Birdeaters and mouse catchers: spiders preying on vertebrates

The darkest of nights: midnight in the Brazilian rainforest. At ground level, the jungle is dusky even during the day, when sunlight barely peers through the dense canopy; after sunset, though, it is pitch black.
A rustle in the leaves; a predator on the prowl. Perfectly equipped for night hunting, he navigates the inky darkness with ease. The three-striped opossum may only be as big as a mouse, but he is nonetheless a bug's worst nightmare. As this tiny terror tirelessly patrols the forest floor, his teeth make short work of any arthropod he encounters. 

Coming down from a tree root, our little hunter freezes. His steps led him to a sinister-looking hole in the ground, where obscurity somehow appears even thicker than around it. Near the hole, the forest floor feels strangely soft under his paws, as if covered with some sort of mat. Even though no distinct smell or sound exhales from it, it seems to radiate a diffuse, looming threat. Better not stay here.

Sudden and overwhelming, the attack meets no resistance. In the blink of an eye, powerful legs grab the marsupial, and huge, sharp fangs pierce his side, delivering a fatal blow before the venom is even injected.
The giant spider was not in her burrow. She was lying in ambush near the entrance, perfectly motionless until she pounced on him from behind.
In a terrifying reversal of roles, the bird-eating tarantula feeds on the mammal.

"Bird Eater". An arachnophobe's nightmare: a giant spider, the size of a dinner plate, that feeds on mice and birds as if they were mere bugs... A sensational depiction, but how accurate is it?

Anna Maria Sibylla Merian (2 April 1647 – 13 January 1717) was a pioneer in many domains. She was one of the first naturalists, more than a century before the "age of scientific naturalism" (which is generally considered to be from the late 1700s to the late 1800s), to enter the history of science. She was also a female naturalist, at a time where describing the scientific community as a "boys club" would have been quite the understatement.

She became famous quite early on, for her scientific illustrations that featured both a unique pictorial style and remarkable anatomical accuracy, and for her studies on insect metamorphosis. 
Her real break into historical significance, though, came in 1697. At the age of 52, she embarked with her daughter on a daring voyage to recently colonised Suriname, thus becoming one of the very first western naturalists to explore the Amazon rainforest. 

Throughout her two-years stay in Suriname, she documented many things, from the atrocities endured by enslaved Natives in Dutch sugar cane plantations (although one thing she wasn't exactly a pioneer in was human rights, as, while her records are useful for historians to know about the exactions of the Dutch colonists, she didn't sound particularly shocked by them, nor was she reluctant to make use of slaves for her own purposes) to countless observations and illustrations of new species.

There she produced her most famous (and most beautiful) artworks, showcasing various butterflies and moths interacting with tropical plants, birds including toucans and hummingbirds, the infamous Suriname toad (Pipa pipa) that incubates its offspring inside the skin of its back, as well as reptiles like caimans, tegus and snakes.
However, her most notorious piece, the one that caused both awe and controversy, was one that featured a few different spiders.

 

Print of the plate #18 of Merian's Insects of Suriname showing an orb weaver getting attacked by ants, a huntsman and its egg sac, and two Avicularia tarantulas, one of them snatching a hummingbird from its nest (source: Natural History Museum, London)

 What made plate 18 so special is the fact that one of these spiders, a pinktoe tarantula (Avicularia avicularia) is shown catching a hummingbird.
Because many of the animals encountered and illustrated by Merian in Suriname were entirely unlike anything known to Europeans at the time, some of her observations were received with understandable (albeit probably excessive) scepticism; yet, it was not the bird with a beak as long as its body, or the toad that incubates its young under its skin, that was deemed "too fantastical to be true", but the spider big enough to feed on a bird.
For some reason, that was considered the hardest thing to believe about the weird and wonderful creatures of the Amazon, and it sparked criticism and doubts about the authenticity of this observation. 

It also created a vivid image in the collective imagination of Europeans, and an enduring stereotype.

