Around 11 million years ago, during the late Miocene, much of what is now northern Honshu in Japan was submerged under fairly deep ocean waters. This offshore environment was inhabited by a variety of ancient sea-going tetrapods such as turtles, desmostylians, seal-like allodesmines, archaic baleen whales, and early oceanic dolphins… and also one very unexpected bird.

Meet the flightless marine swan.

Annakacygna hajimei, also known as the Annaka short-winged swan, was the same size as a modern black swan at about 1.2m long (~4′), but had a combination of features unlike any of of its living close relatives. Its head was proportionally large, and it had a long spoon-shaped bill like a shoveler duck, lined with comb-like structures for filter-feeding on plankton. It also had widened hips that would have helped keep it stable floating in rough waters, its tail was highly mobile and muscular, and its feet resembled those of diving birds like loons.

With thickened heavy bones and shortened forearms it was clearly completely unable to fly, but its reduced wings appear to have been highly specialized rather than just vestigial. Its shoulders were extra flexible while its wrists had a more limited range of motion, allowing it to fold its wings into a distinctive half-raised position similar to modern mute swans.

It probably used its wings and tail to perform elaborate “busking” visual displays, and also to carry and protect its young on its back while out at sea – basically making itself into a living swan boat.

Retro vs Modern #20: Deinocheirus mirificus

Discovered in Mongolia in the mid-1960s, and named in 1970, Deinocheirus mirificus was a famous paleontological mystery for over 40 years.


For a long time all that was known of this dinosaur was a few fragments and an enormous pair of arms – some of the largest of any known theropod at 2.4m long (7’10”) – inspiring its name meaning “wonderful terrible hands”.

Initially it was classified as a new type of carnosaur (which was something of a wastebasket group at the time), but similarities with the “ostrich-mimic” ornithomimosaurs were soon noted in the early 1970s. And despite some paleontologists trying to link Deinocheirus to the similarly big-armed therizinosaurs over the decades, the ornithomimosaur interpretation seemed to have won out by the early 2000s.

Depictions of Deinocheirus during this time period were highly speculative and reflected the uncertainty over its evolutionary relationships, varying from giant carnosaurs to therizinosaur-like forms to “Gallimimus but bigger” – or sometimes simply showing a hilarious pair of monster-arms reaching in from out-of-frame. Many popular dinosaur books just gave up entirely and only illustrated the known fossil material unreconstructed, and an iconic photograph of Mongolian paleontologist Altangerel Perle standing between the arms was commonly used to emphasize the sheer scale of the bones.


In the early 2000s attempts to find more fossil material at the original discovery site had only turned up a few additional fragments, including some belly ribs with evidence of having been bitten by a Tarbosaurus – suggesting that the specimen represented the scattered dismembered bits left behind by a feeding carnivore, and that the rest of the carcass might not even have fossilized.

But then between 2006 and 2009 a team of international paleontologists working in Mongolia found a couple of unusual partial skeletons at sites that had been looted by fossil poachers. While parts like the skulls and feet had been taken, the two specimens were still fairly complete and one still had enough arm material left to clearly identify it as Deinocheirus.

When the discovery was announced at the 2013 Society of Vertebrate Paleontology conference it was massive surprise to most of the paleontological community, confirming that Deinocheirus was indeed an ornithomimosaur, and that it was an incredibly weird one. Heavily-built, it was a much chunkier animal than its other relatives, and most surprising of all it had a humped “sailback” formed by long neural spines on its back vertebrae.

Then things got even better.

And stranger.

A “weird skull” had been spotted in the private fossil trade in Europe in 2011, along with some hand and foot material that perfectly matched the missing pieces of one of the new Deinocheirus specimens. The fossils were acquired and donated to a Belgian museum, and then finally were repatriated to Mongolia in 2014, filling in the rest of Deinocheirus’ appearance with a suitably surprising head to go with the rest of its body.

We now know Deinocheirus lived about 70 million years ago during the Late Cretaceous, in what is now the Gobi Desert but at the time was a river-delta-like environment with numerous river channels, shallow lakes, and mudflats.

