Colin Edgerley

Lost in time

They were among the biggest birds that ever lived, and for millions of years they browsed the shrublands, forests and alpine herbfields of prehistoric New Zealand. Then, in a matter of centuries, they were wiped out. Only their bones remain to tell the story of this country’s most prodigious bird.

Written by       Photographed by Geoff Mason

Most New Zealanders think of moa as tall, ostrich-like birds that roamed the tus­sock-clad plains and foothills of early New Zealand. The popular image is of large flocks of birds grazing the verdant grasslands, with no preda­tors to worry them.

To be sure, some would blunder into swamps and become trapped, their tissues slowly dissolving away while their massive bones were pre­served for human settlers to dredge up thousands of years later—and marvel. But bogs aside, life on these isolated islands must have been an avian paradise, and moa ruled the roost.

It was their size that was their undoing. Like the dinosaurs before them, moa were just too big and cum­bersome to cope with the modern world, and they finally lost the strug­gle for survival.

Or did they? Tantalising rumours persisted, of sightings in Fiordland, Otago and the Takaka Hills. Hadn’t Japanese scientists visited remote South Island valleys, hoping to at­tract moa with electronic calls?

Sadly, technology cannot rescue moa from extinction: no moa will ever answer a computer-generated love-call in the forests of Fiordland. Skeletons in museum cases are as close as most people will ever get to moa now.

In the early 19th century, how­ever, anything seemed possible. Europe had an insatiable appetite for for knowledge of the outside world. Voy­ages of exploration were unearthing new plants and animals from the tropics to the polar seas, and adventur­ers dreamed of discovering lost worlds where fantastic creatures might still be found.

Still, after 70 years of European contact with New Zealand, Europe­ans had not seen one fragment of a moa. Then, in 1839, a Dr John Rule brought a fragment of bone to Eng­land and persuaded the great English anatomist Richard Owen to examine it. Interrupted while writing a lecture, Owen at first dismissed Rule’s claim that the bone came from a flying bird, telling him that it was more likely to be a piece of ox bone from someone’s kitchen rubbish!

Rule persisted. He told Owen that it came from New Zealand, and in­sisted that it was from a bird. Owen, still largely unmoved, asked the en­thusiastic collector to leave the bone with him so that he could examine it more closely after his lecture.

Later, as he compared it with specimens in the collection of the Royal College of Surgeons, Owen re­alised that his off-the-cuff assessment had been wrong. The bone did not match anything from an ox or horse or any other large mammal. Surface markings suggested that it was in­deed a bird bone, and from a bird larger than an ostrich.

This was a major discovery, and, at considerable risk to his reputation and future standing in the scientific community, Owen announced that: “So far as my skill in interpreting an osseous fragment may be credited, I am willing to risk the reputation for it on the statement that there has ex­isted, if there does not now exist, in New Zealand, a struthious [ostrich-like] bird nearly, if not quite, equal in size to the ostrich, belonging to a heavier, more sluggish species.” He named the bird Dinornis Novae Zealandiae, coining the genus from the Greek “deinos”, meaning surpris­ing, prodigious or terrible—adjec­tives which aptly fitted a bird which was thought to reach five metres in height.

Many were not convinced, and the publication council of the Zoological Society insisted that Owen make it clear that they were not to be associ­ated with such an absurd suggestion: the claim was Owen’s and his alone.

Owen, anxious to shore up his reputation, encouraged European set­tlers of the new colony to look for more evidence, and soon boxes of moa bones were on their way back to London, sent mainly by explorer-mis­sionaries such as William Colenso. From the new specimens, Owen de­scribed more and more species, often from single bones, and speculated on their habits and origins.

While some searched for bones, those with an ear for Maori tales col­lected anecdotes about huge birds. One story from East Cape told of a bird which resembled an enormous do­mestic rooster, had the face of a man, lived in a cave, ate air and was guarded by two immense tuatara. At the same time, reports of live moa, always dimly and briefly seen, kept alive the hope that the birds were not extinct. There was even some opti­mism that it was only a matter of time before the birds would be “seen strid­ing among the emus and ostriches in the Regent’s Park.”

But the reports remained unverified, and the hopes unfulfilled. There is no evidence that any moa were alive to see Cook’s Endeavour, even from a far-off hilltop, although Tasman, had he landed, might have had better luck a hundred years ear­lier. For us in the 20th century, the words of the Maori lament hold true: Ka ngaro i te ngaro a te moa. Lost as the moa is lost.

