|
|
||||||||
a 15 Wingate Way, Trumpington, Cambridge, CB2 2HD, UK
r.b.austin{at}dial.pipex.com
| ABSTRACT |
|---|
|
|
|---|
1 wk earlier than old cultivars. In the last two decades, most of these cultivars have carried the rht D1b dwarfing gene. The full yield benefits from modern cultivars have depended on high rates of N fertilization and the use of herbicides and effective fungicides. Data from recent trials with candidate cultivars and F1 hybrids suggest that further genetic gain in yield will be achieved during the next decade. Improved crop protection through chemicals may also enable farmers to obtain greater yields. In the longer term, substantial genetic gain in yield may be achieved if breeders are able to produce cultivars with faster growth rates and greater biomass at maturity. One way to achieve this would be to modify the photosynthetic enzyme rubisco so that its oxygenase activity is reduced. However, cultivars with potentially faster growth rates would require even more N fertilizer if their greater yield potential is to be realized.
Abbreviations: LAI, leaf area index NIAB, National Institute of Agricultural Botany NL, national list RL, recommended list
| INTRODUCTION |
|---|
|
|
|---|
as in many other countries, yields of wheat have increased steadily during the period since about 1950. The improvement has been found to be associated with the greater use of N fertilizer, the development of shorter and lodging-resistant cultivars, the use of crop protection chemicals, improved and more timely cultivation (especially earlier sowing), and beneficial interactions between the effects of these factors. During the period of rapid increases in yield, the question was often asked: can the rate of increase in yield be sustained? The evidence is in the yield statistics. Allowing for year to year variation in yield, wheat yield in the UK has increased by 110 kg ha-1 each year since 1950, and the rate of increase shows no sign of decreasing. Nevertheless, many argue that there must be a biological limit to yield; that weed, disease, and pest control cannot be better than complete (as it already is on substantial areas of the UK wheat hectarage); and that farmers are already applying close to the optimum rates of fertilizer N, P, and K. My aim here is to examine the evidence on the basis of recent improvements in wheat yields, and to speculate about whether substantial further improvements could be made and if so, by what means. The paper is not concerned with wheat quality, but it is taken that future production must have at least the same spectrum of quality characteristics as at present.
The production of wheat in Europe, as for that of many agricultural commodities, is in surplus, and various measures have been taken by planners and politicians in the European Union to bring production better into line with requirement. Up to now, the chief of these policies has been "land set aside" and a progressive decrease in the intervention price. There is also a widespread belief that intensive agriculture, including cereal cropping, is detrimental to the environment and should be discouraged, especially by limiting the rate of application of N fertilizer as well as the use of other agrochemicals. This paper does not address such issues in depth, and is confined primarily to the aim given above.
| Trends in Winter Wheat Yields and Nitrogen Fertilizer Use |
|---|
|
|
|---|
7.5 t ha-1 (Fig. 1)
. Linear regression of yield on years for the period 1948 to 1997 accounts for 91% of this variation in yield. There is no firm evidence from these data that the rate of yield gain is changing. Actually, the rate of increase in yield during the period 1948 to 1975 was 74.2 ± 6.9 kg ha-1 yr-1, compared with 113.7 ± 19.3 kg ha-1 yr-1 during the period 1977 to 1997. In calculating these two rates of increase, the data for 1976 were omitted because it was a year of abnormally severe drought and consequently low yield. There is no significant trend with years for yield to have become more variable (that is, annual deviations from the long-term trend, when expressed as a percentage of the fitted mean yields, show no trend with years; Church and Austin, 1983).
|
185 kg ha-1 (Fig. 2)
. Since 1985, rates have shown no clear trend even though grain yields increased for the same period by 129 ± 26.3 kg ha-1 yr-1.
