FEATURE ESSAY 16.1  A brief history of plant nutrition

Printer-friendly version

Jack F. Loneragan


Figure 1 Emeritus Professor Jack F. Lonerragan, AM FTS FAIAS, with Dr R.W. Bell and Ms Rojare Netsangtip. (Photograph courtesy Brian Richards, Murdoch University) 

Four hundred years ago, the Australian continent and the nature of plant nutrition were unknown in Europe and subjects only of conjecture. That was about to change, with initial steps of both discoveries beginning around the same time and place — the lowlands of Europe at the beginning of the seventeenth century.

1600–1700: beginnings

In the early years of the seventeenth century Dutch merchantmen and explorers encountered, often accidently and sometimes disastrously, parts of the Australian coast. ‘Terra Australis’ then assumed reality for Europeans. At that time, Aristotle’s conclusions that all earthly matter was made from four basic elements, earth, fire, air or water, still held sway among scientists. Most considered water to be the principal or sole nutrient for plants.

A Flemish physician, Van Helmont, broke with tradition by testing the nature of plant nutrition with a quantitative experiment. He grew willow shoots in soil in covered containers, supplying only water. At the end of five years, he recovered everything except 56 g of the 90 kg of original soil and measured an increase of over 70 kg in the dry matter of the willow shoots. As the loss of soil was negligible, he concluded that the increase in willow dry matter arose from water alone, thus confirming existing beliefs. It was a well-executed experiment and, in the context of available knowledge, his conclusions published posthumously in 1652 appeared sound. They were accepted by contemporary scientists, including the eminent chemist Boyle who repeated and confirmed the experiment.

But, as so often happens in science, preconceived ideas about a process under investigation, combined with ignorance of other major inputs (photosynthesis), blinded investigators to a small but vital discrepancy in their data — a loss of 56 g in 90 000 g of soil. While his conclusions were mostly wrong, Van Helmont’s report stimulated others to follow his experimental approach which eventually resolved the problem. In 1699, Woodward published the results of experiments in which he grew spearmint in water from various sources — rain, River Thames, Hyde Park conduit and Hyde Park conduit plus garden mould. He found that plant growth increased dramatically with the level of impurity in the water and concluded that ‘vegetables are not formed of water but of a certain peculiar terrestrial matter’.

Woodward had conducted a brilliant experiment and focused attention on the role of soil in plant nutrition. How-ever, his conclusions again suffered from prevailing orthodoxy and ignorance of photosynthesis. Discovery of that process was itself delayed nearly 100 years until knowledge and techniques in chemistry had progressed sufficiently for its study. Publications by Priestly in 1779, Ingen-Housz in 1779 and Senebier in 1782 finally resolved the role of photo-synthesis for incorporating CO2 and water into plant matter.

So, by the time that the First Fleet landed in Botany Bay in 1788, Van Helmont and Woodward had both been proved partly right and the basic elements of plant nutrition were in place: (1) plant matter was formed largely of CO2 from air, plus some contribution from water, and (2) plant growth depended also upon some ‘peculiar terrestrial matter’.

This information provided farmers of New South Wales with little help for crop production on the nutrient-deficient soils of their young colony, but it did set the stage for investigating the nature of Woodward’s ‘peculiar terrestrial matter’ — investigations which would be of great value to the colonists’ descendants 100 years later.

1800–1900: macronutrients

Researchers such as de Saussure now sought to identify the effective components of Woodward’s terrestrial matter by analysing plant materials. De Saussure found a wide variability in the composition of ash from plants grown on different soils and concluded that some elements which were universally present were essential while others were not. Further progress was defeated by the large number of chemical elements present, variability in the composition of ash and trace amounts of many elements now known to be essential.

Conclusions from plant analysis were complemented by observations on responses of plants to soil additives. An eminent German chemist, Justus von Liebig, presented his view in 1840. He recognised that plants need C, H and O supplied by air and water. He also recognised that plants need P, K and N and suggested, correctly, that soil minerals supplied the P and K. He also proposed that plants grew in direct proportion to their supply of nutrients and that a deficiency of any one prevented growth (codified as his ‘Law of the Minimum’).