[Chapter Break]

Moa belong to  a group of big flightless birds known col­lectively as ratites, a group which includes the African ostrich, the South American rhea, the emu and cassowary of Australia and New Guinea, and the kiwi. They were first grouped together because they all have a flat sternum, or breastbone, with no hint of the plate or keel which provides an anchorage for the flight muscles of other birds. The sternum is raft-like—ratis in Latin.

Moa form a distinct section of the ratites, and are unique to New Zea­land. Unlike all other ratites, which have wing remnants, moa have no such structures; they are not only flightless, they are wingless.

For much of the past century and a half, controversy has raged over how many species of these extraordinary birds there were. Dr Walter Oliver and Sir Gilbert Archey recognised more than two dozen species in their publications in the 1930s, ’40s and ’50s. Although this was an improve­ment on Lord Walter Rothschild’s 37 species of 1907, there were still more than could be justified on ecological grounds. (A basic tenet of biology is that species cannot coexist unless they possess adaptations which al­low them to exploit differing ecologi­cal niches. With moa, there simply weren’t enough niches to go around.)

The surfeit of species arose be­cause moa cataloguing was based al­most entirely on the size of the bones, a parameter in which there is consid­erable variation. It has only been in recent years that this approach has been shown to be flawed: not only can bones of identical size belong to different species, but, to make mat­ters worse, bones that differ in size can belong to the same species!

How can this be?

It is a fact of nature that bigger animals have an energy advantage in cold conditions because they cool down more slowly, and their tissues require less food per unit weight than do smaller animals. New Zealand spans many degrees of latitude, and birds in the cooler southern areas were probably always larger than those of the same species living in the North Island at the same time. (This effect, known as Bergmann’s Rule, is observed in animals such as bears, whose distributions ‘extend over thousands of miles of latitude.)

Compounding this size variation is the fact that the New Zealand cli­mate over the past 20,000 years has warmed by an average of three to five degrees. Even in the North Island, it was an advantage for birds to be larger at some times in history and not at others. It is now known that the aver­age size of several species of moa markedly decreased as the climate became milder, from 14,000 to 10,000 years ago.

Much of the confusion in moa clas­sification came about because until 1950 there was no way of knowing exactly, or even very roughly, when a particular bird had died. Collectors who lumped together their finds from different sites unavoidably mixed birds from different times, and scien­tists examining these bones were prone to separate them into different species when in fact they were merely variations over time. To add to the confusion, it seems that in some species one sex was bigger than the other—probably the female, if moa are consistent with other ratites. Scientists who did not realise this fact gave separate names to bones from males and females of the same species.

The key to the current system of classification (in which just 11 species are recognised) is bone morphology coupled with an understanding of moa distri­bution and evolution. By choosing samples from nearby sites (thus controlling geographic variation) and dat­ing the bones (thus control­ling variation over time) a more accurate picture emerges. This is not to say that the last word has been written on moa taxonomy. The three large Dinornis moa are still difficult to identify from leg bones. Shape characteristics that allow us to separate these spe­cies, irrespective of size, have yet to be worked out.

At present, basic differences in body plan are used to separate the major groups. For example, the three species of Dinornis are distinguished from the other eight moa by having two or three more bones in their spi­nal column, a different-shaped skull with a characteristically broad and down-curved beak, and by the rela­tively longer lower leg bone.

The three Dinornis moa form a family of their own, the Dinorni­thidae. The other moa are all included in the family Emeidae which, in turn, is divided into two groups, or subfamilies. Moa of the genera Euryapteryx and Emeus have rounded bills, an elongated wind­pipe (tracheal loop), and lack one of the joints of the outer toe, and to­gether form the subfamily Emeinae. The other three genera, Megalap­teryx, Anomalopteryx, and Pachyornis, have the toe joint which is absent from the previous group, belong to the sub­family Anomalopteryginae.

These divisions make it seem probable that there were originally two lineages of moa, each of which devel­oped along its own evolution­ary pathway. They may even have arisen from two separate colonisations of New Zea­land.

The conventional picture of moa origins is that they arose from ancestors that lived on the southern supercontinent of Gondwana. They and the other ratites are thought to have been rafted away from the common ancestral bird stock on the frag­ments that became Africa, Mada­gascar, Australia, South America and New Zealand after the supercontinent broke up 60-90 million years ago. The various groups became the ostriches, elephantbirds, emus and cassowaries, rheas, kiwi, and moa. There is, as yet, no fossil evidence to support this view.

