A personal view

I would like to suggest that there should be an increase in investment in wheat breeding in Australia and that geographically distributed programs establish separate pools of germplasm to enhance the adaptation of wheat to regions not being adequately serviced. An increase in investment is well justified.

There are significant opportunities to improve yield, adaptation and quality, and past performance indicates that wheat-breeding programs in Australia have had ex post benefit: cost ratios up to 20. Arguably, they are among the most profitable of all the GRDC's programs.

A proposition is made that the debate on the future of wheat breeding in Australia has focused too much on new technologies and their likely impact on breeding, and too little on the objectives of the breeding programs (see last issue of Ground Cover). More focus should be put on the objectives, as those that are inconsequential, or impossible, can only hinder the progress of the breeding program.

Furthermore, the number of objectives that must be addressed only ever increases and the management of this greater degree of complexity poses major technical issues and stretches ever more the scarce resources available for each of the programs.

The breeder must keep a tight rein on the number of objectives. For every extra objective adopted, the size of the selection population must be doubled.

Ability to adapt breeding to circumstances

Furthermore, breeding objectives must be kept flexible. Since the last issue of Ground Cover we have had:

  • a breakdown in the Lr24 (leaf rust) resistance gene, making the major South Australian variety Krichauff vulnerable
  • another devastating frost in the Wimmera, emphasising the need for a program for late-maturing wheats of the type of Kellalac, and
  • floods in northern NSW, increasing the risk of over-summering of rusts and further breakdowns in resistance.

To meet these problems we need to:

  • change the races of leaf rust fungi that are used for screening wheat breeding lines and introduce new sources of resistance to varieties such as Krichauff which are otherwise well adapted
  • develop a major new germplasm pool in Victoria, and
  • diversify the sources of resistance to rust (and other diseases) in the northern programs.

Wheat genome still far from understood

Another central problem to breeders is the immense size and complexity of the genome (total genetic material) of wheat. Despite the significant progress in molecular biology, it is my opinion that we are decades from a reasonable understanding of the intricate interrelationships of one gene with another, or between the environmental signals and genetic response at a level that impacts on the progress that can be made in such a complex activity as a wheat-breeding program.

Many years ago, Sewell Wright, the famous population geneticist, likened the adaptation of a species to the peaks in a mountain range. In some instances we move further up a peak, from say Spear to Frame or Trident.

In other instances this proves very difficult. Halberd was a peak in Wright's conceptual mountain range and even with hundreds, maybe thousands of crosses, no further improvement was possible. With Halberd we have had to go through a valley of relatively poorer performance to explore a higher peak in the range of which Krichauff, which is 1/16 Halberd, is an example.

Changed emphasis to environmental concerns?

Over the last 30 years, the primary thrust of wheat breeding has been the incorporation of resistances to diseases and pests — stem, leaf and stripe rust, the Septorias, yellow leaf spot and CCN and the two RLNs. It could be argued that wheat has been taken from a high-risk crop to one of our lowest.

But I believe that the future lies elsewhere. Emphasis is shifting to tolerance to aluminium toxicity on low pH soils in southern NSW and WA, to boron toxicity in SA and Victoria, and tolerance to bicarbonate toxicity in coastal areas.

Australian wheats appear to be quite tolerant to salt, probably through unconscious selection over the last century, but this needs to be enhanced, especially in the durums. All of this requires that the breeder be intimately familiar with the local environment and the genetic diversity in wheat and its relatives.

If we look at the distribution of the current programs, we can see some serious gaps including:

  • lack of a separate breeding program (germplasm pool) for the alkaline sodic soils of WA
  • insignificant breeding effort in central west NSW
  • inadequate resources for the longer-season wheats in Victoria,
and minor gaps such as:
  • early-maturing varieties for the far west coast of SA, and
  • adaptation to the Emerald region in Queensland.

Breeding for markets

Markets also exert similar demands. Australia services a variety of specialist markets such as the udon noodle wheats originating out of WA. Given the genetic complexity of achieving market specifications, it is not possible to have a single pool of germplasm which satisfies all markets.

We must avoid the mirage of centralisation (efficiency) on the one side and the mirage of molecular biology (single genes) on the other. We grow wheat in a great diversity of soil types and environments and for a range of markets. The breeders of the future must, therefore, retain maximum flexibility in breeding methods in Australia and have an intimate knowledge of the industry.

While these are not necessarily the views of the GRDC, Ground Cover welcomes informative commentary and debate under our columns Ground Cover Forum and Guest Spot. To discuss a contribution, please contact the Editor, Ross Andrews — 02 6272 5525; or r.andrews@grdc.com.au. In particular, we invite debate with other plant breeders.

Key issues:

  • Does the achievement of economies of scale (by centralising critical resources) in a number of key breeding activities necessarily equate with reduced numbers of lines and germplasm pools?
  • Is it going to be economically viable to develop varieties for all environments? Economic viability must be carefully considered when wheat-breeding objectives are determined.
  • If scientists are able to identify and locate on the genome the key genes for rust resistance, this could enhance the speed and application of new rust resistance. Taking this example, are we really decades away from biotechnology that impacts on the progress in a complex activity like wheat breeding?
  • Can it be scientifically justified that for every extra objective adopted in a wheat-breeding program, the size of the selection population must be doubled?
  • Are benefitxost ratios in research necessarily related to industry profitability?

Region National, North, South, West