Factors influencing the production of basidiocarps and the deposition and germination of basidiospores of Crinipellis perniciosa, the causal fungus of witches' broom on cocoa (Theobroma cacao)

The production of basidiocarps by Crinipellis perniciosa on detached, dead witches'brooms from cocoa was assessed in relation to temperature, light, cocoa clone, age of broom and type of tissue, in cabinets with a daily cycle of 8 h wet and 16 h dry. More basidiocarps formed and matured at 20–25°C than at 25–30°C. In the latter regime the pilei were smaller and white, instead of the usual crimson colour, and the stipes were longer. No basidiocarps formed at 30–35°C. At 20–25°C. more basidiocarps formed and matured with light at 100 μE m^-2 s^-1 during the wet period than at 10 μE m^-2 s ^-1. Only one basidiocarp and five primordia developed on 20 brooms kept in the dark. Brooms from 10 cocoa clones at Pichilingue. Ecuador, differed in basidiocarp productivity. most basidiocarps forming on brooms from Seavina and least on ICS clones. The numbers of basidiocarps produced on brooms aged 1.2.3 or 4 months when detached from cocoa trees were similar but time to initiation of the first primordium differed considerably. More basidiocarps formed at nodes than internodes.
The discharge of basidiospores was optimal at 20–25°C and 80% RH: germination was optimal in water agar films. Neither process was dependent on light.     

Witches' broom disease of cocoa in Rondonia, Brazil: infection of vegetative flushes and flower cushions in relation to host phenology

Formation of vegetative brooms and cushion infections were assessed in relation to host phenology and fructification of Crinipellis perniciosa over 2 years. Vegetative growth occurred in distinct periods or'flushes'based on an endogenous cycle of 8 weeks. Correlation analysis showed that growth patterns of the sample trees were very similar. There were four main flushes in 1984, three in 1985, and the first two in each year coincided with periods of fruiting of the fungus. Total yearly infection increased from 13% in 1984 to 32% in 1985 when there were more periods considered suitable for infection and production of basidiocarps. Many infections on young vegetative tissue produced more than one hypertrophied shoot (broom) and 11 new infections were produced from every fruiting broom in 1984 compared with 3.2 in 1985. The total number of vegetative brooms increased from about 160 per tree in September 1983 to about 650 in September 1985, despite natural broom loss of 68% every 12 months. Infection of flower buds (cushions) was related to flowering intensity and periodicity, and a small proportion of infected cushions produced brooms. The implications of these results are discussed in relation to methods of control.     

Witches' broom disease on cocoa in Rondonia, Brazil: pod infection in relation to pod susceptibility, wetness, inoculum, and phytosanitation

Infection of cocoa pods by Crinipellis perniciosa was examined in the field in relation to pod susceptibility and variations in inoculum and climate over 2 years. The disease was incubated for 13 or 14 weeks in pods inoculated when 2 and 4 weeks old, and for 12 weeks in pods inoculated 6-12 weeks old. Pods were uniformly susceptible to natural infection during the 12 weeks after pollination. Over 75% of rain showers started between 12.00 and 02.00 h, providing sufficient water to rehydrate basidiocarps as well as to wet pods. Pods remained wet after rainfall for decreasing periods when rain occurred after 13.00 h. Incidence of pod disease was shown to be related to the number of wetness periods longer than 4 h, and to the number of open turgid basidiocarps occurring 12 weeks earlier. Wetness periods had a greater influence on disease. Pod infection was modelled by assuming that infection was limited by either wetness periods or turgid basidiocarp numbers; critical numbers of wetness periods when both factors were of equal weight were defined by an exponential equation. The model was used to predict reductions in pod disease associated with different amounts of broom removal in each year.     

Assessing resistance to Crinipellis perniciosa using cocoa callus

Callus cultures from cocoa clones reported to have different susceptibilities to Brazilian isolates of Crinipellis perniciosa , causal fungus of witches' broom disease, including Catongo (susceptible). Scavina-6, CAB 64 and CAB 67 (resistant) were inoculated with basidiospores of C. perniciosa and the subsequent development of mycelia was assessed. Generally, on Catongo callus, mycelium similar to that found in green brooms (parasitic type), developed abundantly but on callus from Scavina-6 and the two CAB clones, mycelial growth was relatively poor and tended to be similar to that in dry brooms and on agar media (saprotrophic type). There was often considerable variation in fungal development on individual callus cultures from the same clone, but there was an overall consistency between successive experiments in ranking the callus material in terms of fungal development. However, growth of isolates from Colombia and Trinidad was similar on callus from Scavina-6 which in the field is known to reac differentially to these two isolates. Experiments with isolates from Colombia indicated that inoculum from basidiocarps produced on brooms which were kept for long periods in cabinets to induce fruiting had a decreased ability to colonize callus.     

