The Biggest Estate on Earth

by Bill Gammage

  41. A clearing near Mudgeegonga, northeast Victoria, 19 December 1963

  Wangaratta run 5, photo 57 (detail), CAD 17, UPGS. Compare pictures 21, 32, 38–40, 43–5.

  I thank Bill Tewson for showing me this photo.

  This air photo shows a grass plain lying in eucalypt forest, drained by a small creek. About 60 by 200 metres, it slopes gently west to where the creek drops over a 5-metre cliff and down a rocky gully into farmland. The landowner says the plain was ‘always’ there.125 Sapling stumps ring its rim but not its centre, indicating land kept open first by fire then by axe. The same soil continues into bordering box and stringybark, which cover steep enclosing hills. This is a brilliantly placed trap. Wallabies panicked on the plain would flee downslope and crash over the cliff, and survivors would be ambushed in the narrow gully.

  This template recurred across Australia. In the Blue Mountains Evans reported

  spaces of Ground of 3 or 400 Acres with grass growing within them that you can scarce walk through; the ground is strong and good with ponds of water which lead to the River; but when within a 1/4 of a Mile or so of it the course becomes a Rocky gully, and so steep between the hills, that no person would suspect such places were up them,126

  41

  and in Western Australia George Grey observed, ‘When kangaroos are surrounded upon a plain, the point generally chosen is an open bottom surrounded by a wood; each native has his position assigned to him by some of the elder ones.’127

  Middens and shield or coolamon trees dot this locality, and near the creek red, white and yellow paintings lie under granite overhangs. They are faint, mostly of animals, but one might be a map of the template. Rock art is uncommon in northeast Victoria. This site is the largest, with 477 motifs. People used it for at least 3500 years.128

  PICTURES 42–52: 1788 FIRE PATTERNS—TEMPLATES

  42. Steve Parish, Wineglass Bay, Tasmania, from the north, c2001

  A Steve Parish Souvenir of Tasmania, Brisbane c2001, 14 (detail). © Steve Parish Publishing,

  reproduced c~ Steve Parish and Kate Lovett. Compare pictures 37, 43–7, 50–1, 53–8.

  For help I thank Joyce Dunbabin, Margaret Harman, Jon Marsden-Smedley,

  PWS Coles Bay and Ian Thomas.

  42

  This view down Freycinet Peninsula shows Wineglass Bay at left and Hazards Beach at right. On the isthmus sharp edges divide trees from grass or heath. These have been described as boundaries between wetland and tall open forest,129 which is so in places, but at centre an edge crosses a hillock and at each end rises to wallaby traps, at the south end on a creek fan. Inland are grass necks, clearings and wetlands. Tasmanian Blue Gum dominates the forest, scattered giants above denser regeneration.

  Who made these edges? From 1824 whalers operated from the Bay’s south end, but only for about ten years, too briefly to clear generations of tree seedlings. The land has been a reserve since 1906, and was never farmed and reportedly never grazed.130 Despite a 1980 fire, this scene matches Frank Hurley’s February 1939 photo of it, except for a little more scrub then, behind the north end of the beach. JW Beattie’s April 1909 photo is the same too, except that more scrub may have rimmed the beach south of the hillock—Beattie’s colleague reported there ‘a slight rise covered with low shrubs’.131

  Here Tasmanians associated grass, heath, open forest, slopes, wetlands and two seas, for the more sheltered Hazards Beach at right has coarser sand, mounds of huge oyster and mussel shell, and more middens than Wineglass.132 In season people camped at Hazards, handy to wetlands but hidden from Wineglass’ grass corridor and traps. Habitats and resources were abundant, convenient and predictable.

  It is striking that the edges remain so sharp. Trees have germinated in the forest but not much outside it. This suggests that whereas on ground recently cleared by axe or fire trees can grow from seed stock, in grassland kept clear for centuries no seed remains, and trees must generate by edge or wind invasion. Here salt or smothering grass has apparently stopped even that.133

  43. James Scott (1810–84), Plan of Branxholm, Tasmania, 1853

  Margaret Newton, Launceston. Compare pictures 13–14, 39, 42, 44–51, 54–8.

  I thank Margaret, Mike Powell and the Queen Victoria Museum for this plan.

  James Scott, a relative of novelist Walter Scott, drew this plan, perhaps of land he meant to buy. About 4 x 3 kilometres, it is now farmland north of the town. James’ nephew JR Scott bought it in 1864, naming it Branxholm after Walter’s seat in Scotland. James’ brother Thomas named Deloraine after a Walter Scott character.

