The stimulating effect of some fungicides on Glomerella cingulata in relation to the control of coffee berry disease

The non-parasitic strain of Colletotrichum coffeanum Noack which colonizes the maturing wood of Coffea arabica produces its perfect stage (Glomerella cingulata) on the branches. Assessments of the relative abundance of this in relation to preceding rainfall have shown that for unsprayed branches, but rarely for sprayed ones, the correlation is negative and highly significant. Various fungicides stimulate G. cingulata greatly, and this stimulation persists for many months. One other saprophytic component of the micro–flora is similarly stimulated, but with two others all the fungicides used depress sporulation. It is suggested that when the main source of infection of coffee berries is the pathogenic conidia from the wood, the mechanism of long–term control by early–season sprays may lie partly in this stimulation of some fungal competitors of the pathogen.     

Some problems of evaluating fungicides for use on coffee in Kenya

As part of a programme testing fungicides for control of coffee berry disease (CBD) promising materials were also tested in the field against leaf rust of coffee. In all, fourteen fungicides were tested, five of which have been recommended for control of CBD. Only Perenox (representative of 50% copper fungicides) and Du-Ter (not highly effective against CBD and no longer recommended) gave good control of rust; Difolatan gave some control but Benlate and Daconil were virtually ineffective.
Where only CBD is present there is generally a good relationship between disease control and yield, but in these experiments where leaf rust was also present, this relationship was seriously disturbed. The implications of this for testing fungicides for use on coffee in Kenya are discussed.     

A comparison of fungicide spray volumes for the control of coffee berry disease

A continuation of previous trials in which three sprayers were used to apply 8.4 g. of Perenox in a range of spray volumes from 0.2 to 1.7 l. per tree confirmed that the volume of spray did not affect greatly the deposit retained on the branch surface. Branches were able to retain up to about 250 mg. Cu/m.² and it is considered that a volume equivalent to 100 gal./acre would provide the greatest margin of safety to compensate for inaccuracies of application, while avoiding serious run-off. Where coffee was well pruned, intensive copper spraying raised the annual yield by about 2 cwt. of clean coffee per acre. Copper spraying did not provide an adequate control of coffee berry disease and, under certain circumstances, increased incidence.     

Inoculum sources for coffee berry disease*

Isolates of Colletotrichum spp. from coffee berries and bark in Kenya fell into four categories, distinguishable by their cultural characteristics on agar. Three of these proved non-pathogenic; the fourth invariably infected both wounded and unwounded berries and caused coffee berry disease. The sporulating capacity on bark of the various Colletotrichum strains showed peaks in the periods before both the long and the short rains. The pathogenic strain made up only a very small proportion of the whole Colletotrichum complex on the bark and its sporulating capacity never exceeded twenty spores/cm²/h. By contrast, a green berry with an active coffee berry disease lesion produced 5×10⁴conidia/h and an infected ripe berry twice as many. It is concluded that in any year, regardless of cropping pattern, diseased berries are likely to play an important role in the development of coffee berry disease.     

Control of coffee berry disease*

Field experiments comparing six spraying programmes and two fungicides for control of coffee berry disease were carried out at three sites during 1968. These established that protection of the crop by sprays applied throughout the long rains (February–July) gave the best control and increased yields threefold. Early season spray programmes, which reduced the sporulating capacity of the pathogen in coffee twigs, but which ceased before the end of the rainy season, gave no disease control in the late crop (harvested October-December) and yields were less than in unsprayed controls. All programmes which gave protection at the onset of the rains gave some disease control in the early crop (harvested June-August). Captafol (‘Ortho-Difolatan’ 80 WP) was better than a 50% copper formulation (‘Perenox’). Conditions suitable for infection, as determined from meteorological records, occurred frequently during both rainy seasons. Estimates of sporulating capacity and data from spore trappings showed that although inoculum from the bark might initiate an epidemic, subsequent disease progress was more dependent upon spores derived from diseased berries; these accounted for most of the inoculum dispersed during the greater part of the season. Removal of the early crop, which was very small, had no detectable effect on disease in the late crop. Although losses were greatest with early season spray programmes, these did nevertheless delay the onset of the epidemic. However, the subsequent rate of disease increase was greatest and ‘scab’ lesion formation least in these treatments. ‘Scab’ lesions were most abundant in those treatments which caused least delay; in unsprayed coffee, it appeared that early exposure to infection resulted in ‘scab’ lesion formation which subsequently retarded the rate of disease progress. Tree height was positively correlated with disease incidence, but there was little effect of tree aspect. Crop density had a small effect on disease incidence within branches.     