It grew up to about 11-12m long (~36-39′) and had a long narrow skull with a wide beak and a deep lower jaw – resembling a hadrosaur more than an ornithomimosaur – and it had a rather small brain for a theropod of its size, proportionally closer to that of a sauropod. Its fairly weak jaw muscles suggest it mainly fed on soft vegetation, possibly foraging for aquatic plants in bodies of water like an enormous duck. Gastroliths in its gut helped to grind up its food, and the remains of fish in its stomach suggest that it was also somewhat omnivorous.

Its characteristic huge arms were actually one of the least strange things about it, and were actually proportionally smaller compared to its body size than other ornithomimosaurs. They were heavily muscled, though, with large curved claws, and may have been used to dig up food from mud and soft soil or to pull clumps of vegetation closer.

Its skeleton was highly pneumatized, full of lightening air sacs, but it was still a very big and bulky animal with relatively short legs that suggest it was rather slow-moving. Its feet resembled those of both hadrosaurs and tyrannosaurs, with blunt claws and adaptations for heavy weight-bearing in a bipedal stance.

The large sailback may have been a display structure, and the tip of its tail resembled a pygostyle and so may have sported a fan of feathers. The rest of its body was probably feathered similar to what’s known from other ornithomimosaurs, although potentially more sparsely due to its huge size.


Brontornis burmeisteri was one of the largest flightless birds known to have ever existed, standing around 2.8m tall (9’2″) and estimated to have weighed 400kg (~880lbs).

Known from the early and mid-Miocene of Argentina, between about 17 and 11 million years ago, it’s traditionally considered to be one of the carnivorous terror birds that dominated predatory roles in South American ecosystems during the long Cenozoic isolation of the continent.

But Brontornis might not actually have been a terror bird at all – it may have instead been a giant cousin of ducks and geese.

The known fossil material is fragmentary enough that it’s still hard to tell for certain, but there’s some evidence that links it to the gastornithiformes, a group of huge herbivorous birds related to modern waterfowl.

If it was a gastornithiform, that would mean it represents a previously completely unknown lineage of South American giant flightless galloanserans. And, along with the gastornithids and the mihirungs, it would represent a third time that group of birds convergently evolved this sort of body plan and ecological role on entirely different continents during the Cenozoic.

Weird Heads Month #29: Giant Saw-Toothed Birds

The pelagornithids, or “pseudotooth birds”, were a group of large seabirds that were found around the world for almost the entire Cenozoic, existing for at least 60 million years and only going completely extinct just 2.5 million years ago.

Their evolutionary relationships are uncertain and in the past they’ve been considered as relatives of pelicaniformes, albatrosses and petrels, or storks, but more recently they’ve been proposed to have been closer related to ducks and geese instead.

Whatever they were, they were some of the largest birds to ever fly, and many of the “smaller” species still had wingspans comparable to the largest modern flying birds.

But their most notable feature was their beaks. Although at first glance they look like they were lined with pointy teeth, these structures were actually outgrowths of their jaw bones covered with keratinous beak tissue. While these bony spikes would have been useful for holding onto slippery aquatic animals like fish and squid, they were actually hollow and relatively fragile so pelagornithids must have mainly caught smaller prey that couldn’t thrash around hard enough to break anything.

The serrations also only developed towards full maturity, and the “toothless” juveniles may have had a completely different ecology to adults.

Pelagornis chilensis here was one of the larger species of pelagornithid, with a wingspan of 5-6m (16’4″-19’8″), known from the western and northern coasts of South America during the late Miocene about 11-5 million years ago.

Like other pelagornithids it was highly adapted for albatross-like dynamic soaring, with long narrow wings that allowed it to travel huge distances while expending very little energy – but with its proportionally short legs it would have been clumsy on the ground and probably spent the vast majority of its life on the wing, only returning to land to breed.


Between about 9 and 7 million years ago, the modern regions of Tuscany, Corsica, and Sardinia were once part of a single island in the ancient Mediterranean Sea.

And since evolution often goes in weird directions on isolated islands, it’s no surprise that some unusual species developed there.

One of which was a very odd duck.

A map of the Mediterranean region during the Late Miocene, showing the location of the Tusco-Sardinian island.
From fig 4 in Williams, M. F. (2008). Cranio-dental evidence of a hominin-like hyper-masticatory apparatus in Oreopithecus bambolii. Was the swamp ape a human ancestor?. Bioscience Hypotheses, 1(3), 127-137.

