Tom Simpson

Deep Secrets

Discovering New Zealand’s Tropical Past

Written by      

There was a sense of expectation as we drove down the broom-lined track towards the Manuherikia River. It was December 2001, and we hoped that exposed layers of sediment in the Bannockburn Formation would reveal a bone or two, bones that would help us understand what New Zealand may have been like 16 million years ago. We could not have known what a treasure trove lay in wait.

The Manuherikia River flows down a valley of the same name in Central Otago, bubbling through riffles and pools and over millions of years of history. This was once an enormous lake stretching from Naseby to Bannockburn and Ranfurly to the Waitaki Valley—an area 10 times that of Lake Taupo.

It swarmed with life, which thrived then perished in the silty lakebed and remains there to this day, a quiet record of creatures unique to New Zealand from a time well before humans.

Palaeobiologist Trevor Worthy excavates fossils from a riverbank in St Bathans, Central Otago. Often metres of rock are removed to reveal a special layer of ancient sediment which is sometimes just a few centimetres thick. Tools of the trade range from the large bucket on a 10-tonne digger to fine dentistry probes repurposed for a new life scraping silt from around fossils. Once exposed, bones are carefully packaged and removed to a laboratory for preparation under a microscope—cleaning, consolidation and reassembly of the fragments.
Palaeobiologist Trevor Worthy excavates fossils from a riverbank in St Bathans, Central Otago. Often metres of rock are removed to reveal a special layer of ancient sediment which is sometimes just a few centimetres thick. Tools of the trade range from the large bucket on a 10-tonne digger to fine dentistry probes repurposed for a new life scraping silt from around fossils. Once exposed, bones are carefully packaged and removed to a laboratory for preparation under a microscope—cleaning, consolidation and reassembly of the fragments.

Alan Tennyson of Te Papa museum, Craig Jones of GNS Science and Trevor Worthy of the University of New South Wales began surveying exposed sections of gravel, mudstone and shale along the riverbank. Within minutes, someone had discovered a bone protruding from a bed of the Miocene sandstone, and in the next hour or two the 100-metre exposure revealed several more. They were small, brittle and bleached where they protruded from the sediment. Some were broken by the turbulent course of the past 16 million years, others were perfectly preserved, the finest details cushioned by the mud as if they were wrapped in tissue paper.

These were discoveries beyond a palaeozoologist’s wildest expectations—the bones of creatures long extinct and unknown to science, simply waiting to be dug out.

The next day, we headed to another site, affectionately known by palaeontologists as Croc Site, because it was from here that a crocodilian bone had been described in 1997—the first, and until then the only, example found in New Zealand. We quarried lumps of the sandstone and put them in our sieves to break down in the flow of the river. Over the course of an hour, the silt was washed away, leaving pebbles and a macabre assortment of bone fragments. First bird bones, and then a great flurry of excitement—a tooth, unmistakably crocodilian. That New Zealand had once hosted a crocodile of sorts was interesting enough, but here we were all gathered around the sieve staring at a 16 to 19 million-year-old fragment of history, complete with root and crown, as though it had fallen from our own mouths.

The presence of a large crocodilian formerly inhabiting New Zealand was first revealed in 1997 by Ralph Molnar and Mike Pole. We now have over 100 fossils, and many teeth, but the definitive skull or even a jaw full of teeth remains elusive. Mysteriously, the fragments we do have show it was rare in the aquatic lake environment of Lake Manuherikia—perhaps it was more terrestrial in habit like those once found in Fiji and New Caledonia. Like modern crocs, this most southern of crocodilians would have lain in wait for prey to cross its path and ambushed at speed. It was likely part of a southwest Pacific group of crocodilians called mekosuchines that were common in Australia in prehistory and were finally displaced by the “salties” of the present day. There is nothing quite like finding a spectacular fossil, especially one the subject of a search lasting years. With luck, many of the questions surrounding this magnificent predator will be answered on our next dig.

