The Greening of the Arid Boundary

The Australian desert fringe has been greened like no other. It is now a place of extraordinary agricultural development, with generally productive low-input farming. On a continent with so little higher rainfall country this region has been able to make a substantial contribution to the national economy. Long desert boundaries on other continents are fragmented between countries or more remote. The Australian achievement is all the more remarkable as there is only one relatively small river—the Murray—discharging through it, so there are no vast deltaic systems or large water supplies like the Nile. Water supplies have had to be secured by other inventive means, especially pipelines.

Many things and people have come together to achieve this development. One was an understanding of ecology, of the soil and its formation in relation to plant growth. Another was the attitude that agriculture can be continually improved; it is not static. Above all it demanded the presence of stubbornly persistent farmers, often through the generations, necessarily inventive and self-reliant.

In the beginning of settlement by Europeans, with broad exploration, and the establishment of Melbourne, Adelaide and Perth in the 1820s and 1830s, there had been recognition of farming possibilities, especially on the moister edge where there is a transition into forest. There appeared to be a substantial belt of land, thirty to 200 kilometres wide, defined by moderate rainfall patterns. There was a moderately reliable winter seasonal element of 200 to 300 millimetres, with a growing season normally of five to six months, May to October. Native vegetation was mainly multi-stemmed eucalypts called mallees, sometimes sparse heath.

The belt begins around Geraldton, arcs inland from Perth across north of Esperance, to the coast, and gives way to the Nullarbor Plain. In South Australia it begins around Ceduna, and includes most of Eyre Peninsula. South from Port Pirie it includes all of Yorke Peninsula, then lies east of the ranges, a wide sweep to Eudunda, across the Murray Mallee then straddles the Adelaide–Melbourne highway. In Victoria a feature is the Big and Little Deserts, with different soils clearly recognisable along the Pinnaroo–Ouyen road, then the well-known Victorian Mallee region across to Swan Hill, through New South Wales around Narrandera, then sweeping north, petering out beyond Dubbo where the ecosystems are different in climate and vegetation.

In due course it was realised that much of this desert fringe had soils that were very unattractive for farming, especially the deep sands. This was so for a large proportion of the better-rainfall land in Western Australia, less, but still considerable in South Australia, and only a small proportion in the well-endowed Victoria. (The Big and Little Deserts were also very interesting ecosystems of native plants, valued in due course as botanical and recreational entities.) Thus Western Australia, despite having 25 to 30 million hectares of quite good rainfall land, struggled to feed its people from settlement in 1829 to near the end of the century. For the remaining land in this fringe, access was an important issue. Land suitable for cropping and near the gulfs in South Australia allowed small ships to draw near and be loaded using shallow-draft barges, and later, extended jetties. Even on these lands yields declined, despite adaptations like alternating cropping with bare fallow.

Goyder’s Line, the limit for cropping, was drawn in the 1870s by the Surveyor General in South Australia, based on rainfall records and vegetation. But cropping with what nature had given was about to change forever. One important event was the arrival of J.D. Custance, recruited from England as Professor of Agriculture to establish an Agricultural College, Roseworthy, north of Adelaide, in 1884. He brought news of the discovery of the need for phosphate on cereal crops, and through the college, began to provide technical support people for farmers. However, more science was needed to underpin good field work, so a university course was established early in the 1900s, eventually bringing Hugh Trumble to South Australia. An agronomist with a strong interest in climatology and an appreciation of the use of grass, he was the son of the Australian Test cricketer of the same name; he wrote an interesting book, Blades of Grass. Trumble typified those whose approach was different from Goyder’s: Why accept the boundary? Why not develop new varieties and better farming practices? If these things could push this 2000-kilometre boundary out ten kilometres, it would create 20,000 square kilometres (2 million hectares) of farm land.

Thus, in effect Goyder’s Line was moved north as better cropping systems successfully increased yield per unit of winter rainfall. In a study that repudiated some sweeping claims by the American geographer Jared Diamond in his book Collapse that Australian farming had collapsed, a farm right on the line was studied: it had shown an increase up to 1970, then had a further doubling of yield per unit of winter rain between 1970 and 2005. 

Soil science is a relatively recent study. It throws light on soil characteristics and their usefulness for plant growth. For instance, much of Europe and North America was covered by glaciers that crept across the landscape grinding a mixture of rocks into fine material, at the same time protecting the forming soil from leaching by rain. Once the glaciers melted, the remaining soils were rich and deep, if at times stony.

Southern Australia was different. In the 1930s the ecologist Bob Crocker put forward a comprehensive suggestion for the geological history of the area and soil formation. A large continent, Gondwana, broke away from South America and South Africa and drifted in the southern seas for millions of years. Eventually Gondwana split up, forming, at various stages, New Caledonia, New Zealand, New Guinea and, about 40 million years ago, Australia—which became the largest separated land mass on the planet. Throughout, normal earth building and landscape processes continued: rock weathering to form soils, erosion to form mountain ridges, valleys and plains.

Plants and animals continued evolving in a largely closed system and, though some had similarities with plants and animals in South Africa and South America, many developed unique features. Thus, though the species are mostly distinct, the broader groups of families and genera are rarely peculiar to Australia. Curiously, the areas with low fertility soils developed a rich, diverse and interesting flora—until recently parts of south-west Western Australia were more notable for their wildflowers than their agriculture.