Indeed, this illustration became so famous that it coined "Vogelspinnen" (bird spiders) as a common name for the family Theraphosidae in German (as well as Dutch and several other European languages, until "tarantula" came back from America as a popular common name for this family). 
In 1758, Linnaeus, the father of modern biology, named this spider Aranea avicularia ("the spider of the little birds") as an homage to Merian's drawing (it would later be renamed Avicularia avicularia). Even though he had actually never met one in the flesh, or even travelled to South America, he included the species in his Systema naturae, making it not only one of the very first spiders, but one of the very first animals, to be scientifically classified with modern methods!

In English, too, "bird-eating spiders" would become a common name for Theraphosidae alongside "tarantulas" (in fact, they were known as "bird-eating spiders" before "tarantulas", was adopted by American English a mere century ago), particularly to very large terrestrial species such as those in the genera Theraphosa, Lasiodora, Phormictopus, Acanthoscurria, Vitalius...
The association with birds is also present in the Spanish language, where "araña pollito" (chicken spider) is a common name for some Theraphosidae, most notably Grammostola and Pamphobeteus.
 

Lasiodora parahybana, one of the largest and heaviest tarantulas (and spiders), is known as the "salmon pink giant bird eater" in English

It's hard to explain why Theraphosidae feeding on birds and small mammals make such a strong impression.
Is it because of our naive tendency to make up arbitrary and illusory "rules" we expect nature to follow, and a spider eating a bird feels like the kind of thing that goes against them?
Or is it anthropocentrism, this bias which makes humans care more about animals the more they look like us, which makes us feel more for a bird or a mouse expiring under a tarantula's fangs than if it was "just" a roach or a locust?  Or simply because this ability emphasises the arachnid's gargantuan size?
Whatever the reason(s), "bird eaters" caused, and still cause, quite the sensation and incredulity among people not used to living alongside them (let's not forget that to millions of people living in the tropics, they're quite an ordinary, and not particularly alarming, sight).

Embedded by Merian's picture, the "bird-eating spider" stereotype still lives on, 300 years later.
Any wildlife documentary or book showcasing tarantulas mentions, besides the fact that some grow as large as a dinner plate, that they are big enough to prey on small mammals and birds.
In 2020, a video filmed in Brazil, showing an Avicularia avicularia, the same species as the one represented by Merian, feeding on a wren, went viral on social media, evidencing that the image of the "bird eater" never lost any of its momentum.

In 2020, a video filmed in Brazil (but often incorrectly captioned "Australian tarantula eats bird") showing an Avicularia avicularia feeding on a wren (Trogodytes aedon) went viral on social media, and was featured in many tabloid articles (image source: reddit)

Nowadays, though, we know that the stories of birds devoured by tarantulas are not fables. Large Theraphosidae are physically more than capable of overpowering and devouring small birds and mammals, and besides Avicularia, there are confirmed reports of large "bird eaters" like Theraphosa, Lasiodora and Acanthoscurria species actually feeding on birds.

The enormous fangs of a Theraphosa are more than large enough, and its chelicerae strong enough, to make short work of a mouse or a small bird

However, despite their name and reputation, these large spiders don't specifically target birds. They are opportunistic, generalist predators that will grab a bird, a lizard or a small mammal if the occasion presents itself, just like they would more commonly do with a large insect.
While Avicularia, as arboreal, mainly nocturnal hunters, will occasionally snatch a sleeping bird from its nest or perch, giant "bird eaters" (Theraphosa, Lasiodora, Acanthoscurria...) are terrestrial, and rarely come across feathered prey. In most recorded cases, large terrestrial tarantulas observed eating birds had in fact found them entangled in mist nets used for ornithological surveys.
Of course, a big terrestrial "bird eater" may, from time to time, naturally come across and prey on a chick fallen from the nest or a wounded adult unable to fly away, but this makes such occasions extremely rare. So rare, in fact, that published records of such cases of predation don't fail to mention how uncommon they are.