Vascular-Streak Dieback: A New Encounter Disease of Cacao in Papua New Guinea and Southeast Asia Caused by the Obligate Basidiomycete Oncobasidium theobromae

ABSTRACT The basidiomycete Oncobasidium theobromae was identified as the cause of a devastating disease of cacao named vascular-streak dieback (VSD) in Papua New Guinea in the 1960s. VSD now causes losses among cacao seedlings and kills branches in mature cacao trees throughout Southeast Asia and parts of Melanesia. The characteristic symptoms include a green-spotted chlorosis and fall of leaves beginning on the second or third flush behind the stem apex, raised lenticels, and darkening of vascular traces at the leaf scars and infected xylem. Eventually complete defoliation occurs and, if the fungus spreads to the trunk, the tree will die. O. theobromae is a highly specialized, near-obligate parasite of cocoa. It is a windborne, leaf-penetrating, vascular pathogen, and may have evolved as an endophyte on an as yet unidentified indigenous host. The rate of disease spread on cocoa is limited because basidiocarps develop only on fresh leaf scars during wet weather, and basidiospores remain viable for a few hours on the night they are shed. Consequently, very few new infections occur beyond 80-m from diseased trees. Transmission of the disease through seed or infected cuttings has not been demonstrated. Strict quarantine measures applied to the movement of intact plants are crucial in reducing spread of the disease. Integrated management, including the planting of less susceptible genotypes, nursery construction and management, canopy pruning and shade management, provides effective control.     

Effects of temperature and leaf wetness on Pyrenophora teres growing on barley cv. Sonja

Conidia of Pyrenophora teres germinated only in the presence of liquid water and at temperatures above 2°C. The speed with which germination occurred was inversely proportional to temperature measured from a base of 2°C, up to the maximum temperature tested of 21°C. Once conidia on leaves had been wetted, about 40% of all infections that would eventually occur were established within 100°C-hours. Subsequent lesion extension was rapid, with area doubling times of about 1 day between 10 and 20°C.
If conidia germinated, up to 80% formed successful infections on young, susceptible leaves. On older leaves fewer spores germinated and the proportion that then infected was smaller.
The latent period, defined as the time before which sporulation did not occur under any wetness conditions, ranged from about 25 days at 5°C to 11 days at 20°C under dry conditions. Under continuously wet conditions it was about 20% shorter at all temperatures. Its inverse had a curvilinear relation to temperature.
Spores were produced after one to several days of humidity above 95%. The precise period decreased with increasing temperature, but at 25°C spores never appeared. The drier a dead leaf was, the longer the pathogen in it look to produce spores.     

Effects of temperature and wetness duration on conidial infection, latent period and asexual sporulation of Pyrenopeziza brassicae on leaves of oilseed rape

Experiments in controlled environments were carried out to determine the effects of temperature and leaf wetness duration on infection of oilseed rape leaves by conidia of the light leaf spot pathogen, Pyrenopeziza brassicae . Visible spore pustules developed on leaves of cv. Bristol inoculated with P. brassicae conidia at temperatures from 4 to 20°C, but not at 24°C; spore pustules developed when the leaf wetness duration after inoculation was longer than or equal to approximately 6 h at 12–20°C, 10 h at 8°C, 16 h at 6°C or 24 h at 4°C. On leaves of cvs. Capricorn or Cobra, light leaf spot symptoms developed at 8 and 16°C when the leaf wetness duration after inoculation was greater than 3 or 24 h, respectively. The latent period (the time period from inoculation to first spore pustules) of P. brassicae on cv. Bristol was, on average, approximately 10 days at 16°C when leaf wetness duration was 24 h, and increased to approximately 12 days as temperature increased to 20°C and to 26 days as temperature decreased to 4°C. At 8°C, an increase in leaf wetness duration from 10 to 72 h decreased the latent period from approximately 25 to 16 days; at 6°C, an increase in leaf wetness duration from 16 to 72 h decreased the latent period from approximately 23 to 17 days. The numbers of conidia produced were greatest at 12–16°C, and decreased as temperature decreased to 8°C or increased to 20°C. At temperatures from 8 to 20°C, an increase in leaf wetness duration from 6 to 24 h increased the production of conidia. There were linear relationships between the number of conidia produced on a leaf and the proportion of the leaf area covered by 'lesions' (both log₁₀-transformed) at different temperatures.     