Climate change has always affected soils. It is generally agreed that in the Post-Miocene there was a period of global cooling with severe ice ages and a lot of water “locked up” as snowfields, glaciers and polar ice-caps. Bob Crocker suggested that, as sea level fell, huge areas of the continental shelf were exposed (fairly quickly in geological time) and not colonised by plants. Prevailing on-shore wind—as along the south-east of South Australia, on Eyre Peninsula and parts of south-west Western Australia—drifted vast amounts of sand onto the land. Between ice ages, the warmer, wetter periods allowed some plant growth on these sands, and also leached much of the slightly soluble lime (calcium carbonate), leaving a sandy soil over limy material, which we call calcrete. Notwithstanding, these sandy soils would have been subject to denudation in dry times, allowing winds to carry sands inland, forming dunes and drifts and leaving the residual calcrete exposed.

Thus, near the present coast of South Australia around Kingston and Robe, we have a limestone (calcrete) landscape of rocky ridges with shallow soils; inland towards Naracoorte limestone ridges parallel to the coast, sand dunes (and some heavy black soils where drainage between the dunes is impeded): a very limy landscape. To the east seemingly endless dunes of siliceous sands make up the Big and Little Deserts. The Bordertown district didn’t get covered with sand but did receive fine limy loess that modified the structure of its soils. Eyre Peninsula reflects these same events, with the undulating landscape and shallow stony soils along the coast from Port Lincoln to Streaky Bay and Ceduna. Anyone who has driven the coast road, the Flinders Highway, will never forget the lunar landscape. There are siliceous sands to the east towards Spencer Gulf contributing to the shallowness of the gulf waters.

Another “landscape maker” was draining and eroding the vast up-lifted plateau along the east coast—the Great Dividing Range. Though the Murray River is a trickle compared with the great rivers of the world (the Yangtse is about seventy-five times larger) it made its mark. Once out of the mountains, for most of its existence it flowed past the site of Yarrawonga, on past Barmah, through Mathoura to Swan Hill—from the air one can see the old river bed. Then, quite unusually for the western plains, a small earth fault block of perhaps twenty metres was uplifted, more or less from Echuca in a northerly direction to Deniliquin. This dammed the river and established conditions for a huge new red gum forest at Barmah, the river splitting: two-thirds of the water flowing through Echuca, the other third down the Edward River through Deniliquin, eventually joining the Murrumbidgee. Downstream near Swan Hill the main Murray River used to flow south, its mouth somewhere in the Wimmera, around Horsham. A wide delta of silty material was laid down, becoming the basis for the rich Wimmera soils, stretching west to around Bordertown, the mouth to the sea somewhere south-west, abundant sand blocking the mouth as is happening further north-west now.

Blocked to the south, and possibly with some more minor faulting, the Murray River formed the Koondrook wetland that became another forest, then wandered north-west to today’s Mildura—so, recently, the Mildura area, like the land around Cairo, got lucky with a river diversion, and, come inventive humans, ceased to be a desert place. On to Morgan, somehow south to Murray Bridge and the lakes, perhaps meeting an old stream from the Flinders Ranges coming down from Burra. The depositions buried some streams, now relics, providing underground water in the Pinnaroo and Bordertown districts, even some water emerging near the coast at Salt Creek on the Coorong.

The large sand particles more or less roll or bounce across the countryside, then tend to gather into dunes. Limestone would have been ground into flour-like material and blown in the wind, much further than sand, landing on the soil as loess and being subsequently leached into the soil profile to varying degrees, depending on rainfall. Thus, further from the coast, beyond the Big Desert, are soils with limy material in the subsoil: mallee soils, riverine plain alluvium, alkaline rather than acid, which is important when considering plant nutrition. And there is salt: some is associated with ancient sea beds, but most owes its origin to the low-intensity rain coming off the Southern Ocean in winter that brings roughly half a million tonnes of salt a year to the river basin. At times when there is too little rainfall to wash it into the rivers, there is salt everywhere, much of it redistributed when land is left without strong vegetative cover (either trees or pasture) such as the period from 1850 to 1950.

Australian soils are often said to be old, but these Post-Miocene soils are in fact geologically young. There is also a tendency to think of young soils as fertile and old soils as poor because they have been leached for a long time by rain. However, this assumes young soils are from the breaking down of complex parent rock or recent alluvium or lava flows, and this is not always so—the Post-Miocene soils have formed from poor marine material, so are low in nutrients though young. 

Plants clothe the planet, taking root in the soil that has been formed from breaking down the rocks, with a huge range of forms with the remarkable ability to form mixed plant ecosystems—and varying in invasive capacity. Ships bringing farm animals needed hay for feed for the journey, and often picked some up at Mediterranean and South African ports. The ships were usually cleaned near the port of arrival and wastes, including seeds, were disposed of nearby. The climate of the new land had similarities with the homelands of these seeds—winter rain and a hot dry summer.