A fascinating observation of a big female Acanthoscurria simoensi feeding on a small bird, photographed by the American ornithologist and INaturalist user Philip Stouffer in Brazil in 2013. The bird was caught in a mist net for an ornithological study, and was therefore unable to escape when found by the spider (source, credit: Philip Stouffer)

How uncommon, though, is difficult to estimate. Tarantulas are sit-and-wait predators that eat rarely and irregularly, and mainly do so at night. After catching a large meal, they may not feed again for weeks or even months, so witnessing the scene takes a great deal of being at the right place at the right time. In addition to that, hunting tarantulas will often disappear in a burrow as soon as a light shines on them. This makes studying the diet of wild tarantulas extremely challenging, and quantitative studies (on how much of each type of prey they eat) are few and far between. 
Qualitative data (about what tarantulas have been observed eating) are somewhat easier to come across; although they don't give us a precise estimate of the proportional composition of their diet, they still bring a wealth of information. These tell us that tarantulas will, in fact, prey on pretty much anything they can overpower. We know, for instance, that the "goliath bird eater" (Theraphosa blondi), considered the largest spider by body mass, is a supreme opportunist whose menu will include insects, lizards, frogs, snakes, snails and earthworms. 

Observing hunting tarantulas in the wild and recording what they catch is difficult; they tend to retreat back in their burrows if a bright light shines on them, and when they don't, the light may artificially attract insects for them to catch, creating a bias in the sampling

We also know that the bulk of the diet of all tarantulas, including of the largest species, is represented by invertebrates. Captures of birds by tarantulas are extremely rare, but so is the consumption of small mammals, even though large terrestrial species encounter the latter more commonly than birds.

Their ability to hunt and eat rodents was greatly exaggerated by a trend among some keepers in the early days of the tarantula hobby (between the 1970s and the early 1990s) who, having heard they occasionally captured such prey in the wild, used to routinely feed live mice to their animals. Fortunately, this habit has now mostly been put to sleep by the current community. Not only is it unnecessarily cruel (although crickets and roaches probably suffer too when fed to a spider, let's face it), but it is extremely risky; while the tarantula most often makes short work of the mouse, it's still a large and dangerous prey, and the fight can go either way. In addition to that, a mouse is a huge meal, very high in fat and calcium, and anyone who had to deal with the smelly, messy aftermath will attest that, not only is it risky and unhealthy, but it is also unhygienic. 
In nature, just like it should be in captivity, a small mammal is a very rare meal for a large tarantula, which will sustain it for months before it feels the need to feed again.

In fact, while they are the most famous for it, large Theraphosidae are far from the only spiders that feed on vertebrates, nor are they the best at it. 

Surprisingly, the title for the best eight-legged mouse catchers is held by a group of much, much smaller spiders. Widows (genus Latrodectus) are by far the most proficient rodent hunters, accounting for 60 to 80% of observed predation events on small mammals by spiders. They have also been seen capturing lizards, frogs, and even snakes, alongside arthropod prey such as large insects, spiders, scorpions, millipedes and centipedes.

Widows (genus Latrodectus) are, by far, the best mouse catchers among spiders

Their ability to tackle such large prey is quite astounding, as the mass of a large female Latrodectus is only about half a gram. Latrodectus geometricus, one of the smaller species of the genus, holds the record for the largest prey item ever recorded in proportion to its body size: a young rat whose body mass was estimated to be around 20g, i.e. about 50 times heavier than its predator!

While this beetle is larger and much heavier than this brown widow (Latrodectus geometricus), it is tiny compared to the largest animals these amazing hunters can catch

 

Widows can take down prey many times their own size, and vertebrates in particular, thanks to a combination of unique adaptations. Their famously medically significant venom is a complex cocktail of neurotoxins, some of which target nerve cells of arthropods, while others, such as α-latrotoxin, are highly active on the nervous system of vertebrates, particularly rodents. While their venom is, of course, crucial for neutralising the prey once it has been ensnared and immobilised, the true secret of their hunting skills is in their webs. 

Although a widow's web looks, at first glance, like a messy jumble of threads, it is in fact an ingenious and formidably effective trap, constructed with thick, remarkably strong silk threads. Their tensile strength, one of the highest among spiders, is high enough to stop an adult mouse without breaking, and will not fail to impress a curious observer brave enough to test it by gently pulling on a thread. 