Production and Germination of Conidia of Trichoderma stromaticum, a Mycoparasite of Crinipellis perniciosa on Cacao

ABSTRACT Growth characteristics of the fungus Trichoderma stromaticum, a mycoparasite on the mycelium and fruiting bodies of Crinipellis perniciosa, the causal agent of witches'-broom disease of cacao, were evaluated under controlled environmental conditions. The ability of T. stromaticum to produce conidia and germinate on dry brooms was evaluated at three constant temperatures (20, 25, and 30 degrees C) and two constant relative humidities (75 and 100%). T. stromaticum produced abundant conidia on brooms at 100% relative humidity and incubation temperatures of 20 and 25 degrees C, but none at 30 degrees C. Sporulation of T. stromaticum was not observed at 75% relative humidity at any temperature. At 100% relative humidity and either at 20 or 25 degrees C, treatment of brooms with T. stromaticum suppressed C. perniciosa within 7 days. In contrast, at 30 degrees C, treatment with T. stromaticum had no effect on the pathogen in brooms maintained at either 75 or 100% relative humidity. Mycelium of C. perniciosa grew from brooms at all temperatures at 100% relative humidity. Conidial germination on broom tissue approximated 80% at temperatures from 20 to 30 degrees C. Results suggest that applying T. stromaticum under high-moisture conditions when the air temperature is below 30 degrees C may enhance the establishment of this mycoparasite in cacao plantations.     

Identification and Spread of Fomitiporia punctata Associated with Wood Decay of Grapevine Showing Symptoms of Esca

ABSTRACT A full understanding of the pathology of esca, a chronic disease of grapevines, has been problematic, in part because the identity of the pathogen (or pathogens) has been difficult to determine. The wood decay symptoms of esca have been most often associated with Phellinus igniarius or Fomitiporia punctata. However, Koch's postulates have not been completely fulfilled because symptoms take many years to develop. The goal of this study was to determine the identity and mode of spread of basidiomycetes associated with wood decay in vines showing esca symptoms in Italian vineyards. Vineyards were intensively studied for the presence of basidiocarps, and mycelium was isolated from symptomatic vines. Fruiting bodies were identified by morphology, while mycelial isolates were identified by restriction fragment length polymorphism analysis of the internal transcribed spacer region of the nuclear ribosomal RNA gene cluster. Fomitiporia punctata fruiting bodies and mycelium were associated with approximately 50% of the vines showing esca symptoms in two vineyards; P. igniarius was not found in any samples. Fruiting bodies of F. punctata were found in five of six vineyards examined, but at low frequencies except in one vineyard. The diversity of somatic incompatibility types was very high; isolates from almost every vine had different somatic incompatibility types. With few exceptions, symptomatic and dead vines were not spatially aggregated within 12 vineyards. The combination of diverse somatic incompatibility types and lack of spatial aggregations are not consistent with the hypothesis that the disease is spread clonally through roots or by pruning tools. The correct identity of basidiomycetes associated with wood decay of vines with esca symptoms is important for understanding the epidemiology of this disease because F. punctata is found commonly on many woody hosts in Europe, which may represent a potential inoculum source for this disease.     

Pathogenic variability amongst isolates of Crinipellis perniciosa from cocoa (Theobroma cacao)

Dry witches' brooms from cocoa were imported from various areas within Bolivia, Brazil, Colombia, Ecuador, Trinidad and Venezuela. Basidiocarps of Crinipellis perniciosa were induced to form on these brooms and seedlings of different types of cocoa were inoculated with basidiospores either on the hypocotyl or cotyledon bud. Host reactions were assessed mainly by recording stem base swelling and broom development at the cotyledon node (hypocotyl inoculations) or the extent of swelling and branching of shoots (cotyledon bud inoculations).
Results from 30 experiments indicated considerable diversity amongst isolates in inducing disease symptoms, but suggested that two groups or populations of C. perniciosa exist on cultivated cocoa. One group (A), comprising isolates from Bolivia and Pichilingue (Ecuador) and most isolates tested from Colombia, induced severe symptoms on cocoa with Scavina 6 as one parent; the other group (B), comprising isolates from Brazil, Trinidad and Venezuela, did not. Within these groupings variants could be further distinguished by particular host reactions. Isolates from Ecuador, especially from the Oriente, a centre of diversity for Theobroma cacao, showed a range of pathogenicity comparable to that found amongst isolates from cultivated cocoa over a much wider area.     