  The plan illustrates how many resources could be associated. Following the Ringa-rooma flats downstream from bottom left, Dogwood scrub on the right and fringing forest or scrub then grass (‘good land’) on the left become Dogwood and Myrtle on the right, then Myrtle on the left, then a grass strip in ‘Wattles with ferns & nettles’ on the right. Wattles are rapid regenerators after fire, ferns are rainforest survivors, nettles flourish in ash. Together they suggest fire clearing rainforest, and no fire letting wattle capture grass, perhaps to move the template (ch 8). At top, flats continue as ‘Open Plains’ on the right and forest on the left, then there is a water-locked clearing, and finally forest.

  The flats extend between low hills, one at left grass, the rest forest. The plan does not say what forest, but distinguishes it from ‘musk and dogwood’ and ‘myrtles’, so it may be eucalypt. At centre, high ground sheds creeks into swamps then ‘Open Plains’, split into remarkable diversity. At top right a small lagoon sits in grassland; another is across the river at far left, near a ford. The big swamp edges forest and hill before dividing the plains. Grass necks parallel it and edge slopes or push into forest; tree clumps give cover. Diverse plant communities blend beautifully with water and slope to feed and shelter different animals while exposing them to convenient harvest in one part without disturbing them in another. A day or two downriver in August 1831, Robinson

  went through some wooded country, the underwood of which had been burnt off by the natives, and across some extensive heathy plains . . . The country was peculiarly favourable for the boomer and forest kangaroo, consisting of heathy and sword grass plains and open forest . . . The inland natives have their hunting grounds for the different species of game.134

  Tasmanians told him this. It shows them thinking of country in terms of specific habitats and resources, and making it to suit. The plan shows how neatly and variably they did it.

  43

  44–5. James Cobon, Cape York survey plans,

  24 January 1891

  Parish Tribulation surveys C153.301 and C153.302 (detail, redrawn), ERM.

  Scale 1:3960. Compare pictures 13–14, 18, 21, 39–43, 52–8.

  For help I thank Bill Kitson and Kaye Nardella of ERM, and Gurol Baba.

  44

  This is north of Cape Tribulation: 301 runs south from Cowie Point, 302 is 3–4 kilometres further south. Cobon had to certify ‘on honour’ that each plan was accurate. Green denotes rainforest; white open eucalypt forest; N a head of navigation; T a head of tidal water; O a head of navigation and tidal water.

  ‘Vine scrub’ fringes creeks, swamps and coast though none entirely, shelters mangroves from fire, and climbs unevenly up slopes. Grassy eucalypt forest covers some but not all crests, and edges some swamps, creeks or coast. Camp clearings, at least one on ‘Good soil’, lie at a head of tidal water. On 301 springs sit on an edge. The land is pleasant, its resources convenient and predictable.

  Without fire this is rainforest country. In 1890 Crown Lands Ranger D Donavon reported of it, ‘The Scrub abounds with Scrub Turkeys, Scrub Hens, and a most gorgeous variety of pigeons, and a great variety of nuts, tubers, and wild fruit in luxuriant profusion . . . its waters swarm with a miscellaneous class of fine edible fish.’135 Yet more plant variety meant more resources, more choice, so rainforest has been cleared. A century later an elder explained, ‘You burn a little patch, for wallabies and kangaroos to liv
e on, instead of you hunting them, they’ll come to you.’136 In panic they keep to the open where they move faster, so rainforest shepherds them into water or uphill, depending on the wind chosen. While men hunted, women could get food in screened places without startling animals.

  45

  Rainforest burns when dry, but it is rarely dry. Clearing it required watchful opportunism over many generations. Maintaining it was easier (ch 7), but burning it from wet valleys but not drier hills, from one water edge but not another, all while keeping eucalypts and grassland open, demands detailed local knowledge of fuel loads, winds and future resource needs, and skilful, timely burning. Yet there were hundreds of balds on Cape York.137 Christie Palmerston’s description shows how prized they were, and how closely managed:

  Its upper part is free of rocks, indeed of everything else, it having been burnt in the early part of this year by an old aborigine named Wallajar, to whom it belongs . . . The S.E. end or horn . . . is ‘Care-ing-bah’, by some called ‘Tachappa’ . . . Its N.W. end is named ‘Koorka-koorka’ [Gourka-Gourka], and is possessed by an old man, whose name I forget . . . four of the young fellows who accompanied me have hunted over [it] often. The other two . . . though born and hunting within its shadow, never were here before.138

