Table Of ContentsNext Page

Integrating crop physiology into the Australian soybean improvement program

Andrew T. James1 and Ian A Rose2

1CSIRO Plant Industry, Qld Bioscience Precinct, 306 Carmody Road, St Lucia Qld 4067 Australia.
NSW Agriculture, Australian Cotton Research Institute, Narrabri, NSW 2390 Australia.


In Australia, soybean cropping extends from 16oS in the tropics to about 37oS in the temperate zone. Varietal improvement is underpinned by research aimed at understanding adaptation to diverse photothermal regimes, tolerance to drought and pre-harvest weathering and grain quality required for higher value culinary use. Breeding is focussed on regions with broadly similar requirements for adaptation, disease, environmental stresses and targeted toward maximum culinary quality possible given the constraints of the environment and cropping system.

Media Summary

Soybean breeders seek to overcome the limitations of extreme sensitivity to daylength, drought and pre-harvest weathering and to understand requirements for culinary quality.

Key Words

soybean, culinary quality, breeding, physiology


Australia is a net importer of soybean. Over time, human consumption or culinary markets have taken an increasing proportion of the domestic production and some high-value export markets have developed for culinary quality soybeans produced out of season to the main northern hemisphere producers (Colton 2000). The Australian breeding program must deliver varieties over the range of 16oS in the tropics to about 37oS in the temperate zone with limited resources. The program has increasingly sought to group together regions of similar environment and potential for production of culinary quality to simplify selection of new varieties, release and maintenance of seed stocks (see Figure 1). Generally, these regions of production span a relatively wide range of latitude and planting dates. Increasingly, varieties must have quite broad adaptation to sowing date and to latitude but specific requirements for key culinary quality traits (James 2003). This paper will report a summary of the requirements for adaptation to each of the regions of soybean cropping in Australia and a discussion of research conducted to improve specific traits.

Figure 1. Regions of soybean cropping in Australia.

Regions of adaptation

Tropical and coasta

Soybean is grown largely rainfed in rotation with sugarcane along the coast or inland is irrigated in rotation with cotton or sorghum. In the sugar cropping system, soybean is planted in rows up to 140 cm wide into 10 – 30 cm deep mulch left after harvest of sugarcane. Soybean crops may be cut and left on the surface of the soil as a green manure or managed to achieve insect control and harvested for grain. Yield of the subsequent sugar crop has been shown to increase by 10 to 30 percent. The increase is thought to be due both to reduction in pest and disease pressures (including soil pests) and in provision of slow-release nitrogen (Sparkes and Charleston 2003). Varietal improvement for this region must deliver varieties with tall growth habit, high vegetative vigour, adaptation to latitudes from 27oS to 16oS and tolerance to pre-harvest weathering of grain (James et al. 1997). More recently, growers have requested light hilum to improve their access to culinary markets and higher rust tolerance. However it is relatively difficult to reliably produce grain of quality suited to human consumption from this region (Christopher and James 1998).


In this region, soybeans of crop duration around 120 to 140 days are grown rainfed on the north coast of NSW and irrigated in northern inland NSW and Qld. On the coast, soybean is grown in rotation with sugar cane, in rotation with maize or winter cereals or as part of a farming system with cattle. Inland rotations almost exclusively involve cotton or winter cereals. The main improvement aims for this region have been on improving pre-harvest weathering tolerance and resistance to phytophthora root rot and other diseases (Ryley 2003). Other important characteristics include high yield under rainfed conditions, tolerance to periodically high levels of soil manganese, good standability, tolerance to Sclerotinia sclerotiorum and resistance to downy mildew. More recently, improving culinary quality of grain for the tofu, soymilk or natto markets has been given high priority.


In northern inland NSW, the approach has been to develop early maturing, hence drought-escaping, varieties. Among the early maturing types, variation was observed in drought response when subjected to duplicate trials in fully irrigated and dryland conditions (Rose, McWhirter and Spurway, 1992). Valder, Intrepid, Valiant and Hale are varieties that have been developed at Narrabri by the NSW Agriculture breeding program specifically for dryland production. They are considered adapted to marginal moisture environments due to a combination of drought escape and drought tolerance. The levels of drought stress encountered have significant effects on seed quality through changes in seed size as well as changes in oil and protein content (Rose, 1988). This has generally had no deleterious effects on acceptance of dryland-produced beans for crushing markets, though this grain is not likely to be suitable for tofu and soymilk markets. Natto and soyflour types have therefore been targeted in selection of future dryland varieties.