Little did we know then that this find would set us on a journey of discovery that would continue for almost a decade, with expeditions every year. Through the heat of droughts and the icy cold of sleet and snow, the Otago hills offered up a hugely significant trove of fossils, revealing with each dig another small piece of New Zealand’s prehistoric past.

Paul Scofield of Canterbury Museum and colleagues Suzanne Hand and Mike Archer from the University of New South Wales have joined the search in recent years, excavating several sites around the St Bathans township. Initially the work was funded by the Foundation for Research, Science and Technology, but, in recognition of the international significance of the find, grants for the past three years have come from the Australian Research Council.

The discovery is unique in part because of its age. Most of the bones are thought to be between 16 and 19 million years old, in the Miocene period, long after the extinction of the dinosaurs.

[Chapter Break]

Think of time as if it were a ruler; if a year was the equivalent of just one millimetre, then 1000 years represents a metre. Join me now on a journey, the first half step of which represents the arrival of humans in New Zealand.

The two-metre mark is when the Roman Empire was at its height, the three-metre mark when early Polynesians began their push into the Pacific, and 100 metres down the track early humans walked out of Africa into Asia and Europe. Some four kilometres on, we reach the point that humans first evolved, and somewhere beyond the 16 km mark, as far as Upper Hutt is from Wellington, we reach our destination in the Miocene period.

The St Bathans landscape looks very different: a vast lake stretches out to the west and south in front of us time-travellers, and from the north to the west a range of high, rounded hills frames the horizon.

Broad, braided riverbeds fan out from these hills, carpeted in white quartz pebbles, and, not far off, one of these enters the lake before us.

It is late in summer, and the land is in the grip of drought. The waters have receded and broad flat mudflats extend far from the lake edge. An abandoned channel from the river delta contains a remnant pool teeming with life—herons stalking in the shallows, kite-like raptors sitting in overhanging palms and, further along, a group of gulls squabbling over a fish. Seeking the source of this excitement, the time-traveller may soon notice beds of dead freshwater mussels, schools of fish swimming back and forth and galaxiids struggling and gulping air in the deoxygenated water. One fish, somewhat bigger than the rest, swims erratically, then finally goes belly up. The kite swoops down from its perch and hauls it from the water.

About 16–19 million years ago, a superlake stretched over 5600 km in what is now Otago. Geological dating is not yet precise, but mapping is detailed and shows that this was one of the largest lakes in the world. A wide braided river entered it via a delta. In Zealandia’s subtropical climate of the period, this shallow lake and surrounds supported a diverse and abundant fauna. Waterfowl were most numerous, and included geese and shelducks (at bottom left). Strange birds vaguely like flamingos waded in the shallows, and herons stalked the margins hunting fish, including gobies or bullies, smelt and galaxiids—descendants of which make up the whitebait in today’s southern rivers. Several kinds of small wading birds and skulking rails foraged along the lake shore. Swiftlets hunted insects high above, gulls scavenged along the shoreline and eagles and other raptors hunted the numerous birds. In the nearby woodlands, vaguely familiar birds including several parrots (pictured on opening spread) were abundant, but one not at all familiar was the adzebill, seen here devouring a lizard. Adzebills, or Aptornis, lived on in New Zealand to become the North and South Island adzebills of the Pleistocene, an endemic family, but were lost like the moa when humans arrived.

Numerous small waders feed on the exposed mudflats, and out in the water there is a group of larger palaelodids, the ancient short-legged cousins of flamingos. Walking along the shore, we come to a small embayment where the water is deeper and extends up to a shallow bank. Here reeds and grasses grow close to the edge and small flightless rails sneak about close to cover. Great flocks of ducks clutter the water. More waterfowl graze on grass on the banks—a slightly longer-legged version of the paradise shelduck, and a goose. Further back, a stout metre-high bird with a huge decurved bill, an adzebill, is ripping rotten wood and bark off a fallen tree, searching for large insects or lizards.