Bambolinetta lignitifila lived during the Late Miocene, about 7.5 million years ago. Known from a single partial skeleton discovered in the mid-1800s, it was initially thought to be a fairly normal dabbling duck and wasn’t properly re-examined until 2014, when its strange features were finally recognized.

It was a medium-sized duck, probably around 50cm long (1’8″), but it had much chunkier wing bones than its relatives, with noticeably shortened forearms – looking much more like the wings of an auk or penguin, and suggesting that it was a similar sort of wing propelled diver. This is incredibly weird for a duck, since every other known diving species uses feet for propulsion instead, and so Bambolinetta may be the only known waterfowl to ever develop this type of underwater locomotion.

It’s not clear whether it was still capable of flying or not. There were few predators in its habitat, so it may well have become completely flightless – and that could also be the reason it later went extinct. Sea levels in the region began to drop around 7 million years ago, reconnecting the Tusco-Sardinian island to the European mainland, and Bambolinetta‘s high level of ecological specialization and its potential island tameness would have given it little defence against an influx of new unfamiliar predators.

Island Weirdness #41 — Talpanas lippa

The big herbivorous moa-nalo weren’t the only unusual waterfowl on the ancient Hawaiian islands — and the island of Kauaʻi had a very odd duck indeed.

The Kauaʻi mole duck (Talpanas lippa) was fairly small, about 50cm long (1’8″), and although it lived alongside one of the moa-nalo species they don’t seem to have been closely related at all. It instead appears to have come from a different duck lineage entirely, with it’s closest living relatives potentially being the stiff-tailed ducks.

It had short chunky legs and an especially weird skull, with eyes so tiny and underdeveloped that it must have been near-blind and flightless. But the areas of its brain associated with the sense of touch were proportionally huge, and although the exact shape of its beak currently isn’t known it seems to have been very wide and flat.

It was probably a nocturnal bird that used an incredibly sensitive bill to grub around in the undergrowth for invertebrates, sort of an equivalent of the modern kiwi but with a face more like a platypus.

The only known Talpanas remains come from mid-Holocene deposits about 6000 years old, but since Kauaʻi is the geologically oldest of the main Hawaiian islands it may have existed there for several million years prior.

Like the moa-nalo it was likely driven to extinction due to human influences on its environment once Polynesian settlers reached the island, sometime between 300 CE and 1200 CE.

Island Weirdness #40 — Moa-Nalo

The islands of Hawaii are part of a larger archipelago formed via hotspot volcanism in the Pacific, and are quite geologically young — the oldest of the main islands was formed just slightly over 5 million years ago, and the youngest less than 0.5 million years ago.

Located almost 3700km (2300 miles) from the nearest continental shore, they’re the most isolated islands on Earth. Their native species are all descended from the rare colonization events that reached such a remote location, either via ocean rafting or island hopping from much older now-submerged islands northwest in the chain.

Like many other Pacific islands no land mammals ever reached Hawaii prior to human arrival, and so it was birds that ended up filling many of the vacant ecological niches.

The dominant herbivores of most the islands were the moa-nalo — no relation to the moa of New Zealand — a group of large flightless birds that resembled giant geese but were actually descended from ducks, with their closest living relative thought to be the modern Pacific black duck. They had reduced wings, chunky legs, and big heavy beaks, some featuring large serrations and others convergently resembling those of giant tortoises.

The Maui Nui large-billed moa-nalo (Thambetochen chauliodous) was one of the biggest species, about 90cm tall (~3′). It originally lived in the highlands of Maui Nui, a large island that formed over 1 million years ago — and when Maui Nui subsided and flooded about 200,000 years ago, it would have then occupied the resulting modern islands of MauiMolokaʻiLānaʻi, and Kahoʻolawe.

Polynesians reached the Hawaiian islands sometime between 300 CE and 1200 CE (the exact dating seems to be controversial). The moa-nalo would have suffered the same fate as many other flightless island birds, lacking any instinctive fear of the new arrivals and falling prey to the invasive pigs, dogs, and rats they brought with them.


Conflicto antarcticus, a recently-named waterfowl bird from the earliest Palaeocene of Antarctica (~65-64 mya).