The web itself is made of three distinct parts, looking roughly like a scaffold. 
In the middle, lies a sheet-like structure made of thin (but strong) strands of silk, criss-crossing densely; this is where the spider circulates around its web, hanging under the sheet, in and out of its retreat. 
Above the sheet, attached to the surroundings, a messy, tangled network of extremely thick and strong lines of silk forms an aerial trap; a prey that might fly or run into these trip lines will tumble onto the sheet below, where the spider lies in ambush... 
Yet, the most fearsome part of the trap lies in the lower parts of the web, between the sheet and the ground (or the flat surface over which the web is built). Widows and several other members of the comb-footed spider family (Theridiidae) build a type of trap known as gum-foot webs. 
While the rest of the web is not sticky, it is connected to the ground by strong, extremely elastic, very tense lines, whose bottom part is covered with large, highly sticky drops of glue. These are connected to the ground by a weak anchor point that breaks easily. Any prey walking underneath the web is at risk of accidentally touching one or several of these sticky gum-foot lines, breaking the anchor point as soon as it tries to pull away. The broken lines will suddenly lose their tension and contract, pulling the prey up and towards the waiting spider. 
Widows are not the only members of their family to build gumfoot webs, but theirs stand out because of their extraordinary strength: they are strong and sticky enough to trap lizards, small snakes and even mice!

Under its messy looks, the widow's web hides a complex, multi-layered, terrifically effective trap

Finally, when the prey is suddenly pulled towards its lair, the widow meets it with one last secret weapon: as a member of the "comb-footed spider" family, the tarsi (last segments) of its back legs have a comb of stiff, thick, serrated setae (hair-like structures). They use this "comb" to grab onto silk from their spinnerets, and throw it like a lasso, wrapping up their prey from a distance, safe from a counter-attack. Some comb-footed spiders, including widows and false widows (Latrodectus and Steatoda) produce a strong, highly elastic wrapping silk laced with large, extremely sticky drops of glue, which will incapacitate even a large, strong prey item, making it safe for the spider to finally approach it and bite it, and immobilising it long enough for the venom to take effect, which can take several hours in the case of large vertebrate prey such as mice.

Theridiidae (in that case a Steatoda nobilis) are equipped with a comb of stiff, thick setae on the inner side of the tarsi of their back legs, which they use to grab and throw silk from a distance like a lasso

Widows, even though they're not very large spiders, are the largest members of the family Theridiidae, which are mostly small to very small spiders. However, thanks to the combination of adaptations of their silk and back legs, species of this family, despite their generally small size, can tackle very large and/or dangerous prey items, such as ants, other spiders, wasps or scorpions. Being particularly big by this family's standards, widows have taken this trend to the extreme, and adapted to routinely include vertebrate prey to their diet.

Closely related to widows, false widows (genus Steatoda) have also been observed preying on small vertebrates, albeit much less frequently than Latrodectus. There are anecdotal reports of Steatoda nobilis, one of the larger species of the genus (up to 14 mm in body length, which is similar to most widow species), feeding on vertebrates, including a shrew, a bat and a small lizard. The smaller species Steatoda triangulosa has been observed preying on a small gecko.
Like Latrodectus, some Steatoda species have been found to possess some toxins in their venom that are active on vertebrates, albeit less so than true widows. Their webs and hunting behaviour are very similar to those of their cousins, making them fairly effective hunters of small vertebrates, although observed instances of them doing so are much more scarce than predation records of true widows. 
While its venom is not known to be specifically adapted to the capture of vertebrates, the much smaller Parasteatoda tepidariorum, has also been seen capturing mice on two instances, and the hunt was successful in at least one of the two. Its web, similar in structure and strength to the webs of widows, can hold large prey items, but its unspecialised venom takes large doses and a very long time (killing the mouse took 10 hours in the observed case) to paralyse a vertebrate. 

While widows are experts in the matter, occasional captures of vertebrates by spiders is much more common than one would think, and doesn't necessarily require the huge size of "bird-eating spiders", or the potent venom of a widow. 
Consumption of birds, lizards, small mammals, fish and frogs has been recorded across many spider families. 

Huntsman spiders (Sparassidae) include many very large species, some rivalling in size (but not in mass) with large tarantulas. Very much like tarantulas, they're opportunistic predators which feed on whatever they can overpower, and their large size gives them access to a wide and varied menu, which includes large Arthropods and small vertebrates such as frogs and small lizards. The larger species of the family have even been seen feeding on small mammals up to the size of a mouse. 