An alternative scenario has arisen out of the discovery of a new group of fossil birds that are very like ratites, except that they could fly well. Geologically much younger than the breakup of Gondwana, these fossil birds may share a common ancestor with kiwis, and, more distantly, with the other ratites. If this theory turns out to be true, then ratites are more closely related to each other than to all other birds, and moa and kiwi may well have flown to New Zealand, not walked.

But how could flying birds turn into enormous wingless birds like moa anyway? Until re‑cently, it was thought that flightlessness took millions of years to evolve. Then a series of dramatic fos­sil discoveries in Hawaii proved that birds can lose the power of flight very rapidly—in just a few million years. In isolated New Zealand, as in Hawaii, there were no large mammals to inhibit such a proc­ess.

Specialists may argue about the evolution of flightlessness or the im­portance of size and shape differences in bones, but what im­presses people the most today is still the moa’s size. Every genera­tion since Owen has concen­trated on their supposedly in­credible height.

They may not have been the heaviest of birds—that distinc­tion goes to the elephantbirds of Madagascar (also ratites)—but they were, in the prevailing popular and professional views, definitely the tallest. And mu­seum curators went out of their way to accentuate this fact.

Most display skeletons have been erected in anatomically impossible ways in an effort to gain the desired height. Conceived by the early ex­perts as “giraffes on two legs,” moa have had neck vertebrae added by enthusiastic restorers, with scant re­gard for the real number in the living animal. Skeletons teeter precariously on vertical leg bones, with over-ex­tended knees and ankles, and with their necks stretched ever upwards.

Reality is much less exciting. Moa had normal legs, horizontal backs, and carried their heads low. The tall­est was about two metres over its back, but its head was normally about the same height. Only when stretch­ing up for a tasty shoot or cluster of berries did this giant come near to the height of a large ostrich.

Much of the mystique surround­ing moa stems from incorrect recon­structions of their skeletons. Even today, controversy surrounds the “correct” skeletal arrangement. The shape and structure of the joint sur­faces where bones meet should be the guide. Joint facets between adjacent bones must be parallel, and oriented correctly. If this rule is followed for each joint, the “rest” or “normal” pos­ture of the animal is reconstructed.

An animal can, of course, alter its joint angles, within the ranges of travel of the joint surfaces, to walk, put its head down to drink, reach up to a twig, or settle on to a nest. What it cannot do is move the joints to positions where the articulating sur­faces do not meet properly. The re­constructions in this article and the accompanying poster, based on nor­mal limb and neck positions, differ substantially from many seen in books or museum exhibits.

Display skeletons are often com­posites, with bones from different individuals, species, or even genera mounted together. This shortcoming has not helped either the faithful re­construction of the birds or the pub­lic’s perception of their appearance.

Although they all had much the same basic stance, moa varied in their body proportions. Some, like the Dinornis group, were relatively slen­der; others, such as the larger species of Pachyornis, were incredibly stout.

One common misconception about the normal outline of moa is that they were higher than they were long. In fact, they resembled cassowaries (now the only large for­est ratites) in being longer than they were high. Therefore, they were not like ostriches, emus or rheas at all. Such birds live in open habitats, not forest.

One reason for the lack of overall height in moa is that their neck verte­brae are much shorter than those of open country forms, such as ostriches and emus. So, although the neck was held in the same descending-ascend­ing loop as the others, the short bones resulted in a much shorter vertical part of the neck. The head was not only not far above the level of the back, but it was attached to the neck at the back of the skull and not under­neath, as in other land birds. This gave the head and neck a characteris­tic, almost snake-like, outline. The only other land bird whose neck met its skull at the same angle was the strange adzebill, another one of New Zealand’s extinct flightless birds.

Despite their many similarities, the two main groups of moa would have looked different, in one respect, even to a casual glance. Dinornis moa had legs whose proportions were like those of other birds. They would have walked with long strides, rather like cassowaries. Emeids, however, spe­cialised in low-gear locomotion, and the stoutest ones were neither fast nor agile. Some species walked and others waddled, but all had the very shortest of lower leg bones. Indeed, the bones were so short that the belly would have nearly touched the ground in the larger species.