|
1.8% of the dry matter of the grain harvested, higher grain yields are unlikely to be obtained in the long term in intensive cereal production systems without increased rates of application of N fertilizer (Austin et al., 1993). Since the early years of this century, breeders have recognized that a major limitation to the effective use of fertilizer N by wheat is the tendency of the crop to lodge. This was the reason why breeders sought to produce cultivars with shorter straw. Before the era of herbicides, cultivars with shorter straw would have suffered more from weed infestation, and even in the absence of weed competition, high yield tended to be associated with tall stature. Intensive selection was needed to obtain new cultivars that were shorter, had stiffer straw, and gave higher yield than the old ones. When the major dwarfing genes were introduced into wheat breeding programs, it was found that the dwarfness they caused was associated pleiotropically with increased grain yield and decreased straw yield. Before the introduction and adoption of cultivars carrying these genes, wheat breeders and agronomists believed that there was little scope for further increases in wheat yield, which appeared to have reached a plateau by the early years of the 1970s. In part, this was caused by losses from diseases, principally mildew (Erysiphe graminis DC), stripe rust (Puccinia striiformis Westend.), and septoria (Septoria spp.), for which at that time there was no effective chemical control and genetic resistance was poor or not durable. The great intensification of wheat breeding during the past 30 yr, forced partly by economic and political circumstances and facilitated by much improved plot machinery (the latter is still being refined and developed, as is the computerization of records and their analysis), has enabled a succession of higher-yielding cultivars to be produced. Most of them carry the rht D1b dwarfing gene (previously known as rht 2, Law, 1989), are more resistant to lodging and have better disease resistance than the "first generation" semidwarfs with the same dwarfing gene. That national wheat yields have continued to increase reflects not only progress achieved in breeding, but the greater expertise of farmers in the appropriate and timely use of machinery, agrochemicals and N fertilizer, and the availability of more effective agrochemicals. A small proportion (probably <5%) of the increase in yield is likely to be attributable to the increase in atmospheric CO2 concentration of
30 µmole mol-1 that occurred during the last half century (Kimball, 1983). | Evidence on Recent and Current Gains in Wheat Yield |
|---|
|
|
|---|
1 wk earlier than the old ones. As they matured at the same time, modern cultivars had a longer grain filling period. Evidence of gain in yield since the widespread incorporation of the rht D1b dwarfing gene into wheat cultivars comes from trials in which the responses of cultivars to N fertilizer were assessed (Foulkes et al., 1998). These trials enabled the estimation of grain yields at the optimum rate of application of N fertilizer. Results, summarized in Table 2
, show that at the optimum rate of N fertilizer, the cultivars introduced between 1977 and 1983 yielded 15% more grain than Maris Huntsman (Cappelle-Desprez was not included in this series of trials). Those introduced in or since 1985 yielded 27% more grain than Maris Huntsman, even though two out of the five did not carry the rht D1b dwarfing gene. Table 2 also shows that the three cultivars which carry the 1BL/1RS rye (Secale cereale L.) translocation yielded more than their near contemporaries which do not carry the translocation. Elsewhere, comparisons of the yields from near-isogenic lines derived from crosses between parents having or lacking this translocation have shown the lines with the translocation yield
10% more than those lacking it, though the effect of the translocation on yield depended on genetic background (Carver and Rayburn, 1994; Moreno-Sevilla et al., 1995). However, results from some other studies (e.g., McKendry et al., 1996) have shown no yield benefit from the translocation. Grain from cultivars with the 1BL/1RS translocation is satisfactory for animal feeding, but as it confers stickiness to dough, it is unsuitable for most breadmaking processes, unless its effects are counteracted by extra-strong gluten conferred by appropriate glutenin subunit alleles. It is notable that `Hereward', the most recently introduced cultivar tested by Foulkes et al. (1998), does not have the rye translocation, and is one of the highest yielding of the cultivars. Clearly, there has been a continued genetic gain in yield since the first generation of semidwarf cultivars was introduced, some of which have been independent of the rht D1b dwarfing gene and of the rye translocation.
|
|
|
|
|
|
| Prospects for Further Gain in Wheat Yields |
|---|
|
|
|---|
Past and possible future improvement in wheat yields may be considered in terms of a simple "evolutionary" proposition. Across a period of decades, if there has been no change in the average environment in which wheat is grown, yields will reach a stable level (for the first 50 yr of this century in the UK, this level was
2.5 t ha-1). In the present context "environment" includes abiotic factors, the climate, the soil, management, and inputs, as well as changes in biotic factors, particularly the evolution of weeds, diseases, and pests. A change in any component of the environment provides opportunities for breeders to modify the wheat genotype to produce cultivars better fitted to the new environment. Changes in management are not generally of the "step" kind, and so the responses to new management practices, fertilizer use, introduction of new herbicides, and others, appear to be gradual. Also, as it takes about a decade to breed a new wheat cultivar, and as new cultivars may not be fully adapted to a given change in the environment, there is inevitably a lag between the a change in management and the production of new cultivars that exploit the new conditions. Although it is a matter for speculation, the fact that UK wheat yields have increased during the last 10 yr while rates of application of N fertilizer have remained practically constant, may reflect one or more of the following: (i) the adoption by farmers of cultivars that are better able to use fertilizer N; (ii) other improvements in cultivar performance not directly related to N use efficiency (e.g., better resistance to diseases); (iii) the use of more effective herbicides that are less toxic to wheat; (iv) the use of more effective fungicides, some of which appear to benefit yield in the absence of recognized diseases; and (v) more skillful management by farmers of agrochemicals and better and more timely cultivation and harvesting.
