The relationship between disease and yield is most often summarized as a simple empirical model that describes average crop performances is the presence of a pathogen. Such models may be robust and useful for surveys but their use is usually constrained to the specific conditions under which the modal was developed. Changes in production system usually invalidate the relationship. The alternative is to base the relationship on an epidemiological analysis of the pathogen population and a physiological concept of host growth and development. This review provides the knowledge and conceptual basis and discusses the limitations to progress in the development of such models. It is shown that a host-based assessment of disease is well suited to yield investigations and to multiple pest constraints, and that disease is logically related to yield via radiation interceptions and radiation use efficiency.

Powdery mildew and leaf rust caused large yield losses in spring barley grown near Christchurch, New‐Zealand, in two seasons. Disease present during early growth stages was as damaging to yield as disease late in the season. Moderate leaf rust severities after anthesis were most damaging when combined with earlier mildew epidemics. Later growth did not compensate for reduced yield potential induced by early infection. This was attributed, at least in part, to an effect on leaf size, and therefore on green leaf area, at later growth stages. There was a closer relationship, by regression analysis, of yield to green leaf area than to disease severity in three cultivars. The three cultivars. which differed in yield potential and disease resistance, were not equally sensitive to disease. It is proposed that high yielding cultivars may be the most sensitive to yield constraint by disease.

The effect of disease in two cultivars of barley, sown at different times in two seasons, on the relative importance of stored carbohydrate reserves and current photosynthesis for grain filling was assessed. Three methods of measuring stem reserve contributions to grain filling are reviewed and compared. Disease reduced stem dry weight and the amount of stored carbohydrate in most situations. In contrast, the total amount of stored carbohydrates used for grain filling was often increased by disease. The magnitude of these effects varied with the method used for estimation, and was also different in the crops sown at different times and in different seasons. The estimates of stem reserve contributions to grain filling ranged up to a maximum of 50% in some cases. At least 10 t/ha of reserve material was retranslocated in the healthy 1984 crop studied using 14C pulse feeding, and up to 0.3 t/ha more was utilized in a diseased crop. The effect of disease on the storage and utilization of stem reserves depended on the time of epidemic development, its duration, and the yield potential of the crop. This suggests that crops could be characterized as those which are very sensitive to disease during grain filling, with low stem reserves or high yield potential, and those with lower sensitivity, with more stem reserves or lower yield potential. Such interacting factors could be incorporated in future plant and yield‐loss mechanistic models.

No abstract

Dicarboximide‐resistant isolates of Monilinia fructicola were detected in New Zealand stone fruit orchards in 1985. The EC50 values of resistant isolates ranged from 3 to 217 mg a.i./l iprodione, compared with 0‐3 to 0‐7 mg a.i./l for sensitive isolates. The degree of dicarboximide resistance was maintained over nine generations in fruit tissues in most isolates, but in four isolates there was a significant decline. Three resistant and two sensitive isolates were selected for further study on agar and host tissues. Resistant isolates caused disease, grew and sporulated as well as sensitive isolates on flowers not amended with dicarboximide fungicides, but some were less fit on fruit. The pathogenicity and spore production of a resistant strain on flowers was reduced significantly by dicarboximide applied prior to pathogen inoculation. Fitness of isolates on flowers and on fruit was poorly correlated. Resistant isolates were significantly less competitive than sensitive isolates when mixed inocula were applied to flowers and fruit. The relative performance on flowers and fruit was not a reliable indicator of competitive ability.

No abstract

Components of quantitative resistance in pea (Pisum sativum) to Erysiphe pisi, the pathogen causing powdery mildew, were investigated. Conidium germination, infection efficiency, latent period and conidium production dynamics on cv. Quantum (quantitatively resistant) were compared with those on Pania and Bolero (susceptible). There was an additional comparison in conidium germination experiments with the resistant cv. Resal. Quantitative resistance in Quantum did not affect conidium germination, but infection efficiency of conidia on this cultivar was 34% less than on the susceptible Pania. More conidia germinated on 5-day-old leaflets than on 15-day-old leaflets but the age of the plant did not affect percentage germination or infection efficiency. The length of the latent period did not differ between cultivars. Total conidium production (AUC) per unit leaflet area on Quantum was 25% less than on Pania. The maximum conidium production per day (CMAX) per unit leaflet area on Quantum was 33% less than on Pania. The time to maximum conidium production per day (TMAX) was 10% longer on Quantum than on Pania. The cv. Bolero, reported to be susceptible, also showed some degree of quantitative resistance, but this differed from that of Quantum. Total conidium production was less on Bolero than on Quantum, but the conidia on Bolero were produced sooner, and for a shorter period, than on Quantum. The stability of these responses was tested by analysing components in three different temperature regimes and testing for interactions with temperature, and with leaflet age. Temperature affected all conidium production variables. AUC per leaflet area was nearly seven times as great and CMAX nearly 15 times greater at 23ЊC than at 13ЊC. TMAX increased by 1·5 times when temperature increased from 13ЊC to 18ЊC or 23ЊC. Several interactions occurred and these are described.

No abstract