Unfortunately, a phosphatic fertiliser which Liebig invented failed to work. In order to prevent his fertiliser leaching, Liebig rendered the phosphate insoluble. That failure, in addition to errors and controversies over some of his claims, led to delay in acceptance of those of Liebig’s claims that were sound. Moreover a number of serious problems remained unsolved. Their resolution took another 50 years and had to await advances in soil chemistry, plant physiol-ogy and soil microbiology.

Meantime, agricultural production in New South Wales and other new colonies languished due to P, N and other deficiencies. In Western Australia, severe nutrient deficiences on the sandy coastal plains south of Perth contributed to the collapse of the notorious Peel settlement in 1829. According to Governor Stirling, ‘the nature of the country in respect of its general inferiority of soil’ contributed to settlers becoming widely dispersed as they tried to select better soils.

Soil chemistry made a major advance when Way showed in 1850 that soluble salts of ammonium, phosphate and potassium reacted with and were held by soil particles. At that same time, Lawes made a soluble phosphatic fertiliser by treating insoluble rock phosphate with sulphuric acid, thus inventing superphosphate, which unlike Liebig’s insoluble product proved highly effective.

Around 1860 German plant physiologists Sachs and Knop independently adopted Woodward’s solution culture technique to define active components of his ‘peculiar terrestrial matter’. By adding pure salts to water, they established that six chemical elements we now know as macronutrients (N, P, K, Ca, Mg, S) and Fe were all essential for plant growth. They also showed that plants grew quite well with nitrate as their sole source of N, overturning Liebig’s claim for ammonium as the sole source of N.

The behaviour of N fertilisers in soils remained puzzling with ammonium salts being converted into nitrates and some experimenters failing to obtain responses. A solution to these problems had to wait until Pasteur had recognised the chemical activities of bacteria and had developed techniques for studying them. Bacteria were then shown to convert fertiliser ammonium into nitrate and in 1886 Helriegel and Wilfarth established the role of nodule bacteria in N fixation by legumes.

Hence, by 1900 all outstanding problems of plant nutrition appeared to have been solved, and this research was impacting on agricultural production in Australia. After experiments at Roseworthy College in South Australia in 1882 had shown remarkable responses of wheat to superphosphate, progressive farmers began importing P fertilisers with good results and production of superphosphate started locally. Two hundred years of plant nutritional research was about to benefit the Australian colonies handsomely as they moved into Federation.

1900–present: micronutrients

Early years of the new century saw an increasing use of superphosphate for wheat production. Although Richardson of the Victorian Department of Agriculture had obtained big responses of clover to P in 1907, it was not until the mid-twenties that its value to pasture production, and via increased soil N to subsequent wheat production, was generally recognised. With alleviation of ubiquitous P and N deficiencies, farmers were now able to obtain good yields of wheat on most soils receiving adequate rain.

Nevertheless, on some soils wheat remained unproductive with symptoms similar to those of infectious diseases. Micro-biologists had observed similar symptoms on cereals in other countries and were convinced that disease organisms were responsible, though puzzled and frustrated by their continuing failure to identify a causative organism.

Meanwhile, in apparently unrelated research, laboratory scientists in Europe and the USA were reporting that tiny amounts of various elements increased the growth of fungi and higher plants. In 1869 the French microbiologist Raulin reported that Zn increased the growth of Aspergillus niger. In 1897 Bertrand (another French microbiologist) observed that Mn was associated with oxidising enzyme laccase, and concluded that Mn was essential for plants. His finding was confirmed in 1914 for maize by Maze in France and for wheat by McHargue in the USA. In 1910 Agulhon reported field responses of several crops to B, and in 1914 Maze claimed from solution culture experiments that small amounts of B were essential for maize. Maze also claimed in 1914 that Zn was essential for maize. In 1916 Grossenbacher reported that applying copper sulphate to soil or spraying trees with Bordeaux mixture (a popular fungicide containing Cu) controlled dieback and stimulated growth of citrus in Florida. Warrington at the Rothamsted Experimental Station in England reported in 1923 that broad bean plants died in water cultures unless B was added.

Hence by 1923 there was strong evidence that, in addition to the macronutrients and Fe, four more elements (Cu, Mn, Zn and B) increased plant growth when present in trace amounts. But the conclusion that these trace elements were in fact essential nutrients, while suggested by some and clearly recognisable now, was obscured at the time by the failure of many workers to repeat the results, by reports that many other elements stimulated growth (Maze, for example, claimed as essential not only Mn, Zn and B, but also Al, Cl and Si), and by a common view that the elements, which were known to be poisonous at relatively low concentrations, acted as stimulants when present in trace amounts. Moreover, their concentrations in plants were so small and demonstration of responses so difficult that they were generally regarded as laboratory curiousities of no concern to agriculture.