New arrivals on new sites is the stuff of biological evolution and ecological history, and after each new arrival—plant, bird, animal, microbe, insect, human—an ecosystem is never the same again. The success of managing the desert fringe rested on accepting and managing new species, making the ecosystems of more economic value. It must be emphasised that this is not an argument to avoid responsibility for shaping the future, or against exercising good stewardship, or for carelessly permitting the extinction of species. Future-making is a much more awesome responsibility than just attempting to “hold” what is, or fruitlessly attempting to work back to what is believed to have been. The desire to utilise materials for food, shelter and clothing brought into play the intelligent application of technology, increasing biomass and so carbon sequestration and animal production, and organising the movement and marketing of products. It does not necessarily mean exploitation and depletion.

Vegetation study is much more rewarding if there is a good guide to the species and their identification, and southern Australia was fortunate. Ill with tuberculosis, twenty-one-year-old J.M. Black was shipped from England to South Australia where the dry climate might help his recovery. He worked on the farm of a relative in the mid-north, and was cured. He took a lively interest in the native vegetation of his adopted land and when twenty-five years later he inherited a good deal of money from his sister, Mrs D’Oyley Carte (of Gilbert and Sullivan fame) he lived like a gentleman, but busy with plants. He produced a four-volume Flora of South Australia, full of line drawings, keys to identification, published by the Government Printer and available to students for a few shillings. There was enough similarity of flora to make it useful throughout southern Australia.

(There is an interesting parallel with a significant contribution to wheat breeding. William Farrer, also ill with TB, came to Australia and worked on a small farm near Queanbeyan, and was also cured. He started wheat breeding in his late forties, but his triumph came in his fifties when he bred his rust-resistant Federation wheat which brought huge benefits and stimulated much more. This fitted the Trumble, rather than the Goyder, approach: “We can!”

The early settlers largely avoided the Post-Miocene soils, especially the sands, settling first some of the “older” country, like the red-gum lands around Hamilton in Victoria and Naracoorte in South Australia, and the reddish brown soils of the mid-north of South Australia. These areas mostly carried an attractive parkland of trees over grass, species soon named for the local animals, kangaroo and wallaby grasses. The legumes that had evolved, or arrived on the continent over the years, were generally not copious fixers of nitrogen, so the soils tended to be relatively low in nitrogen and the ecosystems reflected this. Carrying capacity of livestock and yield of crops were often disappointing, but larger holdings compensated.           

Pasture as deliberate management of plants was an untried activity. Under the feet of these settlers was a skimpy, sprawling plant which in due course was to transform their world: sub clover. Its arrival was arguably one of the most important events in the economic history of Australia. It had been accidentally introduced at several points, probably in the early decades of the 1800s, and made its own sense of its environment. The agricultural systems had some features that helped it: the burrs on or near the surface would have got tangled in the wool of sheep, and so were spread. Sheep ate some seeds and some of these survived passage through the digestive tract and were spread in manure. However, the soils lacked some key plant nutrients, so it wasn’t till nutrition was understood and fertilisers provided that sub clover thrived. Understanding the peculiar interaction with the climate enhanced this further. By the mid-1960s, when a great deal of excellent scientific work had been done, Fred Morley, speaking in a symposium, “The Legacy of a Legume”, estimated that it was probably used over 40 million hectares. It has been estimated that around that time it fixed nitrogen to a value of $3–4 billion each year and there is a sound expectation that it will do this for the next thousand years, climate change notwithstanding (there can be a switch to shorter-season varieties).

Subterranean (sub) clover was first described in 1650, then in 1753 the pioneering plant scientist Linnaeus recorded it in his monumental work Species Plantorum, and gave it its name, based on its peculiar habit of burying its seed. It had been recorded in all the Mediterranean countries of Europe and the coastal region of North Africa and east through the Balkan countries, Asia Minor to the Caspian Sea and northern Iraq, even into southern England. Notwithstanding, it was inconspicuous, unproductive, of little economic consequence, a roadside volunteer, occasionally considered a weed in crops. There does not seem to have been recognition of the various ecotypes—of which there must have been several—partly because the differences would have been minimal under those conditions, so difficult to observe. Also, the era of botanical study had not yet arrived.

Sub clover is an annual legume, winter growing and self-pollinating. Being a legume, it harbours nitrogen-fixing bacteria in nodules on its roots, so has access to nitrogen from the atmosphere, exchanging carbohydrate, made by solar-activated photosynthesis, with nitrogenous compounds made by the bacteria. It is prostrate so, at first glance, not a great forage plant, but it captures the abundant solar energy of this desert fringe well. Low to the ground, it is relatively less severely defoliated than some other plants growing with it—so is persistent under grazing. Being self-pollinating, it is very stable genetically. After pollination it becomes geotrophic, which means the flowers turn down rather than up once the seed is set, burying a proportion of the seed in the ground. This burying is both a marvellous way of escaping seed being eaten by animals and a way of planting the seed in the soil ready for germination next season when the rains come.

Sub clover’s importance in Australia turned on several things: awareness of the need for phosphorus and other plant nutrients; the widespread grazing of woolly coated sheep; the response of sub clover to some features of the unique climate of southern Australia; and the development of seed harvesting technology. We now understand how vital phosphorus is in the capture of solar energy through photosynthesis and in respiration, the oxidising of carbohydrates to release the energy. Part of the cycle allows carbon sequestration in the soil as molecular organic matter, a compound of carbon, nitrogen, sulphur and phosphorus. Enough of all must be present: generally phosphorus and sulphur from superphosphate (sulphur from the sulphuric acid used to change rock phosphate into superphosphate), carbon from solar energy and nitrogen from the legumes fixing nitrogen with their co-operative bacteria.