Regardless of how it moves, a bird’s body shape is filled out by its covering of feathers. Many fossil moa feathers are known, and they lack the fine hooks, or barbules, which hold the vanes of most birds’ feathers to­gether. Without the hooks, feathers look shaggy; a moa would have looked “hairy,” like a kiwi.

Remarkably preserved specimens of the upland moa show that it, alone among moa, had legs feathered to the toes, like the Arctic grouse, the ptar­migan. This unusual feature may have helped keep its legs warm in cold environments, such as the subalpine forests and shrubland where this bird lived.

A bird’s wings contribute much to the outline of its body. The only liv­ing birds where the wings do not show are kiwi, but even they have small, vestigial wings hidden amongst the feathers. Moa are the only birds which had no wing bones at all—not even remnants inside the body. The lack of wings meant that the breast region was flattened, and shallow, too, because the great breast muscles that operate the wings in other birds were lacking. At the other end of the animal, the lack of substan­tial tail bones is evidence that moa had no tail.

In keeping with the differences in height and proportions, the 11 spe­cies varied greatly in weight, ranging from 15-20kg in the upland moa, Megalapteryx didinus, and the North Island coastal moa Euryapteryx curtus, to perhaps 270kg in the huge Dinornis giganteus. (In case readers wonder how one weighs an extinct bird, these estimates are derived from mathematical relationships which link the size of a living bird’s big, load-bearing leg bones to its weight. These relationships can be applied, with due allowances, to extinct spe­cies.)

Unlike living birds, which can be studied directly, our knowledge of moa—how they looked, how they were related, and how they lived and died—comes from studying bones and bone sites. Moa bones have been found in swamps and caves and exposed by wind in sand dunes through­out New Zealand. They may have lain in the ground for hundreds or thousands of years. Highly pol­ished gizzard stones are also widely distributed, and have of­ten outlasted the bones of the birds that swallowed them.

In the early days of collecting, only the best bones were kept. Partly decayed bones and those from immature birds were dis­carded. Bones and the “com­plete” skeletons built from them were regarded as display speci­mens, not as a raw material for scientific study. Only the best were suitable for sale or exchange, and, as a result, many important specimens were thrown away.

Bones last a long time if they are buried rapidly in swamps or dunes, and so kept away from the oxygen that micro-organisms need. In caves, the low and constant temperature in­hibits decay, and the alkaline condi­tions help preserve bone. Some swamps are also alkaline, but in acid swamps bones break down rapidly. In especially favourable sites, bones may be preserved as well as if they had been prepared in a museum—a fact that has fooled many scientists. Younger, exposed bones often look older than better protected, but older, specimens in the same area of cave.

Articulated skeletons are found  more often in caves than elsewhere. They vividly capture the last hours of a moa’s life. Arriving dazed, but un­injured, at the bottom of a shaft, the trapped bird walked about in the gloom. In its search for a way out, and for food, it wandered further into the dark tunnels, stumbling over the rock debris and mixing up the bones of earlier victims. After several days in the wet and cold, the bird was too weak to move, and it sat down and died. Until rats arrived with the first Polynesian settlers, there were no scavenging mammals in New Zea­land, so the bones were often left perfectly articulated as the flesh rot­ted away.

Cave deposits often contain exqui­sitely preserved remains of other birds, lizards, bats, frogs and snails, as well as moa. Bones of all spe­cies, along with the remains of insects and snails, may be depos­ited in stream and rockfall sediments which are layered, or stratified. As layers build up over time, each contains a record of the animals that lived, and died, at that time in history.

Caves persist as landscape features much longer than swamps, which fill up with sediment and vegetation and eventually dry out. The bones then crumble to dust. Stratified deposits in caves may be partially washed away by later floods, but enough of them usually remain to become our most valuable records of past envi­ronments. In some cases, a continu­ous record spanning 20,000 years has been found at a single site.

Remains in caves, swamps, and dunes provide indisputable evidence of where moa once lived. They do not reveal where they were absent. Moa remains in swamps do not mean that they were water birds. Neither do bones in limestone caves mean that moa lived in caves, or even necessar­ily only in limestone areas. The distribution of remains reflects only the availability of suitable pres­ervation sites, and our ability to find them.

But the more sites that lack remains of a species in an area, the more likely it is that the bird was not found there. This is par­ticularly true for comparisons be­tween the islands. Broadly speak­ing, moa lived just about every­where in New Zealand, except on the small offshore islands. Nine of the eleven species were found in the South Island, and seven in the North Island; five were common to both. The South Island had four species unique to it; the North Island, two.

