Take-all disease of cereals

ABSTRACT

This review comprises eight chapters, the first six of which attempt to develop a modern concept of take-all in the UK by providing a general background and discussion of the problems that face farmers and researchers and the issues that arise from them. These first chapters contain the justification for the practical advice to farmers given in the seventh chapter and the reasons for the recommendations for future research found in the eighth chapter. A system of cross-referencing has been used to allow a more open presentation of the information and there is an extensive reference section in which the majority of entries relates to the last decade.

Take-all: background and current perceptions
Take-all is the most damaging root disease of wheat world-wide and among the most important cereal diseases in the UK. It represents a major challenge in plant pathology a) because of the losses it causes and the constraints it imposes on rotational practices, b) because, being caused by a soil-borne, root-infecting fungus, it does not respond reliably to conventionally-applied fungicides and c) because no important cultivar resistance exists. There has been much research on many fronts over the last sixty years and this has contributed to take-all becoming an important model for the study of soil-borne diseases generally. Some relatively recent research from abroad, notably on the use of bacteria as biological control agents and on particular forms of nitrogen-containing fertilizers and chloride-containing fertilizers to control take-all has been much publicized in the UK, but the optimism created has not been supported by the performance of such treatments in British farming conditions. Such geographical differences are therefore discussed and emphasized in relation to the behaviour of take-all and its response to treatments.

The geographical distribution of take-all is determined largely by climate, as is the growing of its host cereals. Its incidence and severity are determined principally by the proportion of susceptible crops in rotations and by environmental factors such as soil type and moisture content. Other farming practices such as sowing date and methods of cultivation, the nutritional status of the soil and application of fertilizers also affect the disease, but usually to a lesser extent. Because significant take-all does not normally occur in a susceptible crop (wheat, barley, rye or triticale) which is grown after a non-susceptible crop in a rotation, it could be mostly avoided, but there are often economic reasons why such rotations are not practised. Therefore the risk of take-all imposes constraints on the husbandry practices of farmers growing sequences of cereals. The ways in which these are manifested are discussed in detail.

Some of the agronomic options open to the farmer for controlling take-all are decreasing inoculum by growing a non-susceptible crop in a rotation, minimizing the effects of take-all by growing a more tolerant cereal host than wheat, delaying sowing, or judicious use of fertilizers. Potential additions to the armoury of control measures are fungicides and biological control agents (BCAs). The most obvious chemical approach is to apply fungicides to the soil. Fungicidal compounds need appropriate physico-chemical properties to be effective in soil; whilst these are reasonably well understood, suitable compounds for use in the complex soils found in the UK are not presently available. Biological control using resident micro-organisms in the soil is already practised, knowingly or unknowingly, on many farms where soil suppressiveness is exploited either by inducing take-all decline (TAD) in long sequences of cereal crops (this usually involves a severe attack of the disease to induce the phenomenon), using fields known not to favour take-all, or inducing the build up of antagonistic Phialophora spp. under preceding grass crops (which delays the onset of severe take-all). On the other hand, BCAs suitable for application to control the disease in high risk situations are not available, despite much research. Bacteria, including fluorescent pseudomonads and Bacillus spp. , have offered most promise, whilst an experimental fungus introduced at the beginning of a cereal sequence is of current interest in the UK. Problems that have to be resolved, and are at the centre of much foreign research concern, the unreliability of BCAs because of their environmental requirements, poor root colonization, lack of persistence, stability and probable regional adaptation of the organisms. There is also much ongoing research (again mostly abroad) on proposed modes of action, including antagonism, competition for nutrients, root stimulation, siderophores and various mechanisms of antibiosis. Integrated control using combinations of agronomic factors has been investigated in the UK. Whilst the importance of delayed sowing was paramount and single fungicide or fertilizer treatments were more or less effective, combinations of treatments rarely achieved better control than the best of the constituent treatments.