Remarkably, the southern Australian climate allows the variation within the species to be expressed by influencing time intervals in the growth cycle: germination and establishment of the new plant as a leaf producer to flower initiation at the growing point; flower initiation to emergence of flowers; flowering to the maturation of the seed. The total—germination to ripe seed—varied. One early variety got through its life cycle in four months—found at Dwalganup in Western Australia; a late variety took ten months—found at high-rainfall Tallarook, north of Melbourne. Between the extremes are a large number taking various times as they interact with the geography and climate. Thus it is a remarkably important plant economically.

How on earth could one explain so much variation in the one species? Yvonne Aitken graduated in agricultural science at the University of Melbourne in the late 1920s, and became assistant to Professor Sam Wadham—and spent her life there. What she had that few scientists have was both a grasp of the science and dexterity with a needle, meticulously dissecting the hidden growing points of plants. Thus she identified the internal change that caused the growing point to change from leaf-producing to flower-producing. Graduate student Bill Collins, who well understood the broader field situation, got to know her well and applied her knowledge, so understanding the subtle changes of climate and the importance of the first stage. Sub clover, though a long-day plant, also requires some exposure to low temperature during the early growth stage, partly over-riding the need for longer days. The total cold needed, accumulated night after night, was different for each variety.

So the peculiar climate of southern Australia came into play. Southerly air streams blowing across the Southern Ocean are cool rather than bitterly cold (as in Europe), so it takes more nights to accumulate the needed cold. With these extra nights come more days of growing leaves and capturing solar energy: in the northern hemisphere a plant may grow two pairs of leaves, in southern Australia eight to ten, much more herbage—great grazing. The time needed for the full growth cycle—seed germination to ripe seed—must fit into the rainfall growing season. For any given total year’s rainfall, southern Australia has a relatively long growing season. It is also wide east–west, thus has a large area in this latitude, whereas the other southern land masses are narrower and higher.

There must have been a number of introductions into southern Australia, some remaining near the point of entry, others soon carried to far places in fodder or on the coats of animals. The most dramatic story around its recognition involves A.W. (Amos) Howard, near Mount Barker in South Australia. From 1889 on, being a seedsman, so perhaps perceiving a business opportunity, he clamoured for recognition of its value. Howard was a prophet of doom for those who did not maintain soil organic matter, and he stressed the need to “recuperate” cropping lands by using legumes to build up organic matter. He bemoaned the fact that “many teachers of agriculture have shown no interest in a plant of such paramount importance to producing industries”. The Department of Agriculture, not at that stage very pasture-aware, advised that sub clover was not likely to become a troublesome weed. We need to remember this was before use of phosphate made it grow luxuriantly. In 1895 the Victorian Government Botanist noted that it provided good pasture for livestock. Maiden, the Government Botanist in New South Wales, commented on it in 1896: “It is not an introduction which needs to render us uncomfortable.” Though it was established in Western Australia before 1893, the soils there were too poor in vital elements. 

Amos Howard on his farm east of Mount Barker was probably right in the middle of a perfect area, both in soil and climatic terms, for that variety. No doubt sub clover had preceded him, and was widespread though not prominent in the district. Howard and sub clover had lived together for more than ten years before he noticed it. When he did, it is likely that the increasing use of superphosphate on crops from 1884 meant that when he sowed some clover where there had been a crop the year before it flourished because there was a raised level of phosphate. Howard’s variety was named after the district, and this Mount Barker variety brought great prosperity to the area. However, it needed too long a growing season—it was called a late variety—to be very useful in the desert fringe.

But even Howard the seedsman couldn’t work out how to gather the seed and then clean it for sales and sowing. In its ripening period it burrowed and screwed its spiny burrs (really curled-up seed pods) into the soil, a rare feature in plants. One remarkably simple piece of technology was helpful: the sheepskin roller, made very cheaply, though it would only pick up a modest proportion of the burrs present. To thresh the burrs, Howard chaffed them, rubbed them between bricks, winnowed them, rubbed them through sieves—and produced limited amounts of fair quality seed in 1903. With great effort he produced more than 1500 kilograms of seed for sale in 1910.

In 1920—thirty years after Howard first drew attention to it—the South Australian Department of Agriculture recommended sowing it. In their defence, it was one thing to decide that a pasture plant was useful, another to have a supply of good clean seed, an essential pre-requisite to widely recommending it. It was really another ten years before full development of the farming systems to utilise it were worked out. In the early 1920s a local blacksmith and metal worker, Ronald Kaesler of Hahndorf, designed, made and patented a complicated threshing machine specifically for the task. Total yield of seed went from 1500 kilograms in 1921 to 176,300 kilograms in 1929, with an increase in area harvested from four hectares to 700. Dense swards sometimes set as much as a tonne of seed per hectare! Thus there were good financial rewards, and a seed industry was established around Mount Barker with small farmers serviced by major contractors. There was a huge surge in soil improvement.