Some factors influencing the performance of spray programmes for the control of coffee berry disease

In 1968 in a field trial at Kamundu, Kenya, early-season (Jan.-Mar.) sprays of copper, captafol and Tuzet failed to control coffee berry disease (CBD), in the late (main) crop of cultivars S.L.28 and S.L.34. Copper and captafol caused some delay in disease development but eventually infection reached over 50 % in all treatments and final yields of sprayed plots were below those from unsprayed plots. In a similar trial in 1969 at Yara on the cultivar French Mission a similar spray programme achieved significant benefits: copper and captafol kept the level of disease below 20% and gave improved yields compared with unsprayed plots. A continuation of the sprays until June increased yields still further. In a second trial in 1968 at Kiamara on the cultivar French Mission substantial quantities of early crop were present and disease in this was effectively controlled by early-season sprays of captafol, although the late crop on the same trees became severely affected. Removal of this overlapping early crop in January caused only a limited delay in disease development in the late crop, despite the fact that with some treatments this stripping resulted in the removal of a large source of potential inoculum from the tree. The results are considered in relation to differences between seasons and fungicides and it is concluded that further evidence has been provided for the importance of the protective role of fungicides against CBD. It is considered that stripping of the early crop is never likely to be a necessary management practice.     

Development of coffee berry disease in relation to the stage of berry growth*

The susceptibility of coffee berries to coffee berry disease (CBD) varies with the stage of berry development. It is well known that mature green berries are relatively resistant, while ripening berries are highly susceptible; however, differences in susceptibility of berries during early stages of development have not hitherto been examined in any detail. Experiments were carried out at a medium-altitude site (5200 ft) to examine the effect of inoculation, in the field, 4, 6, 10 and 14 weeks after flowering. The highest infection rates occurred when berries were inoculated 6 and 10 weeks after flowering when the berries were expanding rapidly. At 4 weeks, when berries were at the ‘pinhead’ stage, and at 14 weeks, when they were attaining full size, much lower rates were recorded. Comparable results have also been obtained at a high altitude (6000 ft) site. In an experiment carried out on a long-rains crop, flowering on 1 March, lesions regularly developed 2–4 weeks after inoculation, whereas in an experiment on a short rains crop, flowering at the end of October, symptom expression was apparently delayed. Possible reasons for this difference are discussed.     

Water-borne spore dispersal in coffee berry disease and its relation to control*

Devices are described which were used to catch rain-dispersed Colletotrichum conidia within and between coffee trees infected with coffee berry disease (CBD). The amount of CBD inoculum dispersed was related to the number of diseased and sporulating berries but not to the sporulating capacity of the fungus in maturing bark. Wet conditions encouraged spore production and dispersal, and spraying with fungicides prolonged sporulation of diseased berries. The amount of pathogenic inoculum, subsequent disease development and crop losses were greatly increased by the presence of only a few diseased berries during growth of the young crop. Spore concentrations were greatest close to diseased berries, but most spores within trees were dispersed downwards through the canopy. Some spores were dispersed between trees by wind-driven rain-splash and vectors such as pickers. Inoculum sources in tree tops are most important during CBD epidemics and disease can be restricted by removing or spraying the tops of multiple-stem trees. Fungicide applied to tree tops effectively controls CBD, because spores dispersed during rain are accompanied by redistributed fungicide     

Strains of Colletotrichum coffeanum, the causal agent of coffee berry disease, tolerant to benzimidazole compounds in Kenya

The detection of Colletotrichum coffeanum tolerant to methyl ester of benzimidazole 2-carbamic acid (carbendazim) and a related benzimidazole compound, cypendazole, followed increases in levels of coffee berry disease observed on Coffea arabica in experimental plots sprayed for 2 yr with these compounds. Sporulation by the pathogen on naturally infected berries removed from carbendazim-, cypendazole- or benomyl-sprayed plots was not checked by a further application of 0–05 % (a.i.) of any of the compounds. Nearly all the isolates from these berries were capable of some growth on agar media containing 1000 ppm (a.i.) of either carbendazim or cypendazole. However, only a few could tolerate 1000 ppm of benomyl and the inability of this compound to reduce sporulation on berries infected with tolerant strains was presumably due to its rapid conversion to carbendazim within the host tissue. Less than 1 ppm of carbendazim, cypendazole or benomyl was needed to give 50% inhibition of conidia of the normal strain. Against the most tolerant strains, however, the LD 50 was > 100 ppm of carbendazim and about 30 ppm of benomyl. Whether isolated from unsprayed or benzimidazole-sprayed plots, all isolates of Colletotrichum acutatum, a saprophytic cohabitant of lesions initiated on berries by C. coffeanum, showed the highest degree of tolerance to benzimidazole compounds. No tolerance of either fungus to the ‘conventional’ fungicide captafol was detected.     

