Crop adaptation in Australasia

Printer-friendly version

Lloyd Evans

CSIRO, Division of Plant Industry, Canberra


A practical consequence of early flowering. Selection of Western Australian wheat varieties of higher yield over the past century has been highly successful, and associated with a reduction in thermal time (in degree-days, °C d) to the first stage of the flowering process. (Redrawn from Richards 1991)


When European settlement came to Australia late in the eighteenth century, it was the only continent without any agriculture and lacking indigenous crop plants. The Aboriginal hunters and gatherers knew well which plants could be gathered for food, fibre or medicine, and when and where. They also practised some agronomic management such as broadcasting millet seed, replanting yam tubers or burning off yam or macrozamia tops. However, when the first fleet arrived in early 1788, they brought a variety of plants, fruits and seeds either from Britain or acquired in Rio or Capetown along the way. A few acres of barley were sown near the site of the Botanical Gardens soon after arrival but they gave a pitiful crop barely returning the amount of seed sown. The next attempts to grow wheat, barley and other cereals were shifted to better soil at Parramatta. Even there, returns were poor. Governor Phillip saw the problem as lack of agricultural expertise among the convicts: ‘If fifty farmers were sent out … they would do more in one year in rendering this colony independent of the mother country … than a thousand convicts’. However, the low yields persisted even after the farmers arrived in 1793, and a botanist in 1800. The real need was adaptation by introduced crops to Australian environments.

In New Zealand the initial problems were less severe, partly because the Maori had already introduced and cultivated several crops and because the soils were more fertile and the climate more suited, at least in some areas, to European crops and pastures. The Canterbury Plains were more readily converted to agriculture, and British grasses such as cocksfoot (Dactylis glomerata), timothy (Phleum pratense), perennial ryegrass (Lolium perenne) as well as white and red clovers (Trifolium repens and T. pratense) flourished widely. Their adaptation has of course been improved since then by regional selection, but some English cereal varieties were still in use in the 1930s.

Plant introduction

Not all domesticated plants from Europe required adaptation to Australian conditions. Blackberry (Rubus fruticosus) spread quickly in Victoria (with some help from Baron von Mueller) while Lantana camara (possibly introduced by Phillip) soon became a nuisance in warmer areas. Many serious weeds were introduced unwittingly such as Bathurst and Noogoora burrs (Xanthium spinosum and X. pungens) as were several useful grasses, clovers and Townsville stylo (Stylosanthes humilis). Given the ready adaptation of these and many other introduced plants to Australian conditions, why were there problems with many of the temperate cereals and seed legumes?

Crop adaptation

By and large, temperate cereals and seed legumes grew well enough. One farmer wrote of his crop: ‘On the 21st of October (1803) a more beautiful appearance of a successful harvest never flattered the expectations of a farmer …’ But in the end the crop was not worth reaping. In this case the cause was rust, but as often it was water stress. Even when autumn sown, the predominantly British varieties tended to flower too late, with the result that seed development occurred in the hottest and driest months. Governor Phillip’s first sowings of wheat in May 1789 were harvested in December, and his barley sown in August was harvested in January. Had Australia been colonised by a Mediterranean country, with Mediterranean cereals and legumes, the problems of crop adaptation would have been less acute.

Nevertheless, early Australian agriculturists persisted with English varieties of wheat that normally ripened in late July, and coinciding with harvest festival in Britain; varieties which, as Farrer said, ‘yielded more of disappointment than of profit. Too late in ripening for our climate, their ears are blasted and their grain pinched by the hot winds of our summers.’

From the 1860s to the 1890s the most widely grown wheat variety was Purple Straw. Previously thought to be a selection from an English variety, Wrigley and Rathjen have now shown it to have been selected from a Tuscan wheat sent from Italy to Scotland and thence to a South Australian farmer. Similarly, the varieties which replaced Purple Straw in the 1890s derived from farmers’ selections in South Australia from non-English varieties. Ward selected rust-free plants from a South African variety to give Ward’s Prolific, from which another grower selected the earlier-flowering Gluyas Early. At about the same time Steinweidel selected tall, early-maturing off-types of an American variety to establish the wheat bearing his name.

Cereal breeding

Although A.B. Robin of South Australia began wheat hybridisation in Australia, it was the purposeful and comprehensive crossing program begun in 1889 by William Farrer of Lambrigg, now almost enclosed by the expansion of Canberra, that transformed crop adaptation in Australia.

The two crucial elements in Farrer’s success were his explicit formulation of his goals as a plant breeder, and the international scope of the wheat varieties used in his breeding program. The clarity and comprehensiveness of his objectives were sharpened in a scathing series of exchanges he had in 1882 with two newspapers, The Queenslander and The Australasian. By the time he discussed them at the Australasian Association for the Advancement of Science in 1898, they had become a well-honed list which included not only resistance to the various rust diseases and high baking quality but also such physiological features as earlier ripening, shorter, stronger straw, upright narrow leaves, reduced tillering and better grain retention. Farrer was also the first to draw on the wheat gene pools of the world to attain his aims. As the great Russian plant geographer Nicolai Vavilov acknowledged in referring to Farrer’s breeding program in 1935: ‘There is probably no region where intraspecific and interspecific hybridization of wheat has been so extensive as in Australia’.

