eventually born, it’s already a metre in length—that’s big enough to protect it from many predators. “Only really big sharks eat baby sand tigers,” says Demian Chapman from Stony Brook University in New York.

Chapman studies the mating behaviour of sharks, and was fascinated by what the sand tiger’s bizarre practices mean for males. If a female mates with many males, her litter could initially include pups from many fathers. But it’s entirely possible that the pups of some males are cannibalised by the pups of others, before they’re even born! So if you gave sand tiger pups a paternity test, what would you find?

Chapman studied the bodies of 15 pregnant sand tiger sharks that died near South Africa’s beaches after getting caught in protective nets. (The nets were put up after a series of shark attacks in the 1960s.) He took tissue samples from mothers and embryos, and analysed microsatellites—short, repetitive pieces of DNA that are used in family tests—at 10 places in their genomes.

A sand tiger shark hatchling next to an embryo (the smaller one on the top left). Credit: D.ABERCROMBIE.
A sand tiger shark hatchling next to an embryo (the smaller one on the top left). Credit: D.ABERCROMBIE.

The results showed that females regularly mate with at least two partners. In cases where the hatchlings hadn’t finished devouring their siblings, the embryos were fathered by two or more different males. And when only the hatchlings remained, they were often half-siblings rather than full ones.

Chapman’s results showed that, yes, some males manage to mate with females but never actually contribute to the next generation of sharks. Their young are devoured in the womb.

This is yet another reminder that sex is just one step towards actual reproduction. In many animals, females exert a surprising amount of choice over who fathers their young. Even after sex, females can store the sperm of different partners in separate compartments and determine which ones get to fertilise her eggs.

For males, this means that sexual competition continues after sex. It’s not just about finding mates, but about ensuring that your sperm fertilises her eggs. This leads to fierce “sperm competitions” and bizarre adaptations, where males scrape away the sperm of past mates, guard or plug females so they can no longer accept partners, “traumatically inseminate” her through her back, or even poison partners with toxic sperm to limit future sexual encounters.

The sand tiger shark’s “embryonic cannibalism” takes these competitions to a whole new level. Even if a male successfully fertilises a female’s eggs, there’s still no guarantee that his offspring will actually be born. This may explain why, in captivity at least, male sand tigers often guard the females they mate with. You’d also expect natural selection to also favour males whose offspring grow particularly quickly, so they become the cannibal hatchlings rather than the devoured runts. This may explain why sand tiger embryos have evolved to have such unusually well-developed eyes and teeth.

Females, meanwhile, may not need to bother with carefully choosing their mates. By having sex with several males, and just letting their offspring duke it out inside her, she ensures that the she gives birth to the offspring of the fittest partners.

Reference: Chapman, Wintner, Abercrombie, Ashe, Bernard, Shivji & Feldheim. 2013 The behavioural and genetic mating system of the sand tiger shark, Carcharias taurus, an intrauterine cannibal. Biol Lett http://dx.doi.org/10.1098/rsbl.2013.0003

More on animal sex:

13 thoughts on “Shark Dads Lose Babies to Unborn Cannibal Siblings

  1. What’s nice about this example is it nicely transitions through conflict between fathers to conflict between siblings (whose interests should align more as they have both inherited alleles from mum). For a nice review of sexual conflict and sib competition in the sea see http://www.sciencedirect.com/science/article/pii/S0169534710001072 . It would be interesting to know (but likely difficult) whether multiple mated broods are smaller, i.e. if there is conflict between mother and offspring.

  2. I’ve been thinking more about this and wondering if intrauterine cannibalism is what drove the evolution of two uteri. It makes sense as it is the only way to have a litter more than one.

    You’ve written before about the dual-uteri in marsupials, Ed. I’m betting this is an example of convergent evolution. Do we know anything about how/why the kangaroos acquired theirs?

  3. I wonder if the sharks can tell whether or not another male has already mated with a female. It would be to the male’s advantage to be the first one in, so to speak, so his progeny develop first and can devour any latecomers.

  4. Regarding Mike’s wonderings:

    I would highly suspect that inter-uterine canninbalism could explain marsupial dual uteri. It doesn’t seem very surprising to me (other than being used to humans only have one uterus) that other species might have two. Human females also have two kidneys, two ureters, and more related, two ovaries and two fallopian tubes. In fact, it seems rather strange to just have one uterus and one vagina, given the dual pipes that lead up to it. We might instead ask why eutherians have only one uterus and vagina (and one penis). It might be more efficient to have one delivery and acceptance point for sperm than two. Or any other number of fun conjectures.

    I would consider asking some more basic questions about the sharks rather than jumping to mammals. For example, are dual uteri common in sharks? Do they also have a pair of ovaries? I would guess so. Is there any evidence in the sand tiger shark that dual uteri was not the ancestral state? And how common are viviparous sharks compared to oviparous? Is live birth among sharks a recent evolutionary development? And so on.

    Maybe I should consider finding some shark researchers to collaborate with – this is really interesting stuff!

  5. Regarding Kathy Orlinsky’s query as to whether males would be able to tell if females have been mated or not I would suggest that they can. Sand Tigers (called Ragged Tooth sharks or, affectionately, Raggies in South Africa) have quite violent mating behaviors where the male grips the female’s fin in his (rather sharp) teeth leaving quite impressive scars. We see them regularly when diving here in the winter and the fresh mating wounds on the females are very obvious so should be visible to the males as well. The females actually have a thicker skin to be able to withstand the courtship. They are extremely placid with divers though – a dive buddy of mine accidentally swam head first into a ~3m male (low vis and no bubbles on a rebreather) and it just sauntered off.

  6. @ Mike and Melissa: Paired uteri are actually the more common condition in vertebrates–humans are unusual in having one central uterus. Most eutherian mammals, including cats and rabbits, have paired uteri. So it’s not that paired uteri evolved in response to cannibalism–they started out that way in the first place. Presumably, the single, midline uterus of humans (and maybe other apes–I don’t remember off hand) evolved in order to better support one large fetus. Remember, when asking questions about the adaptation of a feature, you first have to figure out the starting point of that feature.

    1. Thanks for the clarification and reminder Karen.

      In my reading about this I did discover a couple recent cases of women who have successfully given birth to babies from different uteri, so it seems to still happen in humans from time to time.



    2. Thanks Karen. Except my typo made it appear otherwise at first glance, that’s exactly what I was trying to get at. I learned something new. Do you know of any good references to learn more about mammalian uteri? I’m an evolutionary biologist, but mostly study molecular evolution. The anatomy is pretty cool.

  7. Wow! You guys are amazing… Who would’ve thought? That a few cm unborn Shark can eat other unborn Sharks I mean its crazy but to learn that a female Sand Shark gat to choose which male sperm she wants it to get through? Its just WOW! I Love NG!!!

Leave a Reply

Your email address will not be published. Required fields are marked *