Inland Irrigated

Irrigated production in central NSW river valleys has declined in recent years, this has occurred in part because of the development of a population of phytophthora races which quickly overcame new varieties as they were released, and in part because of higher profitability from other crops.


Irrigated production occurs in this region in southern NSW and northern Victoria, in rotation with paddy rice or pasture systems. A proportion of the production from this region has gained acceptance as having high functional quality for tofu and soymilk manufacture. Generally, attempts to produce natto quality grain in this region have failed due to excessively large seed size. Maintaining functional quality for tofu manufacture, whilst changing to earlier maturity and enhancing phytophthora resistance are the major aims of the breeding program in this region.

Traits enhancing adaptation or quality

Weathering tolerance

Weathering of soybean seed prior to harvest can cause economic losses in humid environments such as the north coast of NSW and southern Queensland. Rainfall at or near maturity causes seed to take up moisture with weathering damage appearing as discolouration, swelling and shrinking, partial germination, loss of weight and invasion by Phomopsis and other fungi. Losses can be total where seed quality becomes unacceptable to markets. On average, where susceptible varieties are grown, substantial losses can be expected in 20-25% of seasons, with lesser, more sporadic losses in a further 25% of years.

Genotypic differences for weathering tolerance are measured in a rainfall simulator using field grown plants harvested at or after R7 and prior to any field weathering. Studies using the rainfall simulator have demonstrated that weathering tolerance is due to the pod wall excluding water from entering the seed for up to 48hrs (Rose and Desborough, 1997) and that hardseededness is relatively ineffectual as it does not develop until grain moisture content has fallen below 17 percent which is too late to prevent damage. The benefits of weathering tolerance were demonstrated in 1996 when heavy, prolonged rainfall caused extensive losses in susceptible varieties but greatly reduced losses in Dune and Manta (Table 1).

Table 1. Comparison of field weathering in two trials (1996 only) and rainfall simulator weathering mean ratings (1995-1998) for commercial soybean varieties with contrasting levels of tolerance.


Field weathering 1996, relative to Dune (%)

Rainfall simulator weathering relative to Dune (mean %)

Unweathered seed
Experiment 1

Unweathered seed
Experiment 2













Soya 791




















Further enhancement of weathering tolerance

The weathering tolerant varieties developed in Australia have all been derived from common southern USA germplasm sources. New sources of weathering tolerance in diverse germplasm have been identified, particularly in landraces from tropical Asia. These are being crossed to Australian developed weathering tolerant varieties to test if it is possible to combine different sources of weathering tolerance and in so doing increase tolerance above existing levels.

Drought tolerance

In much of the potential Australian soybean area, summer rainfall is unreliable in amount and distribution. In many areas, irrigation is essential to supplement rainfall. In others, the development of cultivars better adapted to marginal rainfall areas is central to expansion of the crop. There are two approaches to enhancing the drought tolerance of soybean in the Australian environment. The most successful has been to develop early maturing, hence drought-escaping, varieties for inland NSW. The second strategy has been to develop longer duration drought-tolerating populations with potential adaptation to central Qld.

Early maturing strategy

Yields obtained from the early maturing strategy are clearly dependent on in crop rainfall.As an example, the expected rainfall at Narrabri for the 120 days from 1 December is 260 mm. This translates to a yield expectation of around 1.3 t/ha. In areas to the west of the dryland region, (Figure 1) the corresponding rainfall expectation is less than 200 mm and yield expectations are below 1 t/ha. In the eastern part of the dryland region, the growing season rainfall often exceeds 300mm and yields in excess of 2t/ha are frequently obtained. The yield potential of the dryland varieties that have been developed for the early maturing strategy is limited by their earliness, with full irrigation achieving around 4t/ha.

Longer duration drought-tolerating strategy

This strategy entails enhancing leaf survival in longer duration types so that the crop can survive intermittent severe water stress to go on to produce harvestable yield (James 1998). The key to improving leaf survival has been to select lines with extreme low epidermal conductance, moderate osmotic adjustment and low critical relative water content. Lines with appropriate combinations of traits have been shown to maintain leaf area for around 20 days longer than varieties selected in the early-maturing strategy.

