Search

Cocoa (Theobroma cacao) pod diseases cause tremendous losses, frequently eliminating over 80% of potential yield. Biological control has previously shown promise to supplement cultural control methods practiced by the majority of smallholders. Mixed Clonostachys inocula were most effective against multiple pathogens in Peru, but not in Costa Rica, where Clonostachys spp. seemed to antagonize Trichoderma spp. The objective of our study was to systematically investigate the mycoparasitic interactions between Clonostachys and Trichoderma species, in order to enable the design of mixed inocula in a rational manner. Specifically, we aimed to quantify dominance of one mycoparasite genus over the other using in vitro tests as well as field experiments. The compatibility of Peruvian and Costa Rican isolates of known efficacy was tested by pair-wise confrontations on water agar as well as two host-range assays, one of which mycoparasites were offered another mycoparasite as only host and one where conidia of a Clonostachys and a Trichoderma isolate were mixed at ratios ranging from 1:100 to 100:1 in order to determine dominance on the susceptible host, Phytophthora palmivora. A field study quantified survival and establishment of selected mixtures on surface-sterilized cocoa pods. Hyphal interactions and host-range studies coincided in identifying Clonostachys rosea as the most aggressive mycoparasite, closely followed by C. byssicola. Trichoderma spp. were least aggressive; instead, they were highly susceptible to Clonostachys spp. The outcome of interactions depended on the species used and occasionally even on the isolates confronting each other. The country of origin, however, was irrelevant. Parasitic growth of C. rosea was unaffected by Trichoderma harzianum, even at an inoculum ratio of 1:100. Parasitic growth of Trichoderma spp. progressively slowed as C. rosea concentrations rose, but the effect varied in strength with different isolates. When C. rosea and Trichoderma sp. were applied to cocoa pods in the field either separately or in combination at a ratio of 5:1, C. rosea dominated at the expense of Trichoderma sp., which was suppressed. At equal or ten times higher concentrations of T. harzianum, C. rosea was temporarily suppressed, but from two weeks after application, remained the dominant and persistent pod colonizer. Native Trichoderma spp. were poor pioneer colonizers of surface-sterilized pods and applied populations vanished within less than four weeks. Our results suggest that, although Clonostachys spp. antagonize Trichoderma spp., the incompatibility can be partly overcome by adjusting the inoculum ratio of mixed inocula in favor of Trichoderma spp. Therefore, mycoparasitism among several promising biocontrol candidates do not necessarily rule out their combination into a mixed inoculum, but can be addressed by formulating susceptible isolates at higher inoculum ratios. This, however, will not sustainably improve surface coverage by poorly colonizing Trichoderma spp. Therefore, optimal strains for the habitat and task are needed first.

Frosty pod rot (FPR), caused by Moniliophthora roreri, reduces yields of cocoa (Theobroma cacao) by over 80% within a few years of disease outbreak. Both biological and chemical control approaches are being developed to supplement cultural management. Our objectives were to optimize the agent(s), their formulation and application regime for FPR control in Costa Rica. In order to economize scarce resources for field research, we opted for an iterative optimization of treatments. We aimed to (partly) replace copper hydroxide of proven efficacy with two systemic agents, the coevolved endophyte Trichoderma ovalisporum and the oxathiin flutolanil, and assessed four stickers for improving the agents' performance. Over 3 years, we selected effective agents and their combinations: eight of 18 treatments increased yields. Orthogonal contrasts extracted three general trends: (1) Copper should be applied in water, whereas the systemic agents were best formulated with a sticker. There was no difference between the stickers BreakThru and NP-7. (2) If the first rule was observed, a switch from systemic agent to the contact fungicide in mid season was beneficial. The optimum timing for this change requires further definition. (3) T. ovalisporum-containing agents controlled black pod better than flutolanil-based ones; both systemic agents were equally effective in controlling the more severe FPR. We discuss potential benefits of switching agents in the context of organic agriculture and argue that coevolved endophytes may be particularly suited for the classical biocontrol of Moniliophthora spp. in cocoa planted outside its South American centre of origin.