Extending back from the lake edge is an open woodland with casuarinas, eucalypts, palms and cycads. The grass is dry and sparse. A large nest hangs in an isolated eucalypt, and soon its owner appears, an eagle with a two-metre wingspan, flying in carrying a duck to feed two hungry eaglets. Away across the flats, and along the riverside, the trees are taller and closer together and podocarps tower upwards. Noisy flocks of parrots feed in the canopy, and nearby large pigeons gorge themselves on fruit. Numerous small song birds flit among the trees, and smallest of all, a thimble-sized ball of feathers with no tail, emits a high-pitched squeak. Soon, another joins it and together they rapidly dance over a tree trunk, investigating small cracks in the bark, then disappear from view. Farther along the raised bank of the river, a clearing allows the sun to warm the forest floor.

There, above a well-beaten trail to the riverside, a three-metre-long crocodilian slumbers in the sun.

And off in the undergrowth, a rustle betrays a furry critter the size of a small rat feeding on invertebrates in leaf litter.

[Chapter Break]

It is now 2010. The team has excavated and sorted tonnes of sediment and identified thousands of bones. We have revealed much about the palaeo-Lake Manuherikia and its environs. Geologists have mapped its extent—a staggering 5600 km extending from Central Otago to the Nevis Valley and Bannockburn to the southwest and west, to Naseby in the east, and from north in the Waitaki Valley to some way south of Ranfurly. It existed on Zealandia, long before it took the shape of modern-day New Zealand. This lake was preceded by a broad alluvial plain, across which flowed a braided river. Later the river became impounded and swamps with diverse rain-forest vegetation developed, forming what was to become deep beds of lignite. Then the lake flooded the forests, and persisted for millennia while the climate became seasonally dry.

The layers of rocks record oscillating water levels. Deep water supported ostracods and fish as shown by their fossils in layers of clay. Then came a period of drying out, with cracks forming in the exposed mud. Next, a flood swept through, laying a veneer of sand and small pebbles over the other deposits. It carried with it the bones of dead fish and shells from the dried channels, bones of dead birds and other animals, even shells of tiny molluscs from the nearby shrublands. Some of these were swept back and forth in the shallow littoral zone of the lake and some accumulated a patina of calcite laid down by microscopic algae which lived on the bones. Soon deep water returned, and in time it buried everything in sediment. This was a cycle repeated numerous times over the millennia.

In some places the layers of rock contain fossil accumulations as rich as any found in the world. Often they are as densely spaced as one large bird bone per 30 square centimetres, but it’s the small bits that fascinate us. We wash the sediment, wash it again, and, after it dries, sort it under a microscope, half a teaspoon at a time. In one quarry, we have excavated 110 sq m, putting the entire 10–15cm-thick target layer through this process, amounting to some 10 tonnes of concentrate.

Palaelodids, wading birds most closely related to flamingos, are an entirely extinct family. They were once found across the globe, including New Zealand, between 30 and 2 million years ago. They differ from flamingos by having a shorter tarsus (the fossilised distal tibiotarsus is pictured above), about half the length of the tibia, and lacked the recurved bill of flamingos, their bill being rather more like a heron. Like flamingos, however, palaelodids lived in large flocks on lakes and probably fed in the shallows on aquatic invertebrates such as insect larvae, and crustaceans.

It has been a long and laborious process, but never boring. Placing a teaspoon of processed sediment on a tray and peering at it through a microscope engenders a huge sense of excitement and anticipation. You are the very first person to look at these fossil bones, which are up to 20 million years old. Every spoonful has the potential to reveal a new species, or more parts of one already found.

We had spent weeks searching through muck before it revealed itself, a beautiful sparkling little transparent mammal tooth, with all its cusps and roots. Twenty minutes later, another appeared. Weeks elapsed, and doubts crept in that we were missing them. Then, after sprinkling a few grains of sediment into the Petri dish—grains that looked no different from the last few million—there, as clear as a bell, was yet another beautiful tooth.

But teeth are not the only gems. Perfectly preserved bird and lizard bones appear under our microscopes, and tiny, exquisite snails barely 3 mm across. Each gem is carefully picked up with leaf tweezers, sealed in a gelatin capsule, like those vitamins are stored in, and labelled.

To become lazy, to miss a single tiny tooth, may be the difference between finding a new species or remaining forever in the dark.