Standing around 50cm tall (1′8″), it had a slender body, long legs, a long neck, and a narrow goose-like beak. It also had an unusual pair of bony bumps on its skull which may have supported some sort of small crest superficially similar to the knob on the head of the modern magpie goose.

Temperatures in Antarctica at the time were much warmer than today, and the area where its fossils were found would have been a temperate estuary or river delta. It was probably an omnivorous wading bird, feeding on vegetation, small fish, and invertebrates in shallow freshwater.

Although it somewhat resembled a presbyornithid it was actually part of an even earlier branch of the waterfowl evolutionary tree – so its ancestors must have originated much further back in the Late Cretaceous – and their similar body shapes hint that the common ancestor of all waterfowl may also have been a rather leggy bird. Conflicto’s closest known relative might actually be the similarly-aged Anatalavis (which was previously though to be a primitive magpie-goose) from North America and Europe, suggesting that its lineage was quite widespread and already taking advantage of vacant niches in the immediate wake of the Cretaceous-Paleogene mass extinction.

Island Weirdness #16 – Garganornis ballmanni

While some of the main big herbivores on the Late Miocene Gargano-Scontrone island(s) were the larger species of Hoplitomeryx, they weren’t the only animals filling that ecological niche.

Garganornis was an enormous anatid bird, closely related to modern ducks, geese, and swans. Although only known from fragments of its skeleton it’s estimated to have stood up to 1.5m tall (4′11″), making it the largest known waterfowl to have ever lived.

It probably reached such a size thanks to the lack of large terrestrial predators, and possibly also as protection against the island eagles and owls – literally growing too big for them to be able to eat.

It was flightless, with small wings, and had reduced webbing between its toes, suggesting it spent most of its time walking around on land. It also had bony knobs on its wrists that would have been used to give some extra force to wing-slaps when fighting with each other over territory or mates.

Almost-Living Fossils Month #26 – Angry Land-Flamingo-Ducks

The presbyornithids were an early group of waterfowl birds – relatives of modern ducks, geese, swans, and screamers – that first appeared in the Late Cretaceous, about 71 million years ago. With their long necks, long legs, and duck-like bills adapted for filter-feeding, they seem to have essentially been primitive ducks converging on the body shape and lifestyle of flamingos – and as a result they’re sometimes even nicknamed “flamingo ducks”.

They lived in shallow freshwater environments all around the world, and after surviving through the end-Cretaceous extinction they even became some of the most common waterbirds in the early Cenozoic. Some species have been found in large bonebeds containing fossils from thousands of individuals all in one place, suggesting they were very social and lived in huge flocks.

Around the mid-to-late Eocene (~40-37 mya) they seemed to disappear completely, until some fossils from Australia that were originally thought to be from a species of ancient stone-curlew were reassessed in 2016 and found to actually represent the latest-surviving members of the presbyornithids.

Named Wilaru, this bird lived in South Australia during the Late Oligocene and Early Miocene (~28-20 mya). Two different species have been identified: Wilaru tedfordi and its slightly larger and stockier descendant Wilaru prideauxi. With only partial pieces of their skeletons known it’s difficult to estimate their full life size, but based on similar presbyornithids they were probably both somewhere around 1m tall (3′3″).

As well as outliving the rest of their kind, the two Wilaru species were also rather weird compared to the other known flamingo-ducks, with adaptations that indicate they were spending much more time walking around on land than wading in water. Their feet resembled those of modern screamers (which are also more terrestrial) and may have partially or fully lost their webbing, and since they lived alongside various other species of waterfowl and early flamingos they clearly weren’t competing for the same ecological niches. It’s possible they might have also shifted away from their ancestral filter-feeding diet, perhaps becoming more herbivorous, but without any preserved skulls we can’t tell for certain.

Unlike other presbyornithids they also had large spurs on their wings – and based on the behavior of modern spurred waterfowl this suggests they were much less social. They were probably rather aggressive animals, living solitary or in pairs and fighting each other over mates and territory.

This major departure from the lifestyle of their ancestors may have been what allowed Wilaru to survive for so much longer than all the other presbyornithids. They might potentially have lasted a few more million years into the mid-Miocene, but a cooling and drying climate – especially a sudden temperature drop about 14 million years ago – may ultimately have altered their habitat and food sources too quickly for them cope with.