The seeds were round and hard, not easily chewed by animals, and some even passed entire through the ruminant’s digestive tract, emerging in the faeces. Thus the plant had a remarkable ability to spread to new sites. But that was not all: some seeds had built-in resistance to water uptake—what was called hard-seededness, and even if taking up water, did not germinate. Some dormancy was broken by temperature changes, some even lasted several years. Further, much of the seed is dormant when freshly ripened—so even if there are summer rains it will not germinate, an excellent survival mechanism through to the next winter growing season. All in all, this really meant the possibility of remarkable continuity in pasture. 

Western Australia had some distinctly different varieties from the eastern states, though there was some importation from South Australia. The local ones probably came in through the port of Albany in ships’ cleanings, surviving but not thriving. Though there was some 25 to 30 million hectares of quite good rainfall land, from its establishment in 1829 the town of Perth had struggled through most of the century, hardly feeding itself. So much of the land was Post-Miocene soils or poor lateritic areas, not easy to develop, and other land carried huge forest trees, difficult to clear. In 1891 the gold discovery at Kalgoorlie was a mixed blessing, bring some prosperity but also diverting resources from land development.

Departments of agriculture were being established, the one in Western Australia fully functioning early in the twentieth century. Writing in the very first volume of its journal in 1924 an agricultural adviser, A.B. Adams, gave a useful account of the state of play with sub clover, of which there was possibly as much as 10,000 hectares sown. He described early, mid-season and late-flowering varieties, the suitability of certain soils, the procurement and use of seed, and fertiliser best practice, noting prophetically that “the early kind has possibilities for the drier wheat belt”. It was a classic piece of quality writing for the farming audience.

At last the Perth hinterland got lucky. With superphosphate widely recommended at about 112 kilograms per hectare (one hundredweight per acre) the area of sown sub clover in Western Australia had built up to about 1 million hectares by 1951, but it was to increase nearly ten-fold over the next twenty years and agricultural development and production continued to be a potent economic driver for the state through to the 1980s and 1990s. Then, of course, mining took over—but mining, by definition, is a temporary activity.

The quest for appropriate soil–plant recipes for land development attracted some excellent teams of scientists—the Department of Agriculture, the University of Western Australia, and CSIRO, which had an important field station at Kojonup, where more than sixty varieties were studied. The sandplains of Western Australia were perhaps the last frontier, and attracted a wave of settlers from the eastern states.

Settlers were allotted land, provided they developed it with specified improvements. This sorely tested many new settlers with little cash, but they often made up in energy and commitment what they lacked in capital. A typical settler, Glen Andrew, now living in retirement in Esperance, tells it:

We arrived at our block on 4 March 1955, and moved into the 6 m by 5 m tin shed I had erected on my first visit over summer, then pitched a tent alongside for our three small daughters to live in. The first “developmental” priority was a paddock for a house cow, so with my second-hand Field Marshall tractor and Horwood Bagshaw 14 disc plough I ploughed 10 ha—straight off without any logging or burning of the low heathy vegetation. I back-ploughed it to get a rough seed bed, then in May, with the help of a neighbour (an old hand—he had arrived the year before!) I sowed a cover crop of oats (20 kg/ha) over 4 kg/ha of Mt Barker sub clover seed and a handful of Wimmera ryegrass seed—and of course superphosphate with copper and zinc trace elements.

From then on I put all possible time into “logging” the heath on more land—a six metre log from Salmon Gums (the only reasonable sized trees in the area) with a double railway iron trailing behind on two chains. I had 80 hectares down and ready to be burned in the summer. Some time also went into erecting a 12 m by 9 m shed with living quarters in one half, and storage for seed, fertilisers etc in the other. We also had a bore put down, good water, but only about 5000 litres a day—some help, though.

We had anxiously watched the oats, and were delighted when they made good spring growth, and we cut 4 tonnes of hay. A great day—our first produce, so we bought our first stock—a house cow and calf! I spread about 180 kg of super/ha in the second year (when pastures carried about 2.5 sheep/ha) then cut back to 125 kg/ha. This was basically my development program, with new clovers such as Daliak as they became available. Carrying capacity built up steadily to a steady 7–10 sheep/ha. This was “sure” country in rainfall terms: Over 20 years, the wettest was 800 mm, driest 400 mm (twice) with an average of 520 mm.

With minor variations, this was the story of farmer settlement of the sandplains of Western Australia in the 1950s, driving the economy, greening the Post-Miocene sands. Over the decades since, improvements have continued, for instance, the working of some clay into the surface to enhance water holding capacity. 

In south-east South Australia, the sands were known as the Ninety Mile Desert. They had been disappointing for settlement, despite their reliable rainfall grading from more than 500 millimetres in the south to about 400 to the north-east, and also their location with good services: both the main Bordertown–Adelaide road and railway lines passed through. Like the West Australian sands, in their natural state they had a near-zero livestock carrying capacity. Some of the sands were very deep—five, even ten, metres of sand, pure silica.