  Newcomers at once thought of agriculture. In 1890 one wrote of the lower Cape, ‘The area available for agriculture is cut up into isolated pockets, varying in depth . . . from half a mile to one and a half miles. These pockets comprise rich forest land with belts of dense tropical jungle along the coast.’139 Donavon reported ‘alternate patches of scrub and forest country. The forest land is well grassed . . . [with] a nice friable decomposed vegetable loam well fitted for agricultural purposes.’140 Cobon thought this land ‘highly suitable for an agricultural township and would be readily taken up in small areas for tobacco’.141

  There is no agriculture. By 1994 301 had no grassland and rainforest was edge-invading eucalypts, though eucalypts still dominated ridges. In 1991 302 had a little grass but was similar.142 Alma Kerry recalled that Cowie ‘used to be open, now its grown back really thick, we can’t get wukay [yam] on the beachfront anymore because of the rainforest’.143

  46. Gatcomb (l and c) and Goderich (r) Plains

  near Guildford, Tasmania, 12 April 1949

  47. Gatcomb and Goderich Plains, 16 November 1984

  46: Valentines Run 6/22139, FT, c~ Bill Tewson. 47: Colour M486, 1014–063,

  NW Forests run 25. Base image by TAS MAP (www.tasmap.tas.gov.au). © State

  of Tasmania. Compare pictures 13–14, 18, 21, 29–30, 39–45, 48–51, 53–8.

  These photos illustrate fire and no fire in rainforest. In 1949 this land was part of the VDL Company’s 61,000 ha Surrey Hills block. The Company neither made nor allowed any sustained use of it, though the plains may have been grazed, or fired by wallaby snarers.144 Ridge or valley, damp or dry, in both pictures trees are reclaiming grass and heath. White eucalypt trunks flag pockets once open, especially north of Gatcomb in 1949, and darker patches and edges show rainforest advancing.

  In 1788 Tasmanians worked hard on this land. At bottom the Wandle River runs between open eucalypt forest and grass pockets on and off its flats. People protected the river, called Lare.re.lar, platypus. Robinson found platypus there,145 so the banks were less exposed to cool fires which kept open the ridges along Gatcomb’s south edge. The ridges, thick with snakes, drain boggy creeks alive with frogs to a swamp below Gatcomb’s north tip.146 As their shape conveys, on each plain people used hot northeasterlies to drive fire southwest (ch 8), the fire front expanding and pushing grass neck traps into forest. The trap at Gatcomb’s north tip was murderously efficient. Caught in the open and ringed by forest, prey would flee south, hit the swamp, and flounder into club and spear. Survivors had to battle upslope over boggy ground into a cordon of hunters, then flee to a similar template judiciously located and primed with pick. People could always find them. At least five distinct fire regimes have conveniently associated rainforest, open forest, grass, swamp, river and ridge.

  46

  47

  In 1827 Henry Hellyer, a tireless traveller, followed a Tasmanian track, still clear in 2001, across this country,147 and wrote that plains just north ‘resemble English enclosures in many respects, being bounded by brooks between each, with belts of beautiful shrubs in every vale’.148 He mapped a sequence of plains large and small in rainforest from Emu Bay (Burnie) south to Surrey Hills.149 By 1828 he knew that Tasmanians had made them: like Eyre in the dry South Australian mallee (ch 1), he found logs in the grass. Researchers have identified many such rainforest remnants.150

  Clearing to plant Shining Gum has obliterated these and other beautiful and complex templates in the northwest.

  48. An edge of Gatcomb Plain, 14 February 2002

  BG. Compare pictures 13, 18, 21, 31, 34–7, 40, 44–7, 49–58.

  Gatcomb carried at least three communities: White Grass and open forest on its south ridge and dry slopes, mosses and water plants on creeks and swamps, and elsewhere a wonderland of ferns, orchids, lilies, everlastings and other flowers, and heath.151 Here heath has captured grass, and in its shelter rainforest advances.

  You can see this advance. Edge trees are small, but the skyline shows big pre-1788 Alpine Ash, Messmate or Swamp Peppermint. They can’t grow in rainforest: there is no light (ch 1). They are there because rainforest was not. This can happen only if rainforest is burnt back. Not by a bushfire: they are rare in rainforest, and after a lone fire rainforest slowly returns, as it is now. Decades or centuries of judicious fire drove and kept it back. Once grass and herbs were established fire was needed about every 3–5 years to promote them and kill eucalypt seedlings. The big eucalypts mark a boundary of a plain Tasmanians made.