Within the islands, it is clear that each species had its own dis­tribution, just as living birds such as tui and kaka have theirs. And, like living birds, these distribu­tions were based on preferred habitats. Different species were generally restricted in their habi­tats, which ranged from coastal scrub to high altitude herbfields. For example, bones of the upland and crested moa are generally found only in higher altitude sites, leading us to believe that these birds lived in the montane forests and foraged into the subalpine shrub- and herbfields in summer. In the South Island, the eastern moa, as its name sug­gests,was not found west of the Alps. It is commonest in deposits from the lowland areas of Canterbury where hot, dry summers follow cold winters.

Until 800-900 years ago, Canter­bury had, for thousands of years, been clothed in a mosaic of diverse vegetation types. Dense shrub thickets were interspersed with areas of tall totara, matai, or kahikatea forest, low forest of broadleaved trees, open shrublands and river flats. It was an ever-changing landscape, where the great rivers carved new courses through es­tablished forests, and new forests colonised the raw gravels of the old riverbeds through a succes­sion of shrub and forest stages. In this complex mosaic of habitats, the commonest moa were the stout-legged and heavy-footed moa. Towering over these was the tallest bird ever to have lived, the giant moa, which often ex­ceeded 250kg in weight.

At most, only four species of moa ever lived as part of the same bird community. One or two of the tall dinornithids and one or two of  the emeids were present in most areas, but which species were actually present varied with the island, with geography, and with time. For example, in the continuous tall, wet forests of the King Country during the Holocene (from 10,000 years ago until the present) two Dinornis spe­cies, the large bush moa and slender moa, lived alongside the little bush moa. But in the coastal dune country of the North Island, where the vegeta­tion was a mosaic of open dunes, shrublands, and coastal forest, these were replaced by the giant moa, coastal moa and Mappin’s moa.

Vegetation patterns are not fixed; they have altered many times, with changing climate. Understanding the major changes in New Zealand’s cli­mate over the past 100,000 years has been one of the keys to our new un­derstanding of moa and their habi­tats. Everyone today is aware of the possible effects of global warming, but the world has gone through many cycles of warming and cooling in the past two million years. Every 100,000 years or so, changes in the Earth’s orbit and the way the angle of the Earth’s axis is aligned to the Sun cause temperatures near the poles to drop, and glaciers and ice caps grow larger. Colder and milder periods al­ternate within each Ice Age, but dur­ing the coldest times, ice caps have reached the latitudes of London and Chicago.

In New Zealand at these times the Southern Alps bore a large ice cap, and glaciers spilled out on to the Canterbury Plains. Geography changed in harmony with the cli­mate: sea level fell and rose as water was alternately locked up in glaciers and released again to flood the ex­posed coastal plains. The most recent ice retreat, after the Otiran glaciation had reached its peak 20,000-18,000 years ago (when average tempera­tures were 4-5°C cooler than they are now) and begun to wane, from 14,000 years ago, resulted in sea level rises of 100-150 metres from their glacial minimum (The present outline of New Zealand was reached only about 6,000 years ago, when sea levels sta­bilised.)

Thus, up to 14,000 years ago, the area of Cook Strait between Taranaki and northern Marlborough and Nel­son was dry land, although The Nar­rows was still a deep bay. Further south, the coastline was beyond the end of Banks Peninsula. Climatic and geographical changes affected the vegetation, and hence the birds and other animals. For example, the Oparara valley in northwest Nelson, now under lowland rain forest, was in the subalpine forest and shrubland zone 20,000 years ago. Moa and other birds there then were species charac­teristic of higher altitudes during later Holocene times.

Moa distributions changed with the climate, but only within the is­lands, not between. The Cook Strait region seems to have remained a bar­rier even when it was dry land. Ani­mals which needed warmer condi­tions retreated north and were never close enough to the land bridge to use it. By the time they had moved south again, unbroken sea separated the is­lands.

Vegetation patterns changed rap­idly across New Zealand as the Otiran glaciation gave way to the warmer Holocene period, beginning 14,000 years ago. Certainly, most of New Zealand was under forest by 10,000 years ago, and by 3000 years ago 85­90 per cent of the South Island and 95 per cent of the North Island were forested. Even the Canterbury plains and the central volcanic plateau of the North Island were covered in the characteristic local forest types.