By the 1940s researchers in South Australia well understood the need for phosphate and molybdenum, but what else? There was in South Australia a view, not sustained under critical analysis, that the deep sands were “droughty” compared with the other soils of the area. In fact, because there was no run-off, provided pasture plants like lucerne were used that explored the full depth of moisture penetration, there was more moisture than in other soils. (The exception was some sandy areas that, rather paradoxically, had biological activity that produced water-repellent compounds—non-wettable sands.) The sands were very acid, pH 4 to 5, so plots were laid out, testing the use of lime to counter the effect of acidity on the legumes and their root nodule bacteria. Seed of lucerne and sub clover was inoculated with nodule bacteria and sown with lime:super (including a trace of molybdenum). The results were puzzling: without lime there were a few struggling plants: with lime, nothing! It defied logic. The research officer Newton Tiver joined the navy and used to lie on his bunk at sea and ponder “the riddle of the sands”.

Eventually, solving the riddle followed a not uncommon path: one piece of research needed to be illuminated by another. In the early 1940s the CSIRO Division of Soils discovered the need for copper and zinc applications on certain soils, especially those that were alkaline, as any copper and zinc present was made less available by high pH. The addition of lime to acid soils made what little copper and zinc was present unavailable—so no growth. By the time of the surge of postwar settlement these trace elements tended to be included as a matter of course in the sowing of pasture over a wide range of soils: so with plot work on these sands. The result was that sowing with lime:super gave a huge improvement in the pasture establishment and growth, and another million hectares could be developed. Understanding of nutrition, and the various cultivars of the wonder plant, sub clover, underpinned the land rush of the 1950s and 1960s. The surge in wool prices nicely meshed in with good scientific research and enabled expansion onto these sands. Other CSIRO work determined the need for cobalt for sheep health, especially on the calcareous areas near the coast—it could be supplied by a cobalt “bullet” lodged in the stomach with slow release into the bloodstream.

Taxation rules also helped as costs of development as well as capital gains on the sale of the property were tax-exempt, able to be charged against high professional incomes—this was established by a court case brought by the Director of the Adelaide College of Music! One notable investment was that of the AMP Society insurance company, in a sense emulating the Soldier Settlement Schemes, and developing substantial areas in South Australia and Victoria.

So the Ninety Mile Desert was conquered and became known as the Coonalpyn Downs. Forage was greener than ever before—in every sense: greener in colour because there was much more nitrogen in the system than there was when European settlement occurred; greener, too, in the conservation sense—the soils under most pastures are often higher in organic matter than ever in history, more resistant to erosion, and, in terms of the needs of modern society, more productive than ever.

On these Post-Miocene soils high productivity was often achieved through the total replacement of vegetation, for example, on heath lands where the vegetation had little nutritive value even for the native fauna such as kangaroos. In some other areas improvement occurred as an overlay of existing grasslands. Even on much of the older landscapes, for instance the meadow podsols, the increased capacity was at least four-fold. Some new areas on the basalt plains in Victoria were distant enough from the coast to count as part of the desert fringe. Possibly as “young” as 20,000 years, they were covered in kangaroo and wallaby grasses, presumably locally evolved species (though the close kinship of wallaby grass to the oat plant of western Asia raises some interesting evolutionary questions). Eucalypts, which by and large have poor dispersal mechanisms—seed cups that are not fodder for animals and have no hook mechanisms—had not yet penetrated far into these young grasslands. It is instructive that a relatively “foreign” eucalypt from western South Australia, sugar gum, proved so suitable once planted. 

Wheat growing by innovation had been established in the desert fringe. Quite apart from the lower rainfall, the situation was different from “home”—the people were mostly from the British Isles—needing major and minor inventions. Even in ordinary life there were no trades-people on call, so they had to become jacks of all trades, fashioning wood and metal. Water for animals in summer was more important than winter survival—which demanded shedding in Europe. Thus came the corrugated iron tank, copied in Argentina as the tanque australiensis. The native vegetation was too scanty for hedge-rowing: so to contain stock and exclude wildlife, fences were erected. When it came, fencing wire was a boon for those who could afford it. Perhaps most important were “Mallee” gates of wire and thin wooden stakes, pulled tight with a wooden lever and latched with a wire ring. Shearing sheds and yards allowed for much more innovation, and amazing feats of workmanship gave large tallies of shorn sheep.

The colony in Australia had struggled to feed its people from the outset, so there was an immediate need for grain to feed the people by expanding cropping. This involved the desert fringe, especially in South Australia where the gulfs allowed the small ships of the time to draw near to the wheat-growing areas, loaded at first using shallow draft barges, later long jetties (at Port Germein, nearly two kilometres) connecting the horse-drawn wagons to the ships. From 1843 cropping was given impetus by the invention of the Ridley stripper, the building of a flour mill in Adelaide and the steady extension of railway lines. The abolition of corn duties in Britain gave some hope of export. However, land development was tedious. A log (hauled from forest areas near the coast) was used to roll the mallee scrub, which was then burned and the first crop planted. The burn of that stubble killed many of the mallee shoots, but the stumps remained. In 1876 there was a huge leap forward when Robert Smith invented the stump-jump plough (which enabled clearing the land and cropping to overlap). Over the years the stumps were pulled up and became a good source of firewood income. It is notable that the inventions were of equipment for southern Australian conditions—not an attempt to copy European systems, as asserted by Tim Flannery.