  Robinson came to recognise boundaries: ‘[If] when travelling in these [rain]forests you discover many stringy bark trees—or badger dung—you may always rest assured that you are near to an open country.’152 Similarly, in the Eungai Creek catchment (NSW), Clement Hodgkinson’s Halt near a fern tree scrub (1843?) shows eucalypt forest with no undergrowth almost touching dense rainforest, the narrow space between bridged by tree ferns, which like sunny clearings.153

  48

  West of Alice Springs in September 1872 Giles wrote, ‘The little plain looked bright and green . . . The grass and herbage here were excellent. There were numerous kangaroos and emus upon the plain. [It is] . . . fringed with scrub nearly all round . . . [At] the foot of the hills, I found the natives had recently burnt all the vegetation.’154 He might have been describing Gatcomb.

  49. Wentworth Marmaduke Hardy, Plan in the Hundred

  of Kadina, Yorke Peninsula, SA, 7 September 1874

  Hundred of Kadina diagram 11, LTO . Compare pictures 39, 43–7, 50–1, 53–7.

  For help I thank Mick Sincock and Leith MacGillivray.

  49

  Hardy’s plan covers about 3.76 kilometres in width. He coloured mallee scrub green, open forest yellow-green, grass plain pale green. A ground search confirmed that all three communities straddle different heights and soils.155 Hardy’s diagram 10 of the land immediately south is similar.

  Before 1874 men cut wood here for copper mines, no doubt extending Blackoaks Plain, but not destroying its 1788 pattern. The country is much drier than at Branxholm or Gatcomb, but is shaped similarly. Small grass clearings sit in scrub and open forest, and grass necks jut into both. Trees sheltered animals; clearings or necks lit in sequence moved them predictably.

  In 1839 Eyre noted these patterns further east, reported them common in his ‘various wanderings in Australia’, but could not explain them. He puzzled at how grassy plains suddenly appeared in scrub or forest, when dead timber showed that trees could grow on them. He thought deliberate fire might explain clearings in thick mallee, but not in open forest where the trees stood too far apart for all to be killed by one fire.156 He was a most acute observer, sympathetic to Aborigines, but even he could not ima
gine that the patches were man-made.

  In the 1970s Mick Williams used over 1850 plans to build a picture of the mallee north of Adelaide, including here. He considered various natural explanations for its anomalies, but concluded,

  The mallee cover was by no means as complete as is commonly thought . . . along the river courses [it] tended to disappear and be replaced by open grassy land with wider spaced trees—the eucalypt savannah woodland . . . Of great interest is the evidence of the open, largely treeless ‘plains’ within the scrub . . . [some large], many smaller ones. These ‘plains’ may have been due to aboriginal clearing by fire.157

  50. Frederick Montague Rothery (1845–1928), Ups

  & Downs block, Bundaleer Station, Qld, 1877–8

  [Rothery 37]. Compare pictures 39, 43–7, 49, 51, 53–7.

  Bundaleer is on the New South Wales border east of the Warrego. In 1858 the country was dry, sandy ‘grassy forest, with ridges of dense brigalow scrub’.158 The run was taken up about 1861, and grew to 1176 square miles, divided into blocks. In 1877–8 Rothery, the manager, drew a plan of each block. Ups & Downs, 100 square miles, Rothery’s scale an inch to two miles, got an occupation licence in 1876. Rothery described it as ‘Thick belts of Gidyah, Pine, Brigalow and small Scrub with broken plains scattered all through with open Coolibah flats. Cotton and Saltbush more or less all through.’159 Except that Mulga was common elsewhere, this typified Bundaleer.

  Rothery linked each tree species to its soil type. The pine ridge at left in picture 50 (overleaf), coloured orange, is red sand. Along Nebine Creek’s left bank grey alluvial carries Coolibah, box, saltbush and grass; the right bank is more open. At bottom left blue denotes Brigalow on red sandy loams; elsewhere blue-green patches mark Gidgee or sometimes Belah on heavier red-brown earths. Yellow denotes Mitchell Grass, Never-fail and Queensland Blue Grass plains.160

  Yet soil alone cannot explain a distribution so erratic. The map suggests which tree might occupy adjacent grassland, but not why no trees at all are there. Brigalow and Gidgee are in clumps and belts; grass and scrub alternate. At centre left Wilga sits under the pine ridge, with trees shaping a grass vista to Coolibahs on the creek. North is scrub; behind shielding trees south, grass necks fragment Brigalow. Similar patterns are across the creek.