The above reasoning can be used to speculate that even without the use of more N fertilizer, there remains some scope for the breeding of cultivars with greater yield and for improvements in crop protection chemicals and their use by farmers. In this connection, it is claimed that a new class of fungicides (strobilurins; Clough, 1993) confers a yield benefit of
0.5 t ha-1 even in the virtual absence of leaf diseases, probably by prolonging the functional life of lower leaves in the crop canopy.
The wheat yields obtained by skillful farmers in good years and on heavy, moisture-retentive soils are typically between 10.0 and 11.5 t ha-1. Average UK wheat yields, currently
7.5 t ha-1, are well below this. However, the absence of marked cultivar x environment interaction for yield in NIAB wheat trials (see also Fig. 5) suggests that the genetic component of yield increase is most likely attributable to improved yield potential.
As is now well documented (e.g., Austin et al., 1989) genetic gains in wheat yields have been accompanied by reduced straw yield, aboveground biomass having shown little discernable trend. Does this mean that there is a biomass ceiling (in a given environment)? I believe that there is only limited scope for genetic increase in biomass while the pool of variation available to breeders is confined to the Triticineae (or even to higher plants), but I am optimistic that the photosynthetic characteristics can be improved by genetic engineering (see below) thus raising the biomass ceiling and hence yield.
Evidence from Flintham et al. (1997) suggests that in the UK, the optimum mature plant height of wheat is close to 80 cm. This conclusion was reached by Flintham et al. from trials comparing near-isogenic lines carrying Rht B1b (previously called rht 1), or Rht D1b, and others lacking both of these alleles (Fig. 7) . Efforts to exploit the Rht B1c (previously rht 3) gene, which confers much greater dwarfness, have not been successful, as both in wheat as well as in triticale (x Tritosecale Wittmack) the presence of this gene reduces biomass and grain yield in all backgrounds tested. Cultivars having close to the optimum height have a harvest index of about 50%. Elsewhere, Austin (1994) argued that the scope for further increasing yield through increasing the harvest index alone is quite limited. Assuming that there would be no decrease in the leaf area index (LAI) needed to achieve maximum yield, cultivars with a higher harvest index would have less stem mass. It follows that they would either have weaker or shorter stems, or both. Aside from the evidence from Flintham et al. (1997), shorter stems would be associated with increased canopy density (LAI m-3), and would probably decrease the uniformity of light interception by the canopy and so decrease crop growth rates. Short stems and a dense canopy would also encourage leaf diseases and make harvesting more difficult. Unless there are unforseen changes in plant type, improvements in harvesting machinery and in resistance to leaf diseases, cultivars of the future will probably be
80 cm high and possess a harvest index of not much more than
50%. They will most likely carry either the Rht B1b or the Rht D1b dwarfing gene. Such cultivars will give greater yields under high-input conditions than those that are taller and/or lack one of the major dwarfing genes.
|
|
Evans and Dunstone (1970) and Austin et al. (1982) showed that there was considerable variation in Pmax among wheat species, but that genotypes with high Pmax had small, narrow leaves. High Pmax was correlated with mesophyll cell volume, and much of the variation could be ascribed to the consequences of ploidy. The relationship between leaf width and Pmax was such that the rate of photosynthesis per leaf was greatest in cultivated wheat, with its broad leaves and relatively low Pmax. However, success in plant breeding has often depended on breaking negative correlations between traits, and it is possible that continued selection for high yield will give cultivars with high Pmax, the beneficial effects of which will not be offset entirely by narrow leaf width. Indeed, some recent cultivars have narrower (and more erect leaves) than their predecessors, though it has not been determined whether they have a higher Pmax.