The cropping system was sometimes continuous cropping with wheat, more often wheat–fallow, sometimes wheat–fallow–oats because of the need to feed horses. The fallow enabled a carry-over of moisture from one winter season to the next, increasing yield, but hindsight revealed that releasing nitrogen caused breakdown of soil organic matter, clearly an unsustainable practice. Further, the natural level of phosphorus was being run down, though this was reversed near the end of the nineteenth century when superphosphate became known and applied on many farms. This arrested yield declines on soils with somewhat higher natural stores of organic matter. From about 1900 there was some withdrawal of farming from the outer edge of the fringe, but elsewhere, gains in yield from the use of super. The awareness of the value of sub clover (and medics for alkaline soils) to raise nitrogen levels was not yet high.

One marvellous piece of difficult-to-find literature is Yield Trends in the Wheat Belt of South Australia during 1896–1941, by Dr E.A. Cornish of the Mathematics Division of the CSIRO. It is a shock to see the yields over much of this fringe: a separation of profitable and unprofitable areas was drawn in 1934 at a mean yield over the previous ten years of six bushels per acre (400 kilograms per hectare). There was, however, widespread concern at the decline in yields and the severe erosion of much of the land, and the area sown to wheat steadily contracted, by 1941 losing all of the gains since 1920.

Then came awareness of the value of legumes—especially for clover. For the alkaline soils—ones high in the Post-Miocene lime (even more important than soil pH), it was the Medicago species commonly known as barrel and burr medic. Like sub clover, these species had apparently entered Australia in fodder, and had also only become prominent when higher soil phosphorus levels were achieved. Barrel medic was preferred as it had a less obnoxious seed pod—the burr sometimes caused penalties in wool clips. Commercialisation began at the end of the 1930s, with a variety from Cyprus dominating in Western Australia. Victoria also used specific varieties, which gave remarkably good vegetative growth and enabled strong pasture development to fit into rotations, with Mallee farmers receiving good incomes from wool and lambs. Over the full range of these soils a good deal of carbon was sequestrated in soil organic matter.

Thus farming much of the desert fringe became based on rotations that included alternating periods of cropping with grazed pasture leys. The leys had sub clover and medics, which raised soil nitrogen levels, increased soil organic matter and resistance to soil erosion, and were a good base for highly nutritious hay and silage made with increasingly sophisticated equipment. While phosphorus could be added with the crop-sowing equipment, for pasture, even if littered with stumps and logs and boulders, an extraordinarily clever spinner made application to new land easy. It began with a wrecked Model T Ford—if an engine turning a tail-shaft and diff could push the wheels along, then pulling the vehicle along the ground would spin the tail-shaft. Tip it up and add a plate, drop on fertiliser and the machine enabled seed and fertiliser to be thrown—spread—over rough semi-cleared country. Various models of super-spinner were developed and there was an immediate export market.

At times the main benefit of the legume was more or less indirect through better crops. At other times the price of animal products meant the animals gave a significant surge to income—as in the wool boom of the early 1950s, when high prices for wool sent a surge of capital through this desert fringe, enabling an improvement of farm infrastructure and a raising of nutrient levels that was to be felt for decades. Currently the high price of lambs is causing some increase in stocking levels. At other times the profitability of the animal units has been so low as to see animals almost excluded from the system, with legume crops used in the rotation to raise nitrogen levels.           

Technical and scientific people were early recognised as vital to the development of this fringe. The agricultural colleges in the 1880s were valuable in developing adaptive technologies and maintaining the knowledge base but leaders such as Peter Waite, Chair of Elder Smiths, recognised the need for more than new technologies—science was essential to pushing back the frontiers of knowledge, so university faculties were established. The continuing research from them, departments of agriculture and the CSIRO has steadily, cumulatively, edged efficiency upward over the decades, lowering energy input, making better use of the limited soil moisture, right to the present-day zero-till so ably championed by “No Till” Bill Crabtree of Western Australia. There were a number of research stations near the arid boundary, testing varieties, improving farm practices, sometimes staffed by people who spent decades on site. For instance, John Griffiths, recruited from England as a new agricultural science graduate, spent his whole career dedicated to greening the dry edge of the fringe in the Victorian Mallee. Mr Mallee! A major contribution has also been made by what could be called private research—as by the Birchip Cropping Group, formed by local people like Warwick and Ian McClelland on the family farm after studying agricultural science at the University of Melbourne.

In Cornish’s report covering up to 1940, yields per unit of winter season rainfall were mostly in the range 550 to 700 kilograms per hectare per 100 millimetres of winter rain. From about 1970 to 2000 the yield per unit of winter rain almost trebled. Even on the drier edge it was about 1 tonne per 100 millimetres of winter rain.