Figure 2 First recognised field response to a trace element at Mt Gambier, South Australia. Right side of the central path shows an untreated plot of dead and unhealthy oats with symptoms of Grey Speck disease. Left side shows a plot treated with 84 kg manganese sulphate per hectare. (Reproduced from Samuel and Piper 1928, and supplied by J.F. Loneragan)

Samuel and Piper (1928, 1929) at Waite Agricultural Research Institute (University of Adelaide) challenged these views dramatically when they reported that applying manganese sulphate to an oat crop suffering from Grey Speck disease in Penola, South Australia, increased grain yield from nothing to nearly 3 t ha–1 (Figure 2) and that the ‘symptoms of Mn deficiency in water-cultures corresponded exactly with the symptoms of Grey Speck disease of oats in the field’. If correct, the results would require plant pathologists to adopt new paradigms to account for some plant diseases and physiologists would have to rethink both function and agricultural significance of trace elements.

For some, the change was too much to accept immediately. In a review of the Grey Speck disease that he wrote after seeing Samuel and Piper’s work and hearing their hypothesis, Carne, a plant pathologist from the Western Australian Depart-ment of Agriculture, wrote ‘The solution of the cause of the disease...will probably take longer than to find an effective control. The writer has had some evidence that the lesions are due to bacteria attacking plants in an unhealthy condition as a result of soil defects’.

But Samuel and Piper were vindicated by excited researchers worldwide who, within a few years, had cured a host of ‘diseases’ with micronutrients, including Heart-rot of various root crops in Europe (B), Pican Rosette of pecan trees in Florida (Zn), Mottle Leaf of citrus in California (Zn), Reclamation Disease of cereals in Holland (Cu), and Corky Pit of apples in New Zealand (B). Australian plant pathologists also joined in the fun, with the first published reports coming from Pittman and Owen in 1936 (Mottle Leaf of citrus in Western Australia (Zn), Exanthema of citrus in Western Australia (Cu)) and Carne and Martin in 1937 (Internal Cork of apples in Tasmania (B)).

At the same time, Australian scientists began testing the trace elements as supplements for grazing animals and as micro-nutrient fertilisers for crop production on unproductive soils in areas of good rainfall. In 1935 Lines and Marston announced that Co corrected Coast Disease of sheep-grazing pastures on calcareous sandy dunes near Robe, South Australia. Independently and almost simultaneously, Underwood and Filmer found that Co cured Wasting Disease of cattle around Denmark, Western Australia. Later, small amounts of Cu were found necessary for the complete correction of Coast Disease at Robe. In 1938 Riceman and Donald published their findings that, on the same Robe soils, a range of cereals and legumes suffered acute Cu deficency and, to a lesser extent, Zn deficiency. Meanwhile, Bennetts and Chapman had established in 1937 that Swayback disorder of lambs in Western Australia was associated with low Cu levels in pastures. Teakle and co-workers in Western Australia had also found acute Cu deficiency in potatoes and tomatoes near Albany and in grapevines near Gingin. In 1939 they reported failure of wheat from Cu deficiency at Dandaragan on the coastal plain between Perth and Geraldton.

By 1939, as in 1900, it was generally accepted that all essential plant nutrients had been discovered and that research in plant nutrition need only be concerned with the application of existing knowledge. But once again physiologists curious about abnormal symptoms in their experimental plants in solution culture had an unexpected impact on Australian agriculture. Arnon and Stout (1939) developed new methods for removing contaminants from their cultures and reported that Mo was essential for the growth of tomatoes. They failed to find any symptoms of deficiency in Californian crops and concluded ‘that the amounts of Mo essential for plant nutrition were so very, very small that nature by itself would not be able to clear any soil environment sufficiently to result in a demonstrable Mo deficiency in the field...’. They were delighted when, in 1942, A.J. Anderson of the CSIRO proved them wrong with a ten-fold increase in clover production after adding Mo to an ironstone gravelly soil in the Adelaide hills.