Well before Mendel’s paper was rediscovered, Farrer had stated that he selected the combinations of characteristics he sought from ‘the variable generation’ (i.e. the first selfed generation from his initial crosses), particularly for ‘that delicate and obscure but most important quality … which we include in the inconvenient term of constitutional fitness for the locality’. Early ripening was a crucial component of such fitness, which Farrer derived from Indian varieties, and which helped his wheats to escape rust. Farrer suggested in 1898 that earlier ripening was also drought-escaping and ‘might be the means of extending some of our cultural industries, and certainly that of wheat-growing, appreciably further inland’, and that much could ‘be done by means of artificial selection to expedite the work of acclimatization’.

Indeed, wheat breeding since Farrer’s time has continued many of the trends he initiated. Progressively earlier flowering in more recent wheat varieties continues to be apparent in several states as illustrated for Western Australia  (opening figure to this section; see Figure 5 in Richards 1991). In trials at any one site there is often a sharp optimum ‘time to ear emergence’ in relation to yield, which has become shorter as plant breeding and crop management has improved. This optimum time is strongly influenced by seasonal incidence of drought and heat in each locality, but also by other factors such as frost injury. To date, much genetic manipulation has been empirical, involving the inherent earliness of varieties and their responses to vernalisation by low temperatures and to seasonal changes in daylength. These are ‘the delicate and obscure’ adaptive processes to which Farrer referred, and there is still much to be learned about their compensating interactions in the determination of yield (Section 6.4). Moreover, as economic and agronomic conditions change, earlier trends in adaptation may even be reversed. For example, in breeding feed wheats for higher rainfall zones, emphasis is being placed not only on red grains but also on later-flowering winter varieties with an English background of the kind rejected by Farrer. Crop adaptation is a continually shifting process.

Temperate pastures

This account of the early adaptation of wheat to the more Mediterranean and lower rainfall environments of southern Australia is illustrative of the adaptation of other crops and pasture plants in these regions. With a pasture grass such as Phalaris tuberosa, which probably entered Australia via the Toowoomba Botanical Gardens in 1884, and whose value was recognised after its escape from there, it was more than 60 years before expeditions to the Mediterranean region deliberately collected seeds over a range of altitudes and environments for the purpose of enlarging the gene pool for a breeding program for improved adaptation.

With subterranean clover (Trifolium subterraneum), on the other hand, there were probably many unwitting introductions from the Mediterranean region in grass seed samples. This is implied by the great range of growth characteristics and flowering times found among the many locally adapted varieties. Such variation was noted in subterranean clover after its value as a pasture plant and soil-fertility restorer was recognised by A.W. Howard of Mount Barker. In warmer areas with short growing seasons, only very rapidly flowering varieties could set seed and survive. At sites with longer growing seasons, the earlier-flowering varieties were outcompeted by later-flowering ones which grew larger and set more seed. In general, therefore, later flowering in a local variety was an adaptation to a longer growing season. As with wheat, early flowering in varieties of subterranean clover such as Dwalganup is associated with inherent earliness (evident in an absence of chilling requirement and relative indifference to daylength). Late-flowering varieties such as Tallarook have a requirement for vernalisation and/or long days. In this instance, natural selection had ensured a close adaptation of genotype to habitat which gave rise to the impressive range of local varieties in southern Australia.


Before we turn our attention northwards, two other examples of southern adaptation should be considered. Attempts to grow rice go back to at least 1869 in Queensland but yields remained low even after irrigation was introduced, particularly because of severe weed problems. In the irrigated areas of New South Wales beginning around 1922 it was a different story. Short-grain varieties from California such as Caloro, and later their longer grain varieties such as Calrose, followed by varieties bred from them in Australia, proved well adapted and plantings expanded rapidly. In fact, the varieties and conditions — especially the clear, sunny days and cool nights — were so well matched that for many years Australia had both the highest national average rice yields and the world record crop yield for rice.


The other example of southern adaptation comes from Western Australia where a naturalised species of lupin introduced as a forage plant from overseas (Lupinus cosentinii or digitatus) was domesticated by John Gladstones who selected lines with reduced pod shattering and earlier flowering. In this case the crop was well adapted to the local environment. Further selection will be required to extend the range of lupin crops to other regions of Australia and overseas.