Functional traits for tofu quality

A small-scale method for manufacture of silken tofu was developed after extensive modification of existing techniques (James and Bumstead 2002). The test measures soymilk yield, solid solubility, milk solids, tofu yield, tofu texture and tofu colour on an 80 gram sample of grain. It was found that there is a large genotypic effect on functional traits, a moderate effect of environment and relatively little genotype by environment interaction. Significant correlations among functional traits and between functional traits and grain quality traits such as seed size, protein and oil content and seed germination also occurred.

It is possible to make significant changes to the tofu making potential of soybean through selection. For example there is a large genotypic effect on tofu texture even when using grain of similar protein content (Table 2). Manufacturers prefer cultivars with higher protein content and larger seed size than is usual for varieties selected for oil seed crushing, because this will give higher yields of tofu with firmer texture. However, it is important that protein content not be pushed too high because this will have a negative impact on grain yield, sugar content and flavour of the tofu.

Table 2. Protein content and tofu texture of soybean varieties averaged across seven trials conducted in subtropical Australia.


Grain protein content relative to the average of Jabiru and Manark (% DM)

Tofu texture (0-5 scale where 5 is the highest gelling)



















Soy 791



l.s.d. (0.05)




Christopher, M.J. and James, A.T. (1998) Maturation environment and culinary soybean quality in the tropics. Proceedings, 10th Australian Soybean Conference, 15-17 September 1998, Brisbane. (Ed. A.T. James), pp.25-30. CSIRO Tropical Agriculture: St Lucia, Qld Australia.

Colton, R.T. (2000) The Australian soybean industry – past, present and future: Proceedings, 11th Australian Soybean Conference, 1-3rd August 2000, Ballina. (Ed. P. Desborough), pp. 1-7. NSW Agriculture, Australia.

James, A.T. (2003) Soybean improvement program: Proceedings 12th Australian Soybean Conference, 5-6 March 2003, Toowoomba. Northern Australian Soybean Industry Association, Toowomba, Qld. Australia.

James, A.T. (1998) Breeding for drought survival: Proceedings, 10th Australian Soybean Conference, 15-17 September 1998, Brisbane. (Ed. A.T. James), pp.111-16. CSIRO Tropical Agriculture: St Lucia, Qld Australia.

James, A.T. and E.E. Bumstead 2002 Genotypic variation in Australian soybean cultivars for tofu quality traits. In ‘Plant Breeding for the 11th Millenium’ J.A. McComb (Ed) Proceedings of the 12th Australian Plant Breeding Conference, Perth W. Australia, 15-20th September 2002 pp 770-773. Australasian Plant Breeding Assoc. Inc.

James, A.T., Chotiyarnwong, A., Laohasiriwong, S., Nakiracks, P., Neumaier, N., Pinthongkum, S., Pookpakdi, A., Promdeeraj, T., Sirichumpan, V., Lawn, R.J. and Imrie, B.C. (1997). Exploiting the long juvenile trait in Asian production systems. Pp. 64-68. In ‘Soybean Feeds the World. World Soybean Research Conference V Proceedings’. (Ed. B. Napompeth, Kasetsart University Press: Bangkok).

Rose, I.A., and P.J. Desborough. 1997. Selection of soybean cultivars for diverse environments in New South Wales, Australia. P.26-35. In Proc. World Soybean Research Conf. V, Chang Mai, Thailand.

Rose, I.A. 1988. Effects of moisture stress on the oil and protein components of soybean seeds. Aust. J Agric. Res. 39:163-170.

Rose, I.A., K.S. McWhirter and R.A. Spurway. 1992. Identification of drought tolerance in early-maturing indeterminate soybeans [Glycine max (L.) Merr.]. Aust. J Agric. Res. 43:645-657.

Ryley, M. (2003) Effects of some diseases on the quality of culinary soybean seed: Proceedings 12th Australian Soybean Conference, 5-6 March 2003, Toowoomba. Northern Australian Soybean Industry Association, Toowomba, Qld. Australia.

Sparkes, D.R. and Charleston C. (2003) Adoption of soybeans as a rotation crop in far north Queensland: Proceedings 12th Australian Soybean Conference, 5-6 March 2003, Toowoomba. Northern Australian Soybean Industry Association, Toowomba, Qld. Australia.

Top Of PageNext Page