The haul so far has been impressive. It contains the large hyriid clams (freshwater mussels) and small freshwater gastropods that we expected, but the small 2–4 mm-wide land snails, still retaining colour in their shells, were a surprise. Fragments of yabbies, yabbie-buttons (the calcium depots for shell changing) and even a couple of examples of stomach-grinding plates attest to the presence of freshwater crayfish, much like those found in the waterways of Otago today. But we are vertebrate palaeontologists, so bones were what we wanted, and bones by the million are what we got. Fish accounted for 99.99 per cent, umpteen thousands of them, and many more thousands of otoliths—calcium carbonate structures from within their heads. We now know that these fish derive from several kinds of bullies similar to those seen in New Zealand streams today, smelt, and several kinds of galaxiids—famous as the adults of that New Zealand delicacy, whitebait. Surprisingly, not an eel remnant has been found.

But for every hundred or so fish bones, we get a fragment of something else, and one in a hundred of those is a bit we can identify. After nine years sorting, we have identified many species. We have three kinds of frogs. Reptiles are represented by lizards, a couple of species each of skinks and geckoes much like those of today, and tuatara-like animals as well. But we can also be sure that long ago, New Zealand was not the safe place it is today—it harboured crocodilian creatures up to three metres long, now known from many bones and teeth.

But birds dominate this fauna. There are more than 35 kinds in this fossil record, but as yet only about a dozen have been named.

Excluding fish, the St Bathans fauna was dominated by waterfowl. By far the majority of these belongedExcluding fish, the St Bathans fauna was dominated by waterfowl. By far the majority of these belonged to three species of stiff-tailed ducks named Manuherikia, represented by numerous bones such as the fossilised humerus pictured above. Contemporary stifftailed ducks include the blue-billed duck, musk duck, and the weird pink-eared duck, all of which now inhabit Australia. This group of ducks is one of the most primitive now surviving globally, and while global in distribution they are only few in number. Evolution has marched on and produced the “dabbling” ducks which now dominate duck diversity globally. Also associated with St Bathans fauna is a shelduck and a goose, the latter something like the Cape Barren Goose of Australia. to three species of stiff-tailed ducks named Manuherikia, represented by numerous bones such as the fossilised humerus pictured above. Contemporary stiff-tailed ducks include the blue-billed duck, musk duck, and the weird pink-eared duck, all of which now inhabit Australia. This group of ducks is one of the most primitive now surviving globally, and while global in distribution they are only few in number. Evolution has marched on and produced the “dabbling” ducks which now dominate duck diversity globally. Also associated with St Bathans fauna is a shelduck and a goose, the latter something like the Cape Barren Goose of Australia.

We have moa, or at least moa ancestors, that were large and flightless, represented by eggshell and a few scraps of bone. The eggshell hints at a bird every bit as big as the little bush moa Anomalopteryx that was present in New Zealand immediately before the arrival of the first humans. Eight species in six genera of waterfowl including ducks, shelducks and geese combine to make St Bathans the richest-known site of Miocene waterfowl fauna in the world. The flamingo-like palaelodids are represented by a single species, unlike those found in Australia at a similar time. There is a single species of heron of a new primitive lineage, the first and oldest for Australasia. Raptors are represented by accipitrids, two kinds of eagles and a kite-like raptor, though no falcons or owls have yet been revealed. A large gruiform, a crane-like bird, probably the ancestor of Aptornis, the weird adzebill known from the Holocene, is represented by vertebrae, a part femur and a phalanx. In the same order, but of diminutive size, are at least two species of small rail, one of which was flightless and barely larger than a sparrow.

Even seabirds have turned up. There are the bones of diving petrels of a form very similar to those living now, and doubtless they would have fed at sea during the day and nested on inland hills at night as they do now. Other waterbirds include gulls and waders, somewhat like the dotterels and sandpipers that frequent lake margins today. Of more terrestrial birds, at least two kinds of pigeons, three parrots, an owlet-nightjar, a swiftlet and four or more passerines ranging from magpie-like birds to a diminutive wren existed about the lake. Many of these birds have yet to be described in detail.

