by Bill Schutt
But it’s not just the abundance of eggs and young that makes fish such a popular menu item for members of their own species. Many terrestrial vertebrates produce few or even a single offspring, and most of these newly born or newly hatched individuals already exhibit considerable body size. In many fish species, the extraordinary number of eggs produced imposes a limit on their size, and so a full-grown cod might be a million times (or six orders of magnitude) larger than its own eggs. This fact goes a long way to explain why the majority of them exhibit about as much individual recognition of their offspring as humans do for a handful of raisins. Fish eggs, larvae, and fry (i.e., young fish) are vast in number, minute in size, and high in nutritional value. This makes them an abundant, nonthreatening, and easily collected food source. It’s also why ichthyologists consider the absence of cannibalism in fishes, rather than its presence, to be the exceptional case.
Although not quite as infrequently practiced as it is among the invertebrates, parental care occurs in only around 20 percent of the 420 families of bony fishes (a group composed of nearly all living species except sharks and their flattened relatives, the skates and rays). The primary reason for this trend can be explained by the fact that the natural world is full of tradeoffs. Here the tradeoff works like this: Since females expend a tremendous amount of energy producing eggs (sometimes millions of them), they can’t afford to expend much energy caring for them or their young when they hatch. For this reason, the eggs and fry of most fish species exist in dangerous environments inhabited by a long list of potential predators, including conspecifics. But even in the 90 or so piscine families where parental care does occur, filial cannibalism is an extremely common practice, and here the primary reason has to do with who is doing the babysitting.
Among the land-dwelling vertebrates, females are the principal caregivers, while males take on support roles or simply make themselves scarce. In bony fishes that guard their own eggs, though, it’s usually the males who are involved, undertaking these chores at nests otherwise known as oviposition sites. These can range from slight depressions in the substrate, to rocks, plants, and other materials to which the sticky eggs (generally numbering in the hundreds) adhere in discrete clumps. The male guardians often wind up consuming some of the eggs (partial filial cannibalism), and sometimes all of them (total filial cannibalism).
One reason that male fish engage in this seemingly counterproductive behavior is that generally, they have much less invested in the brood than do females. It is less costly to produce a cloud of sperm than it is to produce, carry around, and distribute an abdomen full of eggs. Furthermore, with their ability to search for food seriously constrained by caregiving duties, males are forced to undertake at least some degree of fasting. This practice decreases their overall physical condition and thus the likelihood of future reproductive success. By consuming a portion of their own brood, males can increase the chances that they’ll survive and produce additional offspring. New eggs are consumed more often than older eggs because there has been less parental investment in maintaining them.
In some examples, though, the loss of eggs from an oviposition site is not the fault of a hungry male guardian. Unrelated conspecific males regularly raid nests in order to consume or steal eggs. Egg theft can be explained by the preference for some females to spawn at sites already containing eggs, even if they’re not hers. In these instances, once a female deposits her own clutch, the male will selectively eat the eggs he previously stole and deposited there.
While we’re on the topic of parental care in fishes, mouthbrooding cichlids deserve a brief mention, if only because they serve to strengthen the often tenuous link between a mouthful of kids and lunch. Mouthbrooding occurs in at least nine piscine families, most famously in the freshwater Cichlidae. Cichlids, especially the African varieties, are extremely popular with aquarium keepers, as well as connoisseurs of tilapia—the Spam of gourmet fish. With more than 1,300 species, cichlids have evolved extremely specialized lifestyles that serve to reduce competition with related species living in the same area.
Mouthbrooding is a common form of behavior in cichlids. Typically, it refers to post-spawning behavior in which parents (usually females) hold their brood of fertilized eggs inside their mouths until they hatch and sometimes even after that. This provides the eggs and fry with a haven from predators, a point commonly portrayed in crowd-pleasing nature videos that depict young fish darting back into their parent’s mouth at the first sign of danger. Conspicuously missing from these lighthearted reports is the fact that parents holding a mouthful of eggs usually eat a considerable portion of them, and sometimes the entire brood. Also destined for the digital equivalent of the cutting room floor are shots showing male cichlids fertilizing the eggs in the females’ mouths, always a difficult topic to explain during family TV time.