In short, moa have always lived in a land dominated by scattered shrublands, woodlands, and forests. But all forests are not the same. Some have a more diverse structure than others, with more species of plants and a greater variety of heights, edges and successional stages. These pro­vide a wide range of diets for herbiv­ores. Others, such as the wet forests of the West Coast and the central North Island, are more uniform, and seem to have provided fewer oppor­tunities for moa.

Whatever their habitat, their beak shapes and gizzard contents (as pre­served in swamps) show that adult moa, at least, were largely vegetarian, and that they browsed rather than grazed. Leaves, twigs and fruit seem to have been the normal fare, though it is unlikely that the three or four species that lived together in any one place competed for the same foods. Gizzards of the largest Dinornis moa usually contain short lengths of twig, with some leaves and berries, along with the many large gizzard stones needed to grind up such a coarse meal.

In contrast, moa of the genera Euryapteryx and Emeus had a more conventional diet of soft leaves and berries, and they had a few small gizzard stones. Their beaks were weak, and the jaw muscles were small, judging from the small attach­ment areas on the skull. They could not cut and shear tough twigs and leaves like other moa. The three spe­cies of Pachyornis and the small Anomalopteryxhad very stoutly con­structed bills, with large jaw mus­cles, and they could crop tough leaves. Indeed, the last meal of one bird was native flax—a challenging meal if ever there was one.

A feature of the New Zealand flora is the number of plants that have divaricating, or densely interwoven, branches. This habit usually ceases if and when the plant reaches more than three metres in height—co­incidentally the maximum height at which the largest moa could feed. Some scientists see this tight branching pattern as an adaptation to moa browsing: the vulnerable tissues of leaves, flow­ers, and fruit are enclosed in an armour of tough, often spiky, branches, away from the pruning bills of moa.

Birds do not, of course, have the rows of grinding teeth and the complex stomachs that allow cows to extract nutrients from coarse herbage. Instead, they have a crop for food storage and a muscular gizzard to grind up their food before it is sent for digestion into a long intestine.

Those that rely on leaves and seeds as their main food source routinely swallow pebbles and other small ob­jects to act as millstones in the giz­zard, speeding up the mechanical breakdown of plant matter. Different species of moa picked up different sizes of stones, and different amounts, according to their diet. Up to 5kg of stones up to 5cm in diameter have been found with skeletons of the largest moa.

The intestine probably oper­ated as a large fermenting vat. Here, as in horses or elephants, micro-organisms broke down the plant cellulose into simple mol­ecules that the moa could use. This monogastric (one-stomach) digestion, which needs a large fermentation vessel to make the best use of low quality food, is one reason why moa grew so large. To get the most out of the food, it had to be held in the gut for a long time, and a long gut needs a big body to house it. Large size not only allowed a long, efficient digestive tract, but also pro­vided the lower metabolic rate typi­cal of larger animals. A lower meta­bolic rate would have been more eas­ily supported by a vegetable diet that was high in fibre and relatively low in nutrients.

There were no mammalian preda­tors in New Zealand to make growing large and flightless hazardous, but large size did not allow moa to avoid predation entirely, as it does for adult elephants. Haast’s eagle (Harpagornis moorei, see New Zealand Geo­graphic, Issue 4), a huge eagle weigh­ing up to 13kg, killed at least the medium-sized (up to 150kg) moa, and the chicks would have been vulner­able to the large extinct harrier (Cir­cus eylesi) as well.

Mixtures of plant species in food remains suggest that most moa spe­cies had a varied diet, but it is also possible that they were not strictly vegetarian. If, as seems likely, moa grew as fast as ostriches, and reached adult size in 12-18 months, the young birds would have needed a high-en­ergy, high-protein and mineral-rich diet. Plants do not contain high enough concentrations of protein, so the chicks were probably omnivo­rous, eating small animals as well as fruit and leaves. Frogs, lizards, and tuatara were abundant in pre-human New Zealand, and there was a wide range of large invertebrates, includ­ing beetles, weta, and snails, to feed on. Moa chicks probably ate anything they could catch. Today, the chicks of another plant-eating bird, the takahe, eat insects, and the adults will even eat lizards.

Moa used a range of senses in their search for food. Most birds have a very weak sense of smell, and rely on vision and acute hearing for obtaining infor­mation about their environment. Kiwi, which are nocturnal, and have poor eyesight, are one of the few groups of birds with a well-devel­oped sense of smell. In moa, just as in kiwi, the part of the brain associated with smell was proportionately very large. Whether this indicates that moa had a strong sense of smell is open to debate, and, indeed, it is difficult to suggest a function of such a sense in a vegetarian bird.