Control strategies are unlikely to be implemented successfully without the ability to assess risk, which depends on a fundamental understanding of the epidemiology of the disease. Modelling approaches to explain the spread of infection and rates of epidemic development are described, emphasizing spatial and temporal aspects and the importance of host development. Detailed information from long-running field experiments is a scarce and valuable resource, useful for validating models and provoking ideas. Recent developments in computer graphics have helped the visualization of such large sets of data, showing trends in disease over decades and making seasonal comparisons clearer. This approach using disease progress curves has revealed a manifestation of TAD in the latter part of the growing season, which calls into question TAD studies based on seedling work only. Much more information is still needed on disease build up and decline in relation to crop sequences and sowing dates. The ability to identify the pathogen is also fundamental to the study, and ultimately the control, of take-all. The take-all fungus, Gaeumannomyces graminis var. tritici, is one of a complex of similar root-infecting fungi, including Phialophora spp. Modern serological and molecular biological techniques are being used to study the taxonomic relationships among these fungi and may prove useful for accurate diagnosis of taxa and quantification of infection.

Field experiments on take-all which rely on naturally-occurring inoculum are beset by problems of unreliability because of the patchy distribution and unpredictability of the disease. Current research to overcome these problems concerns experimental design and use of artificially-produced inoculum. Artificial infestation of field soil can also be used to evaluate losses in grain yield and reductions in grain quality without many of the confounding factors relating to crop sequences and soil conditions. Such results in conjunction with survey data (which are few) help in assessing the importance of the disease. Mathematical models for yield loss can be developed from such data and analysis of disease-yield relationships emphasizes the importance of timing of infection, host growth, and factors influencing both.

Throughout this review a need to assess the importance of take-all nationally and regionally emerges. This requires monitoring of disease and measurements of yields throughout the UK over a series of contrasting seasons and should be supported by a complementary series of field experiments. Take-all has changed, presumably in response to changing farming practices and weather trends. Continued research is required to understand these changes so as to achieve a more reliable assessment of risk. Similarly, understanding the complex nature of the disease demands continued epidemiological, ecological and biological research involving more coordinated and less fragmentary effort from the relevant research groups than hitherto.

Recommendations for minimizing losses from take-all are given. In summary, these are: 1) avoid damaging take-all by using short rotations or growing continuous cereals to exploit TAD; it may be possible to achieve TAD without serious losses by using a more tolerant cereal (e.g. barley) in the year of greatest risk; 2) ensure adequate availability of nitrogen and avoid phosphate deficiency; 3) sow second, third and fourth wheats (i.e. those most at risk) later than first and other wheats; 4) ensure adequate drainage and avoid loose seed beds; 5) apply lime to prevent acid patches, preferably before a break or first cereal, and do not overlime; 6) avoid the build up of perennial grass weeds, some of which are hosts to the take-all fungus.

Research recommendations
In the following summary, topics recommended for further research are listed in four groups. In parentheses against each topic is a) the authors' assessment of its importance to achieving the objective expressed in the title of its group, b) an indication that there is an existing project (which may need funding for continuation or expansion) or that a new one is required and c) suggestions of organizations to undertake the work.

Establishing the importance of take-all
i. Surveys (high; new; ADAS, IACR)
ii. Disease-yield relationship (high; existing; IACR, ADAS)
iii. Diagnosis (medium; existing; IACR, Universities new; Universities)
iv. Economic evaluation (high; new; Universities, ADAS, IACR)

Improving forecasting and risk assessment
i. Data storage and availability
(high; existing; IACR, Universities, ADAS)
ii. Forecasting (medium; existing; IACR)
iii. Agronomic and edaphic factors
(high; existing; IACR and ADAS)

Understanding take-all biology
i. Epidemiology (high; existing; IACR, Universities) ii. Field work methodology (medium; existing; IACR, Universities) iii. Gaeumannomyces-Phialophora complex (medium; existing; IACR new; Universities)
iv. Ecology of pathogen and antagonists (medium; existing; ADAS and Universities)

Controlling take-all
i. Rotations (medium; existing; IACR and ADAS)
ii. Natural biological control phenomena
(high; existing; IACR and ADAS)
iii. Introduced BCAs (medium; existing; ADAS and Universities)
iv. Resistance -breeding (low; ?; ?)
v. Resistance- cross-protection (medium; new; IACR and/or Universities)
vi. Fungicides (low; existing; ?)
vii. Integrated control (medium; existing; IACR, ADAS)

This review, completed in January 1991 has 147 pages.