Frosty pod rot (FPR), caused by Moniliophthora roreri, and black pod (BP), caused by Phytophthora spp., of cocoa (Theobroma cacao) cause combined pod losses of more than 80% in Panama. Biological control of both diseases appeared promising in Peru and is desired by certified organic producers in Panama. We evaluated both local and Peruvian fungal antagonists in participatory trials on smallholdings during two complete production cycles. Furthermore, we tested the influence of a 3% v/v cane molasses amendment on biocontrol efficacy, yield and population dynamics of mycoparasites on the cocoa pod. Significant variation was observed between the two years: FPR was more severe in the first year, BP in the second. FPR was significantly reduced by biocontrol agents (BCAs), but not by the molasses amendment. However, BCAs responded differently to molasses in both years. All BCAs reduced inoculum production by M. roreri with no consistent effect of molasses. BCAs had a lesser and more variable effect on BP, whereas molasses reduced BP slightly by increasing the efficacy of native antagonists. All BCAs and the molasses amendment enhanced the percentage of healthy pods. Molasses was beneficial to absolute yield, but only one inoculum improved yield significantly in the first year. Populations of a Peruvian Trichoderma asperellum isolate remained high for over two months after application to surface-sterilized pods. Molasses had no effect on establishment or survival of this antagonist or recolonization by any native mycoparasite. The reasons for enhanced biocontrol efficacy of the molasses formulation requires further investigation.

Rosellinia spp. cause substantial mortality of cocoa (Theobroma cacao), especially in acidic soils rich in organic matter (O.M.) and, reportedly, low available phosphate. Cultural and chemical control are relatively ineffective and a supplementary biocontrol strategy would be highly desirable. Our aim was the isolation and evaluation of promising mycoparasites against Rosellinia of cocoa. We isolated 45 mycoparasites using a baiting technique. Three of them and a commercial isolate were prioritized for further evaluation based on their broad hostrange and/or effectiveness in pot bioassays. Because Rosellinia is favored by the same soil factors as mycoparasites, we investigated the effect of soil pH (range 4–6), O.M. (6–12%), and phosphate (0– 1000 mg kg −1 ) on their interaction. In vitro, both pathogens and antagonists grew well on a wide range of phosphate (0– 9500 mg kg −1 ) and at various pH (3.0–10.0) levels. The pH optimum usually coincided with the soil of origin and the phosphate optimum depended on the phosphate source used. In pot bioassays using natural soils, mixtures of antagonists exhibited more control than single strains and mixtures were effective over a wider range of soil conditions. High O.M. favored the pathogen more than its biocontrol agents. Soil pH interacted with the treatments such that biocontrol was most efficient at higher pH values and with the highest mixtures of antagonists. The addition of phosphate enhanced plant growth but had little effect on biocontrol efficacy. We conclude that mixtures of mycoparasites should be field tested in combination with liming and removal of O.M.

Three native and two commercial biocontrol agents (Clonostachys rosea and Trichoderma spp.) were evaluated against the cocoa diseases moniliasis, witches’ broom, and black pod. Antagonists were applied either separately or as mixed inoculum in comparison with a copper fungicide and a nontreated control. Cultural control (weekly removal of diseased pods) was practiced in all treatments. Field trials were conducted on neglected cocoa farms in eastern Peru from 1998 to 2000. The fungicide treatment and C. rosea strain G-4 did not reduce disease. The other single-strain antagonists reduced moniliasis. Additionally, Trichoderma longibrachiatum and Trichoderma stromaticum reduced witches’ broom and Trichoderma virens reduced black pod. With strain mixtures, yield increases of up to 15% were obtained. A mixture of four antagonists was superior to mixtures of two or three antagonists with respect to multiple disease control and yield. Biocontrol in combination with cultural control was more economical than cultural control alone; chemical control was least economical. Fertilization improved yields by 11% independent of the disease control measure and compensated for the additional costs with net returns improving by 9%. Increasing application frequencies of a mixed biocontrol inoculum improved moniliasis and witches’ broom control linearly. Moniliasis exhibited a stronger response. Witches’ broom was significantly lower than the nontreated control only if ten applications were administered in two-week intervals. This application frequency increased yields by 15%. It was followed by three applications adjusted to the production cycle (yield increase: 12%). Three adjusted applications were the most economical biocontrol strategy under the conditions of eastern Peru. Net returns were increased by 12%. Recommendations for technology transfer are presented. 2002 Elsevier Science (USA). All rights reserved.