Perhaps, in their forested environ­ment, moa used smell to locate one another, or as territorial marks. Pet­rels, seabirds which breed in bur­rows under heavy vegetation, have a heavy musky odour that permeates their burrows. They have a good sense of smell, and may use it to locate their burrows at night when they arrive back from the sea.

Moa had small eyes, particularly in comparison with those of emus or ostriches, to which they have often been likened. The bony orbit is small, and the nasal chamber extends back between the eyes further than in any other bird except kiwi, leaving little space for large eyeballs. It seems likely that visual cues would have been less useful in forested environ­ments than in the open habitats pre­ferred by emus and ostriches.

Forest birds generally use calls to communicate where they cannot see each other, and there is no reason to  doubt that moa were vocal birds. Some moa, such as the eastern, coastal, and stout-legged moa, had enormously elongated wind-pipes (trachea) which curved back in a loop in the body cavity. By contrast, most birds have a short, simple air passage to the air sacs and lungs. Trumpeter swans and cranes, however, use a similar loop in their wind-pipe to amplify their resonant trumpeting calls, which carry over long dis­tances. Perhaps the early days in Can­terbury and Otago may have been enlivened by the booming calls of moa trying to attract mates or defend their territories.

Apart from this indirect informa­tion on moa calls, there is little firm evidence from fossils about moa be­haviour. Living ratites can provide some clues, if the differences in habi­tat and the absence of mammalian predators are kept in mind. The only large forest ratites today are the cassowaries of Australia and New Guinea. These birds live singly, in pairs, or in small family parties—the widely dispersed food resources in a forest environment will not support flocks of large birds.

Moa almost certainly lived the same way, with each bird or pair occupying its territory, and only a pair or two to a valley. Pairs would defend territories by calling and dis­play fighting. Cassowaries follow tracks to fruiting trees, and it has been claimed that traditional moa tracks are still visible in some unploughed hill country in Hawkes Bay. Moa were long-lived, and traditional food sources would have been sought at the best time of the year. This behaviour can be observed in kakapo and takahe, the last remain­ing flightless vegetarian birds in New Zealand.

Little is known about moa breed­ing habits. The only moa nests that have been found have been in shel­tered sites, in the lee of rocks, or in hollows. The incubating adult, prob­ably the male, as in other ratites, would have had a long duty: perhaps two to three months. Shelter from the elements, and from the moa’s main predator, Haast’s eagle, would have been an important factor in the choice of sites. Once selected, nest sites are likely to have been reused through the years, as suitable locations would have been rare in many areas.

In all cases where we can be rea­sonably sure that only one nest has been preserved, the clutch size has been small—just one or two eggs. The small clutch size also tallies with the cassowaries, rather than with os­triches, emus, or rheas, which live in open habitats and have to cope with mammalian predators.

In most living ratites, including kiwi, the male does the bulk of the large and able to take a range of ani­mal and vegetable foods. Although they would have rapidly reached adult size, like kiwi, their bones did not finish developing until the birds were several years old.

Males probably did not become sexually mature until they were five or six years old, but once they had reached breeding age, a moa’s life expectancy was probably 20 years or more. The main danger to moa chicks would have been the native harrier, which became extinct at about the same time as the moa, after Polynesians arrived. This hawk was about four times heavier than the Australasian harrier now common throughout New Zealand. Harriers normally hunt over open country, but the big New Zealand bird was adapted for forest living, and had relatively short wings.

A fully-grown moa would have been too big for the harrier. It was not too big for Haast’s eagle—the largest eagle ever to have lived. With a wing­span of up to three metres, and talons the size of a tiger’s, this bird would have been a formidable predator.

Clear evidence of the eagle’s prow­ess as a moa hunter came to light only this year in the form of terrible claw marks in the hip bones of moa found in Pyramid Valley and other swamps. Some 10 per cent of the bones exam­ined have marks from the front and rear claws. To make such marks in the living bone, the claws would have had to penetrate up to 5cm of flesh. Sometimes the eagle changed its grip, or clutched twice as it strained to bring down the struggling moa. Sev­eral bones were found to have large gouges, where the bone has been completely ripped away by the huge rear claw.