Mouthbrooders practice filial cannibalism primarily because, as we all know, eating a regular meal is next to impossible while carrying around a mouthful of eggs. Cichlids and other mouthbrooders get around this vexing problem in the simplest way possible: cannibalism. Interestingly, scientists had thought that for the first few days after spawning, female mouthbrooders selectively consumed only unfertilized eggs from their broods. When researchers set out to determine just how mothers were able to distinguish between fertilized and unfertilized eggs, they were surprised to find that 15 percent of the consumed eggs were actually fertile. We now know that, mistakes aside, once all of the unfertilized eggs have been eaten, hungry mothers continue to consume small quantities of their own fertilized eggs. And should the brood reach about 20 percent of its original number, many mouthbrooders will write off the entire batch, and eat them all. As with similar examples of total filial cannibalism, this usually occurs when the cost of caring for the brood becomes higher than the benefit of producing a less-than-normal number of offspring. Rather than investing in a smaller brood, it becomes more advantageous for the female to recover some energy by consuming her remaining young and then moving on to find a new mate.
My personal favorite example of piscine cannibalism is yet another instance in which immature animals are the ones getting consumed. But in sand tiger sharks (Carcharias taurus), the individuals doing the cannibalizing haven’t even been born yet.
Sand tigers, like hammerheads (Sphyrna zygaena) and blue sharks (Prionace glauca), do not deposit their eggs into the environment. Instead the eggs and young develop inside the females’ oviducts, a developmental strategy known as histotrophic viviparity. Scientists who first looked at late-term sand tiger embryos in 1948 noticed that these specimens were anatomically well developed, with a mouthful of sharp teeth—a point (or several) driven home when one researcher was bitten on the hand while probing the oviduct of a pregnant specimen. Strangely, these late-term embryos also had swollen bellies, which were initially thought to be yolk sacs—a form of stored food. This was puzzling, though, since most of the nutrient-rich yolk should have been used up by this late stage of development. Further investigation showed that the abdominal bumps weren’t yolk sacs at all, they were stomachs full of smaller sharks! These embryos (averaging 19 in number) had fallen victim to the ultimate in sibling rivalry—a form of in utero cannibalism known as adelphophagy (from the Ancient Greek for “brother eating”), or sibling cannibalism.
This behavior is possible because sand tiger shark oviducts contain embryos at different developmental stages (a characteristic that also evolved in birds). Once the largest of the shark embryos run through their own yolky food supply, they begin consuming eggs. And when the eggs are gone, the ravenous fetal sharks begin consuming their smaller siblings. Ultimately, only two pups remain, one in each oviduct. According to renowned ichthyologist Stewart Springer, the selective advantage for the young sharks may extend beyond the obvious nutritional reward.
Springer, the first to study the behavior, believed that the surviving pups were born “experienced young,” having already killed for survival even before their birth. H
e hypothesized that this form of sibling cannibalism might afford the young sand tigers a competitive advantage during interactions with other predatory species also looking for meal.
Although the sand tiger is the only species known to consume embryos in utero, several other sharks exhibit a form of oophagy, in which the unborn residents of the oviduct feed on a steady supply of unfertilized eggs. Additionally, a form of adelphophagy occurs in some bony fishes (superclass Osteichthyes) in which broods mature at different rates. Once again, in these species it’s the older members of the brood that cannibalize their smaller siblings.
Cannibalism of the young also occurs in many species of snakes, lizards, and crocodilians, where, for example, it accounts for significant juvenile mortality in the American alligator (Alligator mississippiensis). Although reptiles do not transition through larval stages like most fish and amphibians, the smallest and most defenseless individuals, namely eggs, neonates, and juveniles, run the greatest risk of being eaten by conspecifics.