In 1946 Anderson and colleagues showed that clover required Mo for N fixation and that many acid soils contained unavailable Mo which was released by liming. These findings concluded nearly 300 years of nutritional research from the publication of Van Helmont’s seminal experiment and coincided with the end of the Second World War which had halted agricultural development in Australia for six years. Research findings of those 300 years now provided a sound scientific base for a massive expansion of Australian agri-cultural production and land development which followed in the next 25 years.

‘Sub and super’ (subterranean clover and superphosphate) became a catch cry for correcting P and N deficiencies in pastures and crops. Later, superphosphate was also found sometimes to correct S and Zn deficiencies accidentally. In higher rainfall areas, K fertiliser was also needed. In many areas, too, micronutrient deficiencies and failure of clovers to nodulate depressed production. Jensen, Vincent and colleagues at Sydney University, Bergerson, Hely and Brockwell at CSIRO, Canberra, Parker at UWA and Norris and Date at CSIRO, Brisbane, resolved many microbiological problems associated with nodulation of legumes and supplied effective Rhizobium cultures to nutritional colleagues. In addition, Vincent and colleagues established a laboratory to help the commercial production of effective Rhizobium inoculants, which were critical for legume establishment in many areas of Australasia.

Micronutrients combined with ‘sub and super’ increased productivity of existing farm lands and made possible devel-opment of new areas. Cu and Zn were especially important in South and Western Australia. In South Australia, large areas of native scrub in the 90-Mile Desert (subsequently renamed the 90-Mile Plain) and large areas of calcareous sand dunes of southeastern South Australia were developed following research by Riceman and colleagues in the CSIRO and by Tiver and colleagues in the South Australian Department of Agriculture. Correction of Cu and Zn deficiencies also permitted development of extensive areas of the coastal plain between Perth and Geraldton, Esperance Plains and sandy soils interspersed among the heavier soils of farms in the Western Australian wheat belt. A huge research effort by officers of the Western Australian Department of Agriculture, including Dunne, Shier, Smith, Fitzpatrick and Toms, was required to define the nutrient requirements of the various soils of the 20 million hectares developed between 1950 and 1970.


Figure 3    Response of subterranean clover to Mo in experimental plots on a podzolised soil over granite at Carlsruhe, 80 km northwest of Melbourne, Victoria. Left of white peg (arrow) superphosphate only; right side: superphosphate + 140 g sodium molybdate per hectare. Mo applied in 1956, photographed in 1957. (Photograph courtesy Rex Newman, 1957 (Department of Agriculture, Victoria) and supplied by J.F. Loneragan) 

Molybdenum was especially important in New South Wales, Victoria and Western Australia. On the Southern Tablelands of New South Wales, Anderson and co-workers identified extensive areas where as little as 100 g of Mo combined with 200 kg of superphosphate per hectare allowed establishment of legumes in pastures. Later, they showed that dressings of tonnes of lime being used for establishment of legumes in pastures on acid soils could, in many cases, be replaced by seed coated with Mo and small amounts of lime plus an inoculum of Rhizobium sp. Newman estimated that four million hectares of land in Victoria would also respond to Mo and demonstrated increases in clover yields from 1 to 5 t ha–1 (Figure 3) and the carrying capacity of pastures from one to eight sheep per hectare. Mo deficiency of clover was also widespread in Western Australia as Dunne, Fitzpatrick, Gartrell and Glencross have shown. Gartrell has also delineated a substantial area of soils on which Mo applications increased wheat yields by 20%.

Mn deficiency, though less widespread than Cu, Zn and Mo deficiencies, was also a serious problem in some areas and especially on the calcareous sand of the Yorke and Eyre peninsulas of South Australia where Reuter, Graham and others have increased grain yield of barley two- to ten-fold with Mn applications. Correction of Mn deficiency is unusally difficult on some alkaline, calcareous and organic soils which immoblise soil applications of Mn so quickly that additional leaf sprays are required to correct Mn deficiency entirely.

Following the Second World War plant physiologists had also returned to their solution cultures with increasingly sophisticated purification techniques. They established that five more elements are essential for some species — Cl for many and probably all plants (1954); Co for legumes fixing N2 (1960); Na for C4 halophytes (Feature essay 16.2) and possibly all C4 plants and beneficial to many C3 and C4 plants as a substitute for K; Si for diatoms and a few terrestrial species (1969) and beneficial to many others; and most recently Ni for legumes (1984) and cereals (1987).

