Bacteria belonging to the genus Xanthomonas are, with the exception of X. maltophilia which is cosmopolitan in distribution and not known to be a plant pathogen (Swings et al., 1983), plant-associated bacteria and not usually encountered in other environments. They are not invariably plant pathogens. Isolates resembling X. campestris pv. campestris have been found in asymptomatic association with plant tissue or as epiphytes (Maas et al., 1985; Angeles-Ramos et al., 1991) and there are other xanthomonads which appear to be limited to secondary invasion of plant tissue after infection by a primary bacterial pathogen (Gitaitis et al., 1987b). An analysis of published information on the hosts of Xanthomonas (Leyns et al., 1984) showed that Xanthomonas infections occur on at least 124 monocotyledonous and 268 dicotyledonous plant species. Among monocotyledons the range of hosts extends across 11 families comprising at least 70 genera, and there are hosts in 57 families of dicotyledons comprising more than 170 genera. No strong association of hosts with any of the subclasses of the angiosperm families has been found (Leyns et al., 1984).

The hosts of Xanthomonas.

Historical perspective. Loss of reproductive capacity due to infectious disease is one of many selective forces operating in natural plant populations (Harlan, 1976; Levin, 1975). One result of the imposed selection is the evolution of genotypes exhibiting an enhanced capability to reproduce despite the prevalence of a disease-causing agent. There is now clear evidence that disease resistance and tolerance are important features of natural plant populations (Segal et al., 1980; Ben Kalio & Clarke, 1979).

The Genetic Bases of Relationships between Microbial Parasites and their Hosts

The rust pathogens of cereals exist as populations of races that differ in their ability to attack various varieties. Varieties that are resistant when first released often become susceptible later due to the spread of previously undetected races but the time taken for this to occur in very variable. It often occurs so rapidly as to curtail the commercial use of otherwise satisfactory varieties. Some varieties, however, are widely grown for many years and remain adequately resistant to the prevalent rust diseases. They may aptly be described as having durable resistance. This durable or long-lasting resistance can be detected without any assumptions about, or detailed knowledge of, whether durability depends on any particular mechanisms of resistance, on various degrees of racespecificity or on many or few genes. Cappelle-Desprez is given as an example of a wheat variety with durable resistance to yellow rust. The most powerful test for the detection of durable resistance occurs when a variety is widely grown commercially for several years. A much weaker test is obtained by growing varieties in small disease nursery plots even when the test is repeated for several years. Usually, resistance which is durable is also partial or incomplete. Often, however, partial resistance of wheat to yellow rust has not been durable. Thus the observation that resistance is partial is not, of itself, a satisfactory criterion for the detection of durable resistance. It is suggested that the most obvious sources of durable resistance for use in breeding programmes are varieties which have been widely grown and have displayed this character. The transfer of such resistance during breeding may be achieved if the creation or incorporation of higher levels of resistance that have not been tested for durability is avoided. It should then be possible to derive resistance from the durably. resistant parent. Methods of achieving this are discussed.

Practical breeding for durable resistance to rust diseases in self-pollinating cereals.

Summary
1. Bacterial blight of cotton is potentially one of the most damaging diseases of cotton.

2. Its symptoms and adverse effect on yield are described.

3. Control measures include: (i) cultural techniques of crop sanitation, and the destruction of crop residues, as well as close seasons and crop rotation; (ii) chemical seed treatments and foliar sprays; (iii) the breeding of resistant varieties.

4. Techniques are described for the breeding of resistant varieties, and sources of useful genetic plant resistance identified.

5. The complex nature of the interactions between host genetic resistance, variation in the pathogen and environment is emphasized.

6. Successful resistance-breeding programmes in Africa and the USA are reviewed, and attention is drawn to the pressing need for resistant varieties in India and Pakistan.

7. Continued success in the control of bacterial blight will require integrated programmes that include cultural methods, the use of chemicals (particularly as seed treatments) and the cultivation of resistant varieties with durable resistance.

Bacterial Blight of Cotton

he chestnut blight was first noticed in the New York Zoological Gardens in 1904 (Rice et al. 1980); within 50 years the disease had swept across the eastern deciduous forests, leaving 3.6 million hectares of American chestnut trees (Castanea dentata) dead or dying (Anagnostakis 1987). Once an economically, socially, and ecologically important component of the hardwood forests of the eastern United States (Anagnostakis 1987), American chestnuts persist today only as dwindling stump sprouts from rootstocks of these once magnificent trees (Griffin 1992). The

https://www.jstor.org/stable/pdfplus/1312816.pdf

The Evolutionary Genetics of Plant-Pathogen Systems Understanding the coevolution of hosts and parasites is key to understanding their ecology

Summary Variations in the host–parasite relationship of bacterial blight of cotton, caused by Xanthomonas malvacearum, E. F. Smith (Dowson), axe elucidated in terms of the three main components of variation, namely, genetic variation in the host, genetic variation in the parasite and variations in environment. Although the relative resistance shown by host varieties differed both with the culture of the pathogen used for inoculation and with the environmental conditions, over-riding patterns of host resistance could be detected, showing that certain varieties maintained their resistance over a wide range of conditions. It was found that phage type in X. malvacearum was not related to virulence. Moreover, none of the cultures of the pathogen which had been isolated from different sources, could be shown to be identical when inoculated into a range of host varieties under a range of environmental conditions. It was concluded that the pathogen showed continuous variation in virulence and that it would be difficult and of little value to attempt to define races. Some success was achieved in relating observed variations in the host–parasite relationship to easily measured components of the environment, by using multiple regression analyses. It is suggested that this might provide a means of characterizing the complex variations observed and that the host–parasite relationship could be regarded as a dynamic system, in which disease expression is a function of the interactions of environmental factors and two polygenic systems, that of the host and that of the parasite. Intrapopulation variances were also studied in the host varieties. A population which showed little variation for resistance under one set of conditions might show considerable variation in a different environment or when inoculated with a different culture. It is suggested that selection for resistance under conditions which favour the expression of variation can lead, by repeated selection and inbreeding, to the production of resistant populations which retain their resistance under conditions in which the parental stocks showed no worthwhile resistance. These conclusions are discussed in relation to problems in resistance breeding and genetics.

M. H. Arnold

Papers

Variation in the host-parasite relationship of a crop disease