Growth rates of, and mycelial interactions between, isolates of Crinipellis perniciosa from cocoa

Growth of 87 single-basidiospore isolates of Crinipellis perniciosa , derived from witches' brooms on cocoa in 10 localities throughout South America and the Caribbean, was examined at 25°C on five agar media (Czapek-Dox, prune extract, potato-dextrose, V8 and carboxymethylcellulose). Interactions (designated intermingling or mutually antagonistic) between paired mycelia of 64 isolates were determined on V8 plates.
Six somatic-compatibility groups were identified comprising isolates from; (1) Pichilingue and Rio Palenque in Ecuador, Chigorodo and Manizales in Colombia; (2) Sucua (Ecuador); (3) Manaus (Brazil); (4) Ouro Preto (Brazil); (5) Castanhal (Brazil); (6) Trinidad and Tobago. This geographical separation of isolates was supported by results of the growth tests; growth of isolates within each compatibility group differed from other groups on at least one of the five media.
Separation of isolates by these methods did not conflict with previous results from tests of pathogenicity and could be useful in selecting isolates for screening cocoa lines for resistance to C. perniciosa.     

Early development of light leaf spot (Pyrenopeziza brassicae) on winter oilseed rape (Brassica napus) in relation to temperature and leaf wetness.

In controlled environment experiments to study early development of light leaf spot, lesions developed with leaf wetness durations of 16 to 48 h after inoculation of oilseed rape with conidial suspensions of Pyrenopeziza brassicae at 12 or 18°C, but not with leaf wetness durations of 0 to 13h. The incubation period was 21 to 22 days at 12°C and 14 to 18 days at 18°C for leaf wetness durations of 16 to 48 h. The latent period was 21 to 23 days at 12°C and 18 to 19 days at 18°C, and the total number of lesions increased with increasing leaf wetness duration at both temperatures. In field experiments, light leaf spot always developed on oilseed rape with a leaf wetness duration of 48 h after inoculation in both 1990/1991 and 1991/1992, but the percentage leaf area affected was less on plants placed in an oilseed rape crop than on those placed in a glasshouse. Plants moved to an oilseed rape crop immediately after inoculation nearly always developed light leaf spot symptoms when they were inoculated between 19 October 1990 and 1 March 1991 or between 27 September 1991 and 14 February 1992, but plants inoculated between 31 August and 16 October 1990 or on 20 September 1991, when estimated leaf wetness duration was less than 16 h for several days after they were placed in crops, did not develop symptoms. The latent period of light leaf spot on plants transferred to the oilseed rape crop was 15 to 40 days, and there was an approximately linear relationship between 1 (latent period) and mean temperature during this period. The accumulated temperature during the latent period ranged from c. 150 to 250 day-degrees. The severity of lesions on these plants increased with increasing temperature from 5 to 15°C.     

Effect of temperature and leaf wetness duration on infection of sweet oranges by Asiatic citrus canker

Asiatic citrus canker, caused by Xanthomonas smithii ssp. citri , formerly X. axonopodis pv. citri , is one of the most serious phytosanitary problems in Brazilian citrus crops. Experiments were conducted under controlled conditions to assess the influence of temperature and leaf wetness duration on infection and subsequent symptom development of citrus canker in sweet orange cvs Hamlin, Natal, Pera and Valencia. The quantified variables were incubation period, disease incidence, disease severity, mean lesion density and mean lesion size at temperatures of 12, 15, 20, 25, 30, 35, 40 and 42°C, and leaf wetness durations of 0, 4, 8, 12, 16, 20 and 24 h. Symptoms did not develop at 42°C. A generalized beta function showed a good fit to the temperature data, severity being highest in the range 30–35°C. The relationship between citrus canker severity and leaf wetness duration was explained by a monomolecular model, with the greatest severity occurring at 24 h of leaf wetness, with 4 h of wetness being the minimum duration sufficient to cause 100% incidence at optimal temperatures of 25–35°C. Mean lesion density behaved similarly to disease severity in relation to temperature variation and leaf wetness duration. A combined monomolecular-beta generalized model fitted disease severity, mean lesion density or lesion size as a function of both temperature and duration of leaf wetness. The estimated minimum and maximum temperatures for the occurrence of disease were 12°C and 40°C, respectively.     