But most surprising of all are the mammal bits. A terrestrial mammal, dubbed the waddling mouse, was a find of huge international significance. About the size of a small rat, it is neither a placental nor a marsupial mammal and so, like leiopelmatid frogs and tuatara, represents the last survivor of a lineage that died out elsewhere many millions of years ago. We have several bits of this intriguing critter now, but not the molars that mammalian palaeontologists crave. We do, however, have more than a dozen teeth and bones of several kinds of bats. One is a mystacinid, a short-tailed bat ancestor, revealing the long heritage of New Zealand’s special bat. Another is of an entirely new family unlike any in Australia.

On each field trip we collect new and exciting animals: 2009 saw one of the largest—the first bone of an Aptornis ancestor—and one of the smallest, representing an intriguing new family of bird.

One of the most surprising discoveries in the St Bathans fauna was remains of bats—surprising because lake shores are not where one expects to find many dead bats these days. They may have washed into the lake from surrounding woodlands, or the bats may have eaten fish hunted on the wing over the lake. The fossils (arm and leg bones and teeth) reveal at least four species in three families of bats. An ancestor of Mystacina, the existing New Zealand short-tailed bat is there, and another “micro-bat”, similar to—but not related to—the existing long-tailed bat. Most exciting are bones and teeth indicative of an entirely new family of bat with a similarity to bats now present in South America.
One of the most surprising discoveries in the St Bathans fauna was remains of bats—surprising because lake shores are not where one expects to find many dead bats these days. They may have washed into the lake from surrounding woodlands, or the bats may have eaten fish hunted on the wing over the lake. The fossils (arm and leg bones and teeth) reveal at least four species in three families of bats. An ancestor of Mystacina, the existing New Zealand short-tailed bat is there, and another “micro-bat”, similar to—but not related to—the existing long-tailed bat. Most exciting are bones and teeth indicative of an entirely new family of bat with a similarity to bats now present in South America.

New Zealand has long been considered the land of birds. Now we know for the first time that this was the case in the Miocene period too. But back then there were quite a few differences. We had a terrestrial mammal, whereas none survive today, and a crocodilian, not just a nipper either.

Other clues as to what has happened are seen among the species that we have lost. Ancestral flamingos and swiftlets are strictly tropical today. And in the plant community the story is the same—eucalypts, casuarinas, cycads, palms and a multitude of laurals have all deserted New Zealand for warmer latitudes. From the time of the palaeo-Lake Manuherikia, the Earth cooled 8ºC during a period called the middle Miocene Transition, 14–13 million years ago. On top of this, the Pleistocene glaciations over the subsequent 10 million years brought another 5ºC drop in temperature—rather too much for New Zealand’s tropical heritage, which was all but eliminated.

Nevertheless, many of the animals we had in the Miocene—the frogs, tuatara, skinks, geckos, and many of the birds (owlet nightjars, shelducks, pigeons, diving petrels)—were allied to those surviving today. None of these species existed in Australia, and few even shared the same genus, an exception being the Palaelodus, which was globally distributed. Even the waterfowl, which are relatively well described and included at least eight species, were all different from those then in Australia. Clearly the fauna, 16 to 19 million years ago, was already very much Zealandian, reflecting a long and unique heritage dating back to our physical connections with Gondwana. We have lost many of our heritage taxa since, allowing a few Australian immigrants to arrive, and evolution has seen the replacement of many other species such as the waterfowl and rails and mystacinid bats, but Zealandian we remain.

Those that survived the ever-descending temperatures, the retreat of the subtropical forests and the raising of the mountainous landscape we know today, became masters of adaptation and now stand out among animals elsewhere on Earth by their appearance, habitat and behaviour. As much as New Zealand is unique for the fauna we have lost, it is also famous for what has remained. This is a nation of survivors, which have been encouraged by circumstances to do things somewhat differently than elsewhere. The survivors of St Bathans were adaptable and unspecialised, qualities which have lived on in the cultural psyche of New Zealand society that followed.

More by