The effect of fungicidal treatments on sporulating capacity in relation to the control of coffee berry disease

The capacity of the maturing wood of the branches of Coffea arabica to produce conidia of Colletotrichum coffeanum Noack, the causal agent of coffee berry disease, is profoundly affected by fungicidal treatments. The effect differs according to the epidemiology of the disease. When the main source of inoculum is from the wood, the reduction in the sporulating capacity persists throughout the crop-season, and good control of the disease results. When the main source of inoculum is from diseased berries, reinfection of the branches takes place, the sporulating capacity of sprayed branches rises above that of unsprayed ones, and there is indifferent or no control. Evidence from field trials suggests that the correct timing of early season spray schedules may be related to the part of the annual rhythm of sporulating capacity at which they are applied. As this rhythm is determined by rainfall, and varies from year to year and from place to place, the correct time for the start of a schedule should be predictable from meteorological data.     

In vitro screening of coffee genotypes for resistance to coffee berry disease (Colletotrichum kahawae)

An in vitro protocol was developed to screen Coffea arabica genotypes for resistance to coffee berry disease caused by Colletotrichum kahawae. Initially, cultural conditions which influenced the growth of isolates of C. kahawae on agar media suitable for callus growth were determined. The growth of the fungus on the callus derived from susceptible and resistant genotypes was then assessed. This ensured that no detrimental competition for nutrients between the pathogen and the calli occurred. Optimisation of the concentration of the phytohormones added to the media, the temperature and incubation period were found to be important in the expression of differential responses of calli to inoculation with the pathogen as detected by measurement of hyphal growth. The screening of calli of nine C. arabica genotypes showed that this method identified genotypes highly resistant or susceptible to the disease and was sufficiently sensitive to distinguish those genotypes with moderate or low resistance.     

Seasonal variations in the sporulating capacity of the fungus causing coffee berry disease

Assessments of the sporulating capacity of Colletotrichum coffeanum Noack in coffee branches (defined as the number of conidia produced per cm²per hour under optimum conditions) have been made for 4 years on a number of different sites in Kenya. An annual rhythm is evident which is essentially similar on all sites and for all years. Early-season and late-season peaks are separated by a mid-season trough. The change in sporulating capacity at any time is negatively correlated with wetness in the immediately preceding period, and daily rainfall can be used as a measure of wetness. The position of the peaks and their duration is defined by the timing of the onset and the duration of the monsoon rains, and is, therefore, subject to considerable variation. It is suggested that the correct timing of control sprays might be related to the form of the annual curve of sporulating capacity, and be predictable from rainfall records.     

Rainfall and cropping patterns in relation to coffee berry disease*

Both the increase in severity and extent of coffee berry disease (CBD) during the 1960s and the failure of early-season sprays to control it have been attributed to changes in climatic conditions and cropping patterns. A critical analysis of rainfall data since 1951, when CBD first appeared in the East Rift coffee districts of Kenya, produces no evidence that conditions affecting CBD development were any different in the 1960s than in the 1950s, in either high-or low-altitude coffee-growing areas. A marked increase in overlapping crops occurred during the late 1950s, and this was caused mainly by a change in pruning methods. The increased disease hazard from this change of cropping pattern cannot account for the failure of early season sprays, which were successful before 1962. An estimate of the time of main crop flowering shows that only in those years when flowering was early and the rainy season finished early did early-season sprays achieve substantial benefits. It is concluded that their effect in these years was obtained by fungicidal protection of the crop throughout most of the rainy season, and that their failure in other years contributed to the worsening disease situation.     

Chemical investigations on resistance to coffee berry disease in Coffea arabica. An antifungal compound in coffee cuticular wax

Investigations have been made into the possibility of a chemical basis for resistance to coffee berry disease in certain Coffea arabica varieties. Chloroform extracts of the cuticular wax of berries, purified by fractionation and chromatography, proved toxic to a range of fungi including Colletotrichum coffeanum, the causal organism of coffee berry disease. Leaf cuticular wax also contained this fungitoxic principle, though at a lower level. The possible importance of this antifungal material in determining resistance to coffee berry disease is discussed.     

Testing ant predation on the coffee berry borer in shaded and sun coffee plantations in Colombia

Soil‐dwelling ants, many of which are generalist predators, are more diverse in shaded than in sun coffee plantations without trees. We compared ant predation on the coffee berry borer, Hypothenemus hampei (Ferrari) (Coleoptera: Curculionidae: Scolytinae) in three shaded and three sun coffee plantations in Apía, Colombia, in both the wet and the dry seasons. We found that H. hampei adults exposed to ants for 5 days suffered higher removal in shaded plantations and in the wet season. In the laboratory, we observed that ants killed 74–99% of H. hampei adults over the course of 5 days. Ants appear to be important predators of H. hampei, particularly in shaded coffee plantations and in the wet season.