Like tarantulas, huntsman spiders (Sparassidae) can get very large, accessing a wide range of prey thanks to their size. This Palystes superciliosus is feeding on a large roach, but this species, like all large Sparassids, can occasionally add small vertebrates such as lizards and frogs to its menu

With their large, strong orb webs, large orb weavers in the family Araneidae, particularly in the genera Nephila, Trichonephila, Nephilingis, Argiope, Caerostris and Araneus, mainly target large flying insects, but, as they're strong enough to capture a large cicada or a giant katydid, these webs can also trap very small birds, bats or lizards. The record for the heaviest vertebrate found in a spider's trap is held by a Nephila or Trichonephila species (golden orb weaver) : a laughing dove (Streptopelia senegalensis) weighing 80 grams. 
It would, however, be incorrect to say it was caught by the spider, as the dove was found dead in an empty web. In fact, these big orb weavers generally avoid taking on such large prey and tend to try to cut them free from the web before they damage it too much. Smaller birds, such as hummingbirds and very small songbirds, on the other hand, are occasionally consumed by the spider if they get trapped in the web. 

The immense webs of golden orb weavers (genera Nephila and Trichonephila) are strong enough to catch small birds, which are occasionally, albeit rarely, consumed by the spiders

While the predation of warm-blooded vertebrates gets most of the spotlight, the majority of vertebrates falling prey to spiders are frogs and small reptiles. There is not much of a difference in size or strength between a large insect and a small frog or lizard, and any spider capable of overpowering a cicada or a locust can potentially catch and kill one of those.  
In fact, frogs are the only vertebrates known to make up a significant part of the diet of some spider species. Specimens of two Australian Mygalomorph species, Idiosoma rhaphiduca and Hadronyche formidabilis, have been found with numerous frog remains in their retreats, indicating the amphibians make up a substantial proportion of the prey taken by these spiders.  

Frogs, tadpoles and small fish are also regulars on the menu of many "true" spiders. Members of four different families (Dolomedidae, Trechaleidae, some Pisauridae and some Ctenidae) have evolved as large, semi-aquatic fishing spiders, which hunt from the surface and dive to catch aquatic Arthropods, but also small fish such as minnows, small amphibians, and their larvae. 

Dolomedes plantarius (family Dolomedidae)* is one of Europe's largest and strongest spiders; European Dolomedes are the only spiders of the continent that frequently hunt vertebrates

In humid tropical rainforests, where small frogs abound even well away from bodies of water, non-aquatic Ctenidae and Trechaleidae (which are generally large to very large spiders) will also frequently prey on these tiny vertebrates, whose size is, after all, no larger than the insects they also catch. 

In fact, that is probably what the people who doubted Maria Merian's accounts may have failed to realise: tropical rainforests simply have more of everything. More spiders, big and small, more insects, more vertebrates, some big, but many tiny. 
The Suriname jungles are home to hummingbirds lighter than cicadas, cricket-sized frogs, and mammals the size of a large katydid, all fair game for the spiders adapted to hunting large Arthropods. 

Feeding on vertebrates is not unique to large tarantulas, nor are they the best at it. Thus, calling them "bird eaters" is a bit of a misnomer, albeit not entirely undeserved. 
However, it perfectly encapsulates their formidable bulk, giving them an almost legendary aura. 
What Merian created in the European collective imagination with her drawing was a fantastical monster; an eight-legged King Kong, lurking in a jungle where everything is possible. Yet, what makes bird eaters so special is that they are entirely, undeniably real.

* Dolomedes plantarius is a rare and threatened species, protected in several countries. This specimen was in a captive setting because it was legally captured as part of an authorised scientific sampling campaign. This species should not be maintained in captivity except by facilities carrying out research and captive breeding for conservation purposes.

References are integrated into the text of the article; the words in blue are clickable and will redirect you to the sources of the information. 

Except when the source is explicitly mentioned, the photos illustrating this article are mine and are not free to use without permission 

All of the contents of this blog are proudly and painstakingly produced without the use of AI.