Cannibalism is relatively rare among birds, a fact that may be related to one aspect of their specialized anatomy—their beaks. These keratinous structures are responsible for the designation of most bird species as “gape-limited predators.” In other words, their lack of teeth limits them to consuming prey small enough to be swallowed whole. Existing under this anatomical constraint, when cannibalism does occur in birds it’s generally the eggs and young that are consumed.
According to Cornell ornithologist Walter Koenig, “Since brood reduction is widespread in birds, it’s likely that sibling cannibalism would be a lot more widespread than it is if birds had beaks that were capable of tearing dying offspring to pieces, or opening their gape wide enough to swallow them whole.”
Heterocannibalism (in which non-kin get eaten) has been reported in seven of the 142 bird families and is most common in colonial sea birds like gulls. Here, the practice of consuming eggs or young is an integral part of foraging strategy and it can have a significant effect on bird populations. In one study of a colony of 900 herring gulls (Larus argentatus), approximately one-quarter of the eggs and chicks were cannibalized. Heterocannibalism also occurs in acorn woodpeckers (Melanerpes formicivorus). In this species, two female woodpeckers share a single nest and will even feed and care for each other’s young. But before that occurs, the nestmates may also destroy and consume each other’s eggs if one bird should lay an egg first. Presumably this is because the oldest hatchling would be the most likely to survive. To eliminate this advantage, the birds will keep eating each other’s eggs until they both lay their eggs on the same day, a process that can take weeks.
Sibling cannibalism, in which brothers and sisters get eaten by other brothers and sisters, is best known among the raptors, a non-taxonomic name for predatory birds like eagles, hawks, kestrels, and owls, all of which possess strong eyesight, powerful beaks, and sharp talons. The latter two characteristics exempt raptors from the gape limitations seen in many other birds and may help explain the increased frequency of cannibalism in these birds. In some species, sibling cannibalism is often the end result of asynchronous hatching, in which two eggs are laid with one of them hatching several days before the other (one can almost hear the acorn woodpeckers gasping in horror). As a result, the firstborn chick uses its extra bulk to win squabbles over food with its younger brother or sister. In instances where the parents are unable to provide their young with enough to eat, the firstborn will kill and consume its younger sibling. Researchers sometimes refer to these types of victims as “food caches,” as sibling cannibalism becomes an efficient way to produce well-nourished offspring (albeit fewer of them) during times of stress.
Something similar happens in the snowy egret (Leucophoyx thula), which commonly lays three eggs. The first two get a serious dose of hormones while still in the mother’s body. The third egg receives only half the hormone boost, resulting in a less aggressive hatchling. If food is abundant, the larger nestlings simply throw the passive chick out of the nest, but if alternative sources of nutrition become scarce, the smaller sibling is stabbed to death and eaten.
According to Koenig and fellow ornithologist Mark Stanback, filial cannibalism in birds has been reported in 13 of 142 avian families but is not well understood, perhaps because it is relatively infrequently observed. On rare occasions, birds like roadrunners (Geococcyx californianus) will eat undersized chicks. Similarly, barn owls (Tyto alba) are reported to consume their own chicks during extreme environmental conditions, such as when food is in short supply and the chicks are either starving or sick. It has been suggested that filial cannibalism of dying or decayed offspring can prevent infection and deterioration of the entire clutch. Presumably there are also benefits to getting rid of dead chicks before they attract legions of carrion-eating flies and maggots. In most cases, however, it’s the lack of alternative forms of nutrition that initiates the behavior.
“A lot of examples of cannibalism in birds are clearly associated with food limitation,” Koenig told me. “We’re basically talking about a lifeboat strategy, where the strong cannibalize the weak.”