As many have argued, if the catalytic properties of the CO2-fixing enzyme rubisco (ribulose bisphosphate carboxylase/oxygenase) could be engineered to reduced or eliminate its oxygenase activity, photosynthesis should be increased at all light intensities (assuming no dominating internal or external limitations). In principle, this change could be brought about by increasing the maximum rate of carboxylation (VCmax) or reducing the maximum rate of oxygenation (VOmax), or altering the respective Km's. I say "engineered" because there appears to be little variation in the catalytic properties of the enzyme among higher plants. Some argue that this is because rubisco has already evolved to be the most efficient possible, consistent with the biophysics of the enzyme-substrate interactions involved. Additionally, there may be a selective advantage in the retention of its capacity for oxygenation, particularly in environments where appreciable stress is common. However, Read and Tabita (1994) and Uemura et al. (1997) have reported that some thermophilic and other algae have rubisco with a much greater relative specificity for CO2 than for O2, strictly (VCmaxKmO)/(VOmaxKmC). This feature of their rubisco is associated with changes in the base sequence in a particular region (loop 6) of the large subunit of the enzyme. However, engineering of the rubisco of Synechococcus to give it the base sequence in loop 6 common to the thermophilic algae did not benefit its relative specificity (Read and Tabita, 1994). As far as I am aware, results of corresponding mutation experiments in higher plants have not been published. From an extension of the model of Farquhar et al. (1980) the possible effects of altered rubisco can be calculated. Suppose that the relative specificity of the enzyme in wheat could be increased from its present value of about 95 to a new value of about 195, similar to that of the thermophilic alga Galderia partita (Uemura et al., 1997). This might be made by increasing its VCmax and decreasing its VOmax so that the sum of VCmax and VOmax was unchanged, its other properties remaining the same. In wheat plants so transformed, leaf photosynthetic rate at current ambient CO2 concentrations and at a leaf temperature of 20°C would be increased by
20% compared with physiologically relevant light intensities. At doubled CO2 concentrations, the corresponding benefit would be
10%. These beneficial effects of increasing rubisco relative specificity are calculated to be greater at higher temperatures. Assuming that the benefits were expressed across the entire growth cycle of wheat, we may speculate that biomass and LAI at anthesis would be increased, and that the benefit to yield would be greater than the 20% which would be expected if photosynthesis were increased only during grain filling. Whether such benefits will be realized will only become evident after much more experimentation involving site-directed mutagenesis and the evaluation of transformants in the field. A reason for strong optimism is that the growth and yield of wheat, like that of many other C3 plants, is increased at above-ambient atmospheric CO2 concentrations (Kimball, 1983; Kimball et al., 1995). A reason for pessimism is that the activity of rubisco is under close control in response to external (light, temperature) as well as to internal (demand or "sink") factors. Such effects could counteract any beneficial effects of increased relative specificity. However, they might be overcome by further selection for high yield among plants that have been transformed so that their rubisco has high specificity.
| Conclusion |
|---|
|
|
|---|
Received for publication December 28, 1998.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
A. J. Challinor, F. Ewert, S. Arnold, E. Simelton, and E. Fraser Crops and climate change: progress, trends, and challenges in simulating impacts and informing adaptation J. Exp. Bot., July 1, 2009; 60(10): 2775 - 2789. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. A. Semenov and N. G. Halford Identifying target traits and molecular mechanisms for wheat breeding under a changing climate J. Exp. Bot., July 1, 2009; 60(10): 2791 - 2804. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Sylvester-Bradley and D. R. Kindred Analysing nitrogen responses of cereals to prioritize routes to the improvement of nitrogen use efficiency J. Exp. Bot., May 1, 2009; 60(7): 1939 - 1951. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Bertheloot, P. Martre, and B. Andrieu Dynamics of Light and Nitrogen Distribution during Grain Filling within Wheat Canopy Plant Physiology, November 1, 2008; 148(3): 1707 - 1720. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. K. Ransom and M. V. McMullen Yield and Disease Control on Hard Winter Wheat Cultivars with Foliar Fungicides Agron. J., June 23, 2008; 100(4): 1130 - 1137. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. A. Ainsworth, A. Rogers, and A. D.B. Leakey Targets for Crop Biotechnology in a Future High-CO2 and High-O3 World Plant Physiology, May 1, 2008; 147(1): 13 - 19. [Full Text] [PDF] |
||||
![]() |
M. Brancourt-Hulmel, G. Doussinault, C. Lecomte, P. Berard, B. Le Buanec, and M. Trottet Genetic Improvement of Agronomic Traits of Winter Wheat Cultivars Released in France from 1946 to 1992 Crop Sci., January 1, 2003; 43(1): 37 - 45. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Vadose Zone Journal | |||
| Journal of Natural Resources and Life Sciences Education |
Soil Science Society of America Journal | ||||
| Journal of Plant Registrations | Journal of Environmental Quality |
The Plant Genome | |||