The plant biodiversity of Australia was increased as some other uninvited guests arrived in the hay and ships’ cleanings and flourished with the raised fertility levels of the desert fringe. Barley grass, also from Europe, has great seedling vigour and winter growth and a superb seed distribution system—an awned seed which clings to wool and human clothing. Capeweed, a rosette member of the daisy family, travelled from the Cape of Good Hope, and relied on drought resistance and prolific seed setting from a fluffy head which enables good establishment on bare ground such as gateways. Annual rye grass, with a similar growth form to cereals, was something of a Trojan horse in crops, building up its seed burden in the soil through the crop cycle, curbed by clever selective herbicides, then hitting back with herbicide resistance. Paterson’s Curse (Salvation Jane) arrived from the western Mediterranean, possibly the Azores and Channel islands, to adorn the hills and provide honey in the moister parts of the fringe. Onion grass, a rare perennial, spread quietly and still obstinately holds a place in poorly farmed land and public places. And there were others, each with its subtly different factors affecting flowering time, seed setting, mechanisms for seed dispersal and survival of seed over summer, longevity, and re-establishment conditions. Thus there were large year-to-year fluctuations in botanical composition, resulting in variations in nutritive value and digestibility.

Most of the plants involved in both pasture and crop developments have been annuals, despite the original occupants being perennials. Lucerne, with its remarkable root system, is the most notable introduced perennial, at times contributing strongly. Some people argue that perennial crops should be used: sow it, just get out the harvester every summer and harvest the grain. A little critical analysis suggests that, especially in these environments with a sharply defined moist, cool winter season and a very dry and hot summer season, annuals are superior for grain and seed production. A plant which simply produces a large number of seeds, then dies off, can be very successful. Add on effective means of dispersal, so that many of the new plants will be at a distance or in different niches, and the annual becomes very powerful indeed. Tim Flannery describes them as having “weedy” characteristics, but this is unfair. Rather, this is such a successful survival strategy that such plants can thrive as a continuing element in pastures and successfully invade new situations.

The most powerful argument for annual, rather than perennial, crops is the proportion of the substance of the plant that is transferred to the seed heads and can be harvested—called the Harvest Index. Annuals are often in the order of 0.45 to 0.5, perennials rarely above 0.3, as the old plant must store nutrients to survive the adverse period. Couple this with the flexibility of the farmer to change the crop: from wheat to canola or barley to wheat or to include a new wheat variety with disease resistance or some special marketing attribute and the case for annuals becomes compelling. Finally, the move to zero-till systems, in which the new crop is sown with fertiliser as the first and only tillage pass (and that only a scratch) means that the operations for growing the annual are little different from the maintenance for a perennial: both need weed and pest control, addition of fertiliser, and leaving all residues on the surface to be incorporated as molecular soil carbon. It should be noted that if the plant is grown for pasture, with the animal consuming stem and leaf rather than the seed head, the perennial becomes much more desirable—plants such as lucerne have a place in such environments as the desert fringe. Even though the response to big summer rains is occasionally spectacular, a planned fodder conservation reserve is necessary.

Some people make claims that there can be a huge increase in carbon storage in the soil of the farming areas of Australia—of which the fringe is a substantial part. In fact carbon sequestration has been a significant unintended consequence of this greening, as the object of stabilising the soils to almost eliminate erosion through more or less continuous soil cover and higher soil organic matter has had this effect. A healthy soil has an abundance of life, much of it simply living invisibly in the soil and cycling residues. As pointed out earlier, the core of the cycle must be molecular organic matter, a compound of carbon, nitrogen, sulphur and phosphorus, limited in the natural state of the soils by levels of nitrogen and phosphorus and possibly sulphur. Thus the effect of adding phosphorus and sulphur in superphosphate, stimulating legume growth so raising the nitrogen level and producing copious amounts of carbohydrates was to raise organic matter levels and thus carbon storage. This level comes into an equilibrium with soil moisture and temperature and the agricultural activity. Many farms are at or near this equilibrium, and those that are not are farmed by people who are difficult to persuade. 

Drought is a recurring theme in talk of this fringe. Rainfall does fluctuate—and droughts are notable events, but we get (less newsworthy!) good years. It is important to recognise that drought is not aridity: rather, it is periodic failure of nature to deliver normal rain for that area—and “normal” is broadly defined. The science of farming enables a good yield of crop from quite a variety of rainfall patterns, as soil moisture storage is improved, for instance by zero-till. Then the management must be underpinned by technologies that enable minimal damage: fodder storage for livestock and efficient feeding systems, transport of stock to areas not affected. In reality most rain-fed farming is risk farming and there must be financial resilience, especially at the arid boundary. The trick is to knowingly manage the risks, underpinning all with precise measurements and modelling. Australian scientists have developed strategies for the desert fringe and the special events, an approach well ahead of other fringe farmers, such as in the countries of Saharan Africa. Greening is more than a biological term, it has a human behaviour component.

Is the climate changing? If trends are detected, there will be gradual adaptation to changes, including recognition of shifts in the arid boundary. All efforts must be underpinned by the same attitudes to complex systems that have enabled substantial production over recent decades, carefully incorporating new technologies such as genetic manipulation as they become available.

We can farm the desert fringe forever provided we continue to have adequate research and development, apply new knowledge sensibly, have the right attitude to innovation and change, have a proper sense of time and history, are humble enough—and do not ever claim to have arrived. We is an inclusive term: scientists, farmers, urban people, all of us.

Dr David F. Smith has lived and worked right across southern Australia, from studies in native plant ecology to land development and research supervision. The author of Natural Gain: In the Grazing Lands of Southern Australia (2000) he is with the Department of Agriculture and Food Systems at the University of Melbourne.