The claw marks are exactly where they would cause most damage. One foot held the broad back of the moa, while the other tore through the flesh around the spine to penetrate the lungs, nerves, and major blood ves­sels along the spine.

Eagles preyed on other birds as well, including flightless geese and takahe, but moa formed their staple diet. Even so, eagles would have just taken the excess from the population. They were not responsible for the extinction of moa. Eagles depended on moa for food, and were in rela­tively low numbers, so did not have much effect on any species in the long term. They were also very rare in many areas, including the whole North Island and western regions of the South Island. Their distribution closely matched that of a group of small to medium-sized moa of the more varied forest and shrubland habitats of the eastern lowlands and open mountain basins.

Extinction came rapidly for all moa, throughout their ranges. In the period from 800 to 500 years ago, most populations of moa simply van­ished. The usual reasons proposed for their extinction have been that the climate changes over the past 1000 years drastically affected the birds’ habitat, or that the group had been “evolving itself towards extinction.”

Having survived the many envi­ronmental changes over the previous 1-2 million years, it is inconceivable that all eleven species should suc­cumb to the minuscule climatic changes of the past 1000 years. And, as we have seen, far from being ill-adapted, grotesque creatures strug­gling to survive, moa were part of a thriving, dynamic ecosystem, were very well adapted to their own habi­tats and had coped with substantial changes before.

Some other influence had to be involved, and that influence was man. All known species survived until Polynesian times. Then, as man­made fires altered the New Zealand environment on a vast scale, moa declined to extinction. Fifty percent of the forest area of the South Island was destroyed in the first 500 years of human settlement. By 400 years ago, moa were rare; by 300, they were all but extinct.

A critical factor was that most of the forest destroyed was on the more vulnerable, drier downlands, plains and river valleys of the east. The very forests which supported most moa were the ones destroyed.

While habitat destruction was the major factor in the decline of moa populations, man the hunter tipped the balance already heavily weighted by man the destroyer. In a land de­void of all land mammals except bats, bird flesh would have been an impor­tant food source, and a moa would have made a sizeable meal. Hundreds of thousands of the big birds found their way into earth ovens, to end their days as the main course in a moa-hunter feast.

Maori traditions indicate that moa were caught in nooses and snares, driven into lakes to be decapitated from canoes, chased to exhaustion by teams of runners, herded into mass killing pits and even, it is claimed, induced to swallow heated pebbles. Dogs may well have been used to aid in tracking and capturing the birds, and there is evidence that some were bred with powerful heads and shoul­ders for the hunt.

When Europeans arrived, the re­mains of camp sites littered the ground at the mouths of streams all over the country. Some southern sites covered many hectares and con­tained thousands of birds. Eggshell is common in many middens, showing that the early inhabitants consumed not only the adults, but the next gen­eration, too.

The bones, skin, and feathers of moa undoubtedly provided an im­portant industrial resource, in addi­tion to the immediate demands for food. The large, thick leg bones were ideal for the manufacture of fish hooks, chisels, and awls. Bones were carved into an array of ornaments, including distinctive reel necklaces. It is probably beyond dispute that the skin and feathers were used for cloth­ing, although few remains are known.

Eggs were obviously highly val­ued, because they have been found in moa-hunter burials along with other grave goods. The usual interpretation is that the eggs were used as water bottles, but this explanation seems unlikely: eggs with only a single small hole in the end would have been painfully slow to fill and empty. Perhaps the hole simplified the cook­ing of the egg as food for the depart­ing spirit. It is possible that eggs had some other symbolic value, now un­known and unknowable, and that they were blown so that they would keep.

A mere heartbeat of geological time, the Polynesian moa-hunter pe­riod was twice as long as the Euro­pean era in New Zealand. Annual hunting trips would have prevented the slow-breeding moa from recolonising remote western valleys, where their populations had always been low. Towards the end, as stocks dwindled, demand for meat, feathers and bone outstripped supply, and the birds were hunted with ever more persistence and cunning.

The final elimination of the moa was as inevitable as it was rapid. Hunted everywhere, and with much of the food for the young already gone—eaten hundreds of years be­fore by plagues of kiore rats brought in by the first settlers—moa simply died out. Millions of years of adapta­tion to a unique environment ended in a few centuries.

So complete was the process, and so rapid, that little recollection of the birds was left. The problems of living in the changed, emptier, environment were so pressing that there was little time to look to the past. The bones themselves were left to pass on the story of a lost world, uniquely New Zealand, that had gone forever.