Sandplain lupin is not endemic to Australia, whereas wild relatives of rice, sorghum, cotton, soybean, tobacco, yams and other warmer climate crops are. Wild relatives pose problems for pest and disease control in crop species, but also offer a possibility of closer adaptation to Australian conditions if not to others. For example, Kanaka cane cutters partly domesticated a wild Australian yam, Dioscorea transversa, while in Queensland, in an effort to replace their own noble yams. When taken with them back to Melanesia on their return, the Australian species failed to form tubers and was rejected because it ‘does not know the seasons’.

Sugar cane

Knowing the seasons, through their responses to daylength, is an important component of crop adaptation at low latitudes. Although sugar cane was brought to Australia by the First Fleet, it did not become established until reintroduced from Tahiti about 30 years later. However, adaptation to North Queensland conditions began in earnest at the end of the nineteenth century when collecting expeditions to New Guinea brought back several Saccharum officinarum clones which were used by the Queensland Acclimatisation Society, the Colonial Sugar Refining Company and, later, the Bureau of Sugar Experiment Stations in selection and breeding programs. Many locally adapted varieties were produced, more than 50 of which were being used in 1971, although a small number of widely adaptable varieties accounted for half of the sugar production.


There are nine wild species of cotton (genus Gossypium) endemic to Australia, but this crop was not grown on a significant scale until the American Civil War, and became a major crop only when extensive irrigation areas became available. In the Murrumbidgee Irrigation Area the growing season proved to be too short and too cool for profitable production, and most Australian cotton is now grown in the Namoi Valley. Although American varieties have yielded well in that area, better-adapted Australian-bred varieties are now being grown to a greater extent.


Soybean is another warmer zone summer-growing crop with many wild relatives in Australia, but was of only minor significance in Australia until the 1970s. As with other irrigated crops, soybean varieties bred elsewhere can be highly productive when their responses to daylength and growing season are appropriate. Availability of varieties adapted to the southern areas of the USA was a key factor in the expansion of the soybean crop in Australia. Even-later-maturing varieties are being bred for North Queensland conditions where local adaptation calls for greater sensitivity to daylength.

Tropical pastures

Just as different crops have been adapted to northern Australia, so different pasture plants were also required which were beyond the adaptive range of British and Mediterranean grasses, clovers and medics. Indeed, the search for grasses and forage legumes suitable for cattle grazing in northern Australia was an unprecedented adventure in acclimatisation.

An unwitting introduction of Townsville stylo (Stylosanthes humilis) in 1904 showed the way. Stylo productivity in semi-arid conditions induced N. Pollock (Queensland Department of Agriculture) to throw seed out of train windows wherever he travelled in the state. However, stylo competed poorly in many parts of Queensland, as did several other forage legumes adapted to wetter conditions, and in 1947 an expedition to four South American countries was organised by CSIRO to collect grasses and legumes likely to adapt to northern Australian conditions. So well adapted did several of the legumes brought back prove to be that the carrying capacity of Queensland pastures was transformed by their introduction.

Understanding crop adaptation

Although Governor Phillip, in 1789, had not recognised the need for adaptation of crop and pasture plants to local conditions, various state acclimatisation societies (beginning in Victoria in 1861 and in New Zealand in 1864) sought to spread knowledge of acclimatisation and to understand the causes of the success or failure of introduced plants. So soon after the publication of Darwin’s Origin of Species in 1859, they were often hampered by a belief in the fixity of species and in their perfect adaptation, as well as by ignorance of the crucial role of such processes as vernalisation and photoperiodism. Yet just as Farrer was an effective plant breeder before Mendel was rediscovered, so also could he select for earlier maturity without knowledge of the effects of daylength on flowering. Nevertheless, such knowledge has greatly increased our power to screen plant introductions or breeding materials for particular environments, together with our continually improving knowledge of local soils and climates. Yet there is still a great deal to be learned about interactions between genes and environments, about physiological states such as inherent earliness of flowering, about factors determining local adaptation and, as importantly, about why some genotypes have much wider adaptability than others. Plant adaptation is still a relatively dark continent on physiological maps.

Further reading

Burt, R.L. and Williams, W.T. (1988). ‘Plant introduction in Australia’, in Australian Science in the Making, ed. R.W. Home, 252–276, Cambridge University Press: Cambridge.

Evans, L.T. (1980). ‘Response to challenge: William Farrer and the making of wheats’, Journal of the Australian Institute of Agricultural Science, 46, 3–13.

Evans, L.T. (1993). Crop Evolution, Adaptation and Yield, Cambridge University Press: Cambridge.

Farrer, W. (1898). ‘The making and improvement of wheats for Australian conditions’, Agricultural Gazette NSW, 9, 131–168, 241–250.

Lovett, J.V. and Lazenby, A. (eds) (1979). Australian Field Crops,
Vol. 2, Angus and Robertson: Sydney.

Wrigley, C.W. and Rathjen, A. (1981). ‘Wheat breeding in Australia’, in Plants and Man in Australia, eds D.J. Carr and S.G.M. Carr, 96–135, Academic Press: Sydney.