Canopy-microclimate effects on the antagonism between Trichoderma stromaticum and Moniliophthora perniciosa in shaded cacao

The collective impact of several environmental factors on the biocontrol activity of Trichoderma stromaticum ( Ts ) against Moniliophthora perniciosa ( Mp ), the cause of cacao witches' broom disease, was assessed under field conditions of shaded cacao ( Theobroma cacao ) in south-eastern Bahia, Brazil. Biocontrol experiments were performed adjacent to an automated weather station, with sensors and Ts -treated brooms placed at different canopy heights. Sporulation occurred at the same dates for all Ts isolates, but in different quantities. Broom moisture >30%, air temperature of approximately 23 ± 3°C, relative humidity >90%, solar radiation intensities <0·12 KW m⁻² and wind speed near zero were the key environmental parameters that preceded Ts sporulation events. A multiple logistic regression indicated that these weather variables combined were capable of distinguishing sporulation from non-sporulation events, with a significant effect of wind speed. Analyses of environmental factors at ground level indicated similar pre-sporulation conditions, with a soil moisture content above a threshold of 0·34 m³ m⁻³ preceding all sporulation events. The sporulation of five selected Ts isolates was compared at four different canopy heights. Isolates responded differently to weather variation in terms of sporulation and antagonism to Mp at different canopy levels, indicating that different microclimates are established along the vertical profile of a shaded cacao plantation. The potential of these findings for development of predictive mathematical models and disease-management approaches is discussed.     

不同树形结构梨枣的冠层特性及节水性分析

以4 a生梨枣树为试材,观测分析了不同树形经过拉枝、摘心修剪后的树体指标与冠层特性之间关系及其对水分生产效率的影响.结果表明:树体越大,叶面积指数越大,透光率越小,光截获密度越大;自然圆头形单位体积内的坐果数最多,较对照(CK)提高353.62个/m3,每667m2产量按顺序排列为:自然圆头形>Y字形>柱形>开心形>立...     

Random amplified polymorphic DNA (RAPD) analysis of Crinipellis perniciosa isolates from different hosts

DNA sequence polymorphisms among isolates of Crinipellis perniciosa , causing witches' broom disease in several genera and species in the Sterculiaceae, Solanaceae and Bixaceae, were revealed by random amplified polymorphic DNA (RAPD) banding patterns. Comparisons were made among isolates from Theobroma cacao (cultivars SCA6 and SVB), T. grandiflorum, T. obovatum, T. subincanum, Herrania sp. (all collected near Belém, Brazil, at three plots within two kilometres of each other), and Bixa orellana and Solanum rugosum (150 and 1440 km from Belém, respectively). Genetic similarities, as determined by Sorensen's coefficient ( Sc_ij ; fraction of shared bands), ranged from 1·00 ( T. cacao SCA6 and T. grandiflorum ) to 0·56 ( S. rugosum and T. subincanum). Results indicate that proximity is more important than host species for determining the genetic relationships among isolates, that there were two independent sources of inoculum at the site near Belém, and that the isolate from B. orellana is most closely related to the isolate from T. cacao SVB. Comparisons were also made among multispore cultures from different basidiocarps on the same T. cacao SVB broom, as well as monospore cultures from the same S. rugosum basidiocarp. Banding patterns were similar among basidiocarps on the same broom, but differences were detected among monospore cultures from the same basidiocarp.     

An optimized screening method for identifying levels of resistance to Crinipellis perniciosa in cocoa (Theobroma cacao)

The effects of host age, leaf number, host type (clone or seedling), pathogen spore concentration and incubation time on inoculation with Crinipellis perniciosa (witches' broom disease of cocoa) were studied in greenhouse experiments using susceptible cocoa genotypes. Three methods of inoculation (agar-drop, water-drop and spray) were also tested. An optimized inoculation method was selected and tested for its repeatability as well as its ability to discriminate between various levels of resistance to C. perniciosa in cocoa. The optimized method (350 000 viable basidiospores per mL, 60 h incubation, agar-drop technique) produced 100% infection repeatedly, on both clonal and seedling plants of a susceptible genotype. Seedling age (2–12 months) and leaf number did not significantly affect the percentage of plants with symptoms or broom characteristics. This method discriminated effectively between the various levels of resistance in 14 cocoa genotypes and is recommended as an inoculation method to identify levels of resistance in germplasm collections. Symptom severity was shown to be a better measure of resistance than infection success.     