* * *
8 Insectivora and other traditional vertebrate classes like fish, reptiles, and birds are no longer considered to be valid units of classification, either because scientists have determined that species included in them aren’t as closely related as they once believed, or that species that should be in those classes aren’t.
3: Sexual Cannibalism, or Size Matters
Whoever authorized the evolution of the spiders of Australia should be summarily dragged out into the street and shot.
— Mira Grant, How Green This Land, How Blue This Sea
While it’s fairly common knowledge that the praying mantis is the co-holder (along with the black widow spider) of the title “Nature’s Most Infamous Cannibal,” fewer people know that the name praying mantis is shared by nearly all of the 2,200 species making up the order Mantodea.
The moniker comes from the curious manner in which the insects hold their forelegs while resting. As a result of this prayerlike attitude, they’ve become some of the most popular insects in mythology and folklore. Many of these mantid myths have religious or semi-religious overtones. In France, par exemple, they’re known as prie-dieu and are said to point lost children homeward. The Khoi people of South Africa regard praying mantises as gods, while Arab and Turkish folklore holds that the insects direct their prayers toward Mecca. Always looking on the bright side, Americans once believed that praying mantises blinded people and killed horses, this perhaps as a nod to the fact that rather than being used for prayer, the anterior-most limbs are actually modified into lethal, spike-covered weapons. Often well camouflaged, most species are ambush predators, lashing out with their “raptorial legs” to capture, crush, and secure their prey while a set of sharpened mouth parts slice, can-opener style, through the toughest exoskeleton.
Although mantises feed primarily on other insects, the largest species can reach around six inches in length and these giants will attack and consume small reptiles, birds, and even mammals. It is likely that this type of predatory behavior is responsible for the common misspelling “preying mantis.”
As a child in the 1960s I was told that there would be a $50 fine for anyone caught killing a praying mantis (and my friends have the same recollections). Since I was unable to uncover a record of any such federal or state law, I can only assume that the story was a scare tactic designed to keep nasty little boys from slaughtering an uncommonly pious insect known to eliminate an array of less religiously inclined pests.
Many people are familiar with the praying mantis’s supposed penchant for cannibalistic sexual encounters, reports of which began showing up in the scientific literature in the late 19th century. Back then, several authors claimed that female mantises regularly bit off the triangular heads of their partners during sex. These same sources also claimed (to many a reader’s astonishment) that the decapitated males continued to copulate, abdo
mens pulsing away as if nothing of much importance had just happened. According to these references, several hours later the female would stride off, full and fertilized, while the male, having been reduced to a tiny pile of wings and hard bits, stayed put. Similar tales about mantid mating continued into the early 20th century, when members of a new generation of entomologists began investigating the function of this rather puzzling behavior.
One hypothesis reasoned that the male mantis’s brain actually inhibited sexual behavior. With their heads removed, however, males became “disinhibited,” found the rhythm, and eventually pumped out a full load of sperm. Other mantid mavens suggested that getting oneself cannibalized made sense for praying mantis males that might have limited opportunities to mate over their lifetime. It made evolutionary sense, therefore, to fatten up the only female they might ever run into—especially one now carrying their sperm. Furthermore, it would be a plus for both sexes since headless males reportedly pumped out more sperm than those equipped with heads, leading to more fertilized eggs and more offspring. These accounts contributed to an overall impression that the decapitation of male mantises was a normal and perhaps necessary copulatory stage and, soon after, the concept became entrenched in textbooks and the popular literature. Unfortunately, what never quite made it into print was the fact that most observations of mantis cannibalism were made in laboratory settings and only after females had been deprived of food.
In reality, cannibalism varies across this large and diverse group. The behavior has gone unobserved in most species, not necessarily because it doesn’t happen, but because it hasn’t been studied. Researchers now believe that rather than being a required component of mating behavior, the consumption of males is more likely to be a foraging strategy employed by hungry females unable to wrap their raptorial forelegs around an alternate form of nutrition.
