Thambetochen: the flightless, tooth-billed moa-nalo

The so-called bill teeth were blunt bony projections, not true teeth; other fossils preserve evidence of wing reduction and plant feeding.

A flightless Thambetochen moa-nalo browsing on a Hawaiian island
A flightless Thambetochen moa-nalo browsing on a Hawaiian island. Preserved anatomy is distinguished from the reconstructed setting and appearance.

Thambetochen was a genus of large, flightless ducks from the Hawaiian Islands, part of the distinctive group called moa-nalo. Its jaws carried rows of blunt bony projections that resembled teeth but were covered by a keratinous bill. A near-complete skeleton from Molokai and additional bones from Maui and Oahu document major changes to the wings, legs and skull. Coprolites attributed to one species preserve abundant fern spores and other plant traces. Those deposits support herbivory, though they do not prove a fern-only diet. The extinct bird catalogue places the moa-nalo among island birds whose anatomy evolved independently from other flightless giants.

Quick facts

GenusThambetochen Olson & Wetmore, 1976
SpeciesT. chauliodous and T. xanion
GroupAnatidae; Hawaiian moa-nalo
IslandsMolokai, Maui and Oahu
AgeLate Pleistocene and Holocene deposits
MaterialSkulls, jaws, axial skeleton and limb bones
FlightFlightless, from reduced pectoral and wing anatomy
DietHerbivory supported by attributed coprolites
Evidence guide

What can the fossils tell us?

Jaw bones preserve repeated blunt projections along their margins

The fossil does not preserve the keratin covering, so the exact living edge profile is unknown.

Discovery and species

Storrs Olson and Alexander Wetmore named Thambetochen in 1976 from a remarkably informative skeleton found in the Moʻomomi dunes of Molokai. The type species, T. chauliodous, was described from a specimen with much of the skull and skeleton represented. Joan Aidem found the skull and neck vertebrae; Alan Ziegler collected additional elements. The associated material allows the bird to be reconstructed more confidently than taxa represented by scattered cave bones. Small fragments of eggshell were found nearby. The original authors considered whether they might have come from an egg inside a female, but the pieces were not attached to the skeleton. They do not establish the individual's sex, the egg's dimensions or the size of a clutch. A dated land-snail shell from the deposit was used to estimate the age of the locality; it does not directly date the bird bone. The associated land-snail shells were estimated at about 25,000 years old, but an age for nearby material is not necessarily the age of the bird's death. The distinction matters because locality age, the age of one fossil and the final survival date of a whole species are different questions.

A second species, T. xanion, was described from Oahu in 1991 from a complete rostrum found at Barbers Point. It was smaller and more lightly built than T. chauliodous, with differences in the bill and limb proportions. Hundreds of moa-nalo bones from Maui caves have not been cleanly divided into additional species; variation could reflect age, sex or geography. The island record should not be treated as if every population lived on every island at the same time.

The genus occurred on Oahu, Molokai and Maui. Quaternary changes in sea level altered the size and connections of islands, so a map of fossil localities records preserved finds rather than one simultaneous distribution.

A bill with tooth-like edges

The most striking feature is the robust bill with rounded bony projections along the upper and lower jaw margins. These are sometimes called teeth, but they are not teeth with enamel and dentine. They are processes of the jaw bones that would have been covered by the living bird's keratinous rhamphotheca. The soft bill sheath is not preserved, so the exact shape of its edge cannot be observed directly.

Measurements of the type skeleton give a rostrum around 55 millimetres long and a mandible around 98.6 millimetres. Those values describe the specimen rather than a universal species size. The bill's stout construction and blunt projections are compatible with grasping or processing plant material, but they do not identify a single food without direct dietary traces.

Its body was large and goose-like, but reliable genus-wide mass estimates are lacking. A comparison with New Zealand's flightless duck Cnemiornis gives a sense of scale, not a precise weighing. Illustrations should avoid assigning a specific body mass as measured fact.

Why it could not fly

The sternum of the type skeleton has almost no keel and remains incompletely fused along much of its rear midline. The coracoid lacks a well-developed acrocoracoid process, and the wing bones are strongly reduced. In the type, the humerus measures about 56.5 millimetres, the ulna 37.3 millimetres, the radius 35.3 millimetres and the carpometacarpus only 17.9 millimetres. The hindlimbs, by contrast, are robust and well suited to supporting a heavy terrestrial bird.

A small furcula-like element has been reported from Maui material, so complete absence of the wishbone should not be generalised to every specimen. The combined reduction of the sternum, shoulder and wing nonetheless supports flightlessness. Strong legs indicate terrestrial movement, but bone proportions alone do not yield a measured running speed or exact posture.

The independent evolution of flightlessness is common on islands where dense forests and a reduced community of mammalian predators can make flying less essential. That ecological explanation is plausible for the moa-nalo, but no individual fossil reveals why a particular ancestor lost flight.

Coprolites and a plant-rich diet

At Puu Naio Cave on Maui, researchers studied 21 coprolites attributed to T. chauliodous. The assignment used their size, the abundance of moa-nalo in the local fauna and the absence of an equally likely large herbivore. It is a reasoned attribution rather than a dropping found attached to a skeleton.

Four samples with more than 200 counted pollen grains and spores contained over 95 percent monolete fern spores. Plant fibres and chemical evidence also fit herbivory. One coprolite yielded a radiocarbon age of 4,040 ± 100 years before present, calibrated in the original study to roughly 4,835–4,240 calendar years ago. These observations make plant feeding secure and fern consumption well supported.

The high fern-spore proportion prompted the idea that moa-nalo specialised on ferns. Cave vegetation, transport and differential preservation can alter what appears in a coprolite, however. The evidence supports a plant-rich diet that included ferns; it does not prove that ferns were the only food.

An ancient duck lineage in Hawaii

Moa-nalo skeletons are so specialised that early anatomical classifications disagreed about their ancestry. Mitochondrial DNA extracted from fossil bones showed that the Hawaiian group belonged within the duck family and was close to dabbling ducks. It was an ancient island lineage rather than a recent offshoot of one named living species.

The likely scenario is that a flying duck ancestor reached the Hawaiian Islands and descendants independently became large, terrestrial herbivores. DNA supports the broad relationship, while the exact branching point within living ducks depends on the available sequence data and comparison taxa. A modern Hawaiian duck or mallard should not be labelled the direct ancestor of Thambetochen.

Other island waterfowl, such as the New Zealand Cnemiornis, arrived at large flightless forms by separate evolutionary histories. Similar body plans can reflect similar ecological opportunities rather than close kinship.

Frequently asked questions

Were Thambetochen's bill projections real teeth?

No. They were bony projections of the jaw covered by the keratinous bill, without enamel or dentine.

Could Thambetochen fly?

The reduced sternum, shoulder and wing bones show that it was flightless.

What did it eat?

Attributed coprolites contain plant fibres and many fern spores, supporting herbivory. The evidence does not establish a fern-only diet.

Was it related to New Zealand moa?

No. Ancient DNA places moa-nalo among ducks, near dabbling ducks. The name moa-nalo describes a Hawaiian group and does not imply close relation to moa.