Effects of temperature and humidity on diseases caused by Phoma medicaginis and Leptosphaerulina trifolixs in lucerne (Medicago sativa)

The post-inoculation temperature and humidity conditions most favourable for development of Phoma black stem and leaf spot (Phoma medicaginis) and Leptosphaerulina leaf spot (Leptosphaerulina trifolii) diseases on lucerne (Medicago sativa) were determined using controlled-environment cabinets. P. medicaginis caused more severe leaf disease at day/night temperatures of 21/16 and 18/13°C compared with 15/10°C; petiole disease was most severe at 21/16°C compared with 18/13°C, and least severe at 15/ 10°C. L. trifolii generally caused more severe leaf disease at 18/13°C compared with 21/16°C or 15/10°C, but the most severe petiole disease occurred at 21/16°C with least at 15/10°C Leaf disease caused by either P. medicaginis or L. trifolii was always more severe with the longest period (168 h) of high-humidity incubation and, generally, increased with increasing periods of high humidity. The severity of petiole disease was much less affected than that of leaf disease by an increase in the period of high-humidity incubation.     

Sporulation of Pyrenopeziza brassicae (light leaf spot) on oilseed rape (Brassica napus) leaves inoculated with ascospores or conidia at different temperatures and wetness durations

In controlled environment experiments, sporulation of Pyrenopeziza brassicae was observed on leaves of oilseed rape inoculated with ascospores or conidia at temperatures from 8 to 20°C at all leaf wetness durations from 6 to 72 h, except after 6 h leaf wetness duration at 8°C. The shortest times from inoculation to first observed sporulation ( l ₀), for both ascospore and conidial inoculum, were 11–12 days at 16°C after 48 h wetness duration. For both ascospore and conidial inoculum (48 h wetness duration), the number of conidia produced per cm² leaf area with sporulation was seven to eight times less at 20°C than at 8, 12 or 16°C. Values of Gompertz parameters c (maximum percentage leaf area with sporulation), r (maximum rate of increase in percentage leaf area with sporulation) and l ₃₇ (days from inoculation to 37% of maximum sporulation), estimated by fitting the equation to the observed data, were linearly related to values predicted by inserting temperature and wetness duration treatment values into existing equations. The observed data were fitted better by logistic equations than by Gompertz equations (which overestimated at low temperatures). For both ascospore and conidial inoculum, the latent period derived from the logistic equation (days from inoculation to 50% of maximum sporulation, l ₅₀) of P. brassicae was generally shortest at 16°C, and increased as temperature increased to 20°C or decreased to 8°C. Minimum numbers of spores needed to produce sporulation on leaves were ≈25 ascospores per leaf and ≈700 conidia per leaf, at 16°C after 48 h leaf wetness duration.     

In vitro effects of temperature and wet periods on infection of oilseed rape by Alternaria brassicae

The influence of temperature and duration of wet periods on infection of oilseed rape by Alternaria brassicae was studied on detached leaves and pods, leaf disks and intact seedlings. Infections increased with age of leaf and the interaction between temperature and leaf age was highly significant. On older leaves infection was optimal at 25°C. There were many infections also at 15, 20 and 29 C but relatively few infections at 10°C. On pods most infections were observed at 20 C, the highest temperature studied. Infection at each temperature increased progressively with duration of surface wetness. The minimum wet periods for infection of leaves were 3 h at 20–25°C, 4 h at 15°C, 6–9 h at 10 C and 12–24 h at 5 C and for infection of pods, between 6 h and 9 h at 10°C and 6 h (or less) at 15°C and 20 C. On leaves, dry periods interrupting wet periods limited lesion development to that obtained with the initial wet period only; on pods some further infections developed when pods were re-wetted. Dry periods of 3 h and 6 h following the inoculation of pods reduced subsequent infection but there was no further reduction by longer periods of drying to 48 h.