Of these additional elements, only Co has been shown to depress plant production of commercial crops, and then only in lupins sown with low Co seed on lateritic sands in Western Australia. However, the levels of Co, Se and I, essential for animals, have been sufficiently low in pastures to have depressed animal production over large areas.

The total number of essential nutrients in Woodward’s ‘peculiar terrestrial matter’ now stands at 17 although no one plant would require more than 14 or 15. Nevertheless, it would still be ‘courageous’ to claim that all essential nutrients have now been discovered; it is wiser to say that plants do not need other elements at a concentration above the lowest level so far measured.

In the light of this complexity, the abysmal nutrient poverty of so much of southern and southwestern Australia and the understanding of plant nutrition at the time, we can now appreciate the impossible task faced by those unfortunate early settlers who tried farming it. We can also appreciate the outstanding contribution which plant nutritional research has made to Australian agricultural production in this century and particularly since 1950. Unfortunately, that same information has also permitted overdevelopment of some areas resulting in problems of erosion, salinity and pollution. Increased soil N due to legume-based pastures has also led to soil acidification resulting from accelerated leaching of cations accompanying increased amounts of nitrate being leached. In addition, removal of nutrients in agricultural products imposes a continuous drain on the soil’s nutrient resources which will require replenishment in order to maintain production.

In looking for solutions to these problems, research in plant nutrition has a major role to play. Increasingly since 1970, research in plant nutrition has focused on efficient use of fertilisers in agricultural production with the twin aims of decreasing costs and eliminating pollution from surface and groundwaters. Sustainable and non-polluting agriculture requires a continuous monitoring of soil nutrient supplied and the nutrient status of crops and pastures. Many plant physiological and soil tests have been and are being developed to these ends. Data from such tests are being used in models of crop production which incorporate environmental variables in order to match fertiliser applications with crop nutrient requirements. In addition, there is great interest in exploiting the huge variability known to exist among plants in their ability to obtain nutrients from soil as shown, for example, by a 30-fold range in Mn concentrations in the tops of 24 annual crops and pasture species growing on the same soil. Mycorrhizal contributions to nutrient absorption by plant roots (Section 16.4) and manipulation of plant genomes for Al tolerance (Section 16.5) are two other areas of active research in this exciting field.

References and further reading

Anderson, A.J. (1942). ‘Molybdenum deficiency on a South Australian ironstone soil’, Journal of the Australian Institute of Agricultural Science, 8, 73–75.

Anderson, A.J. (1956). ‘Molybdenum as a fertilizer’, Advances in Agronomy, 8, 163–202.

Arnon, D.I. and Stout, P.R. (1939). ‘Molybdenum as an essential element for higher plants’, Plant Physiology, 14, 599–602.

Chapman, H.D. (1966). Diagnostic Criteria for Plants and Soils, Division of Agricultural Sciences, University of California, Riverside.

Donald, C.M. (1975). ‘Trace elements in Australian crop and pasture production, 1924–1974’, in Trace Elements in Soil–Plant–Animal Systems, eds D.J.D. Nicholas and A.R. Egan, 7–37, Academic Press: Sydney.

Marschner, H. (1995). Mineral Nutrition of Higher Plants, 2nd edn, Academic Press: New York.

Russell, E.W. (1988). ‘Historical’, in Russell’s Soil Conditions and Plant Growth, 11th edn, ed.A. Wild, 1–30, Longman: UK.

Samuel, G. and Piper, C.S. (1928). ‘Grey-speck (manganese deficiency) disease of oats’, Journal of the Department of Agriculture of South Australia, 31, 696–705, 789–799.

Samuel, G. and Piper, C.S. (1929). ‘Manganese as an essential element for plant growth’, Annals of Applied Biology,
16, 493–524.

Stephens, C.G. and Donald, C.M. (1958). ‘Australian soils and their responses to fertilizers’, Advances in Agronomy, 10, 167–256.

Stiles, W. (1946). Trace Elements in Plant and Animals, Cambridge University Press: Cambridge.

Williams, C.H. and Raupach, M. (1983). ‘Plant nutrients in Australian soils’, in Soils: An Australian Viewpoint, Division of Soils, CSIRO, 777–793, CSIRO: Melbourne/Academic Press: London.