Five native mycoparasitic strains of Clonostachys rosea and three of Trichoderma spp. were isolated from healthy cocoa tissue or basidiocarps of Crinipellis perniciosa using a baiting technique. They were compared singly or in combination with the commercial biocontrol agent Trichoderma virens (SoilGuard) for their potential to control three cocoa pod diseases: moniliasis, caused by Moniliophthora roreri; witches' broom, caused by C. perniciosa; and black pod, caused by Phytophthora palmivora. All isolates except Trichoderma T-1 inhibited basidiocarp formation of C. perniciosa under controlled conditions. The remaining isolates except Trichoderma T-3 reduced vegetative broom formation in a seedling bioassay. Clonostachys rosea G-3 and Trichoderma strains T-2 and T-3 significantly reduced symptoms caused by M. roreri in a seedling bioassay. Host-range studies identified P. palmivora as most susceptible to mycoparasitism and M. roreri as most resistant; C. perniciosa was intermediate. Different degrees of susceptibility were discovered at the pathogen strain level, especially for P. palmivora. However, broad host-range mycoparasites which attacked all three pathogen species were identified. Under field conditions, all selected treatments except a combination of C. rosea G-2 + G-3 reduced moniliasis, the main disease, significantly by 14.6–24.9% as compared with optimized, cultural control alone. No significant reduction of witches' broom or black pod was achieved but a combination of five C. rosea strains (G-1, G-2, G-3, G-4 + G-5) performed consistently best against all three diseases simultaneously. Yield increased by 16.7% and net returns by 24%. Control of moniliasis and yield were positively correlated to the number of mycoparasites in the inoculum. The results suggest that simultaneous biocontrol of the three major cocoa pod diseases with mycoparasite mixtures is highly promising. Future development strategies are discussed.

Thirty mycoparasites were detected using plates precolonized with pathogens of the crown-rot disease complex: Colletotrichum musae (6 strains), Fusarium moniliforme var. subglutinans (2 strains), F. moniliforme, Fusarium pallidoroseum, Botryodiplodia theobromae , and Nigrospora sphaerica (1 strain each). Dry leaves, dry flower residues, green leaves, and green peel were compared as mycoparasite sources. Dry leaves were the most prolific source, whereas green material collected in the canopy yielded the least number of isolates. Host species differed significantly in their suitability as bait for mycoparasites. C. musae was the best host for the isolation of most mycoparasites; fusaria were least colonized. No difference between strains of the same host species was detected during isolation. In contrast, in dual culture, different C. musae strains exhibited significant differences in susceptibility to mycoparasitism. The consequence of this finding for practical biocontrol is discussed. Thirteen aggressive antagonists, including Pythium sp., Trichoderma sp., Gliocladium sp., and Verticillium sp., were selected for further evaluation. Preliminary efficacy, host range, aggressiveness against survival structures (conidia and haustoria), and tolerance to fungicides presently used for crown-rot control were investigated. Eight mycoparasites had a reasonable number of desirable traits, but no single strain exhibited all features required for consistent biocontrol. It is suggested that future studies should consider combining two or three strains and/or to apply them with a reduced rate of a fungicide combined into the inoculum.