Multi-trophic level interactions in a cassava–maize mixed cropping system in the humid tropics of West Africa

Multi-trophic level interactions in a mixed crop, involving cassava and maize, were studied in derived-savanna in Benin, West Africa. Two trials were planted, one during the short rainy season two months before onset of the dry season and one during the long rainy season in spring. Key pests under study on maize were the noctuidSesamia calamistisHampson and the pyralidsEldana saccharinaWalker andMussidia nigrivenellaRagonot, and on cassava, the exotic mealybug,Phenacoccus manihotiMatile-Ferrero and its encyrtid parasitoidApoanagyrus lopeziDe Santis. Both crops received insecticide treatments to assess the crop loss by a pest species. On maize, intercropping with cassava reduced egg and immature numbers ofS. calamistisby 67 and 83%, respectively, as a result of reduced host finding by the ovipositing adult moth and of higher egg parasitism byTelenomusspp. Both trials showed similar effects on maize yields: on insecticide-treated maize, intercropping with cassava reduced maize yields by 9–16%, while on untreated maize the net effect of reduced pest density and increased plant competition resulted in zero yield differences; yield losses were lower in inter- compared to monocropped maize. For cassava, cropping system had no effect on parasitism byA. lopezi. Yield differences between mono- and intercropped cassava depended on time of harvest: they were large at the beginning and zero at final harvest. Land equivalent ratios were mostly > 1.5 indicating that a maize/cassava mixed crop, protected or unprotected, considerably increased the productivity per unit area of land.

The role of wild grasses in the management of lepidopterous stem-borers on maize in the humid tropics of western Africa

Sites in the humid forest of Cameroon and the derived savanna of Benin were selected to evaluate the effect of planting border rows of wild host plants on lepidopterous stem-borer infestations and on maize yield. Grass species were chosen that in surveys and greenhouse trials were highly attractive to ovipositing female moths but with offspring mortality of close to 100%, thus acting as trap plants. In Cameroon, elephant grass Pennisetum purpureum Moench significantly lowered infestations of Busseola fusca (Fuller), Sesamia calamistis Hampson and Eldana saccharina Walker and increased yields of maize though the differences were not significant during all three cropping seasons. In 1998 in Benin, the only grass tested, Pennisetum polystachion L., significantly increased parasitism of mainly S. calamistis eggs by Telenomus spp. and larvae by Cotesia sesamiae Cameron and reduced numbers of the cob-borer Mussidia nigrivenella Ragonot. In 1999, three grass species; P. polystachion, Sorghum arundinaceum (Desv.) Stapf and Panicum maximum Jacq. were tested. Panicum maximum was the most efficient species for suppressing S. calamistis and M. nigrivenella infestations and enhancing egg and larval parasitism. In the Benin trials, with the exception of M. nigrivenella damage to cobs, the grass species tested had no beneficial effect on yield because pest densities were too low and also rodent damage to maize was enhanced with grasses in the vicinity of the crop. By contrast, stand losses due to Fusarium verticillioides Sacc. (Nirenberg), were significantly reduced by border rows of grasses.

Natural enemies of the maize cob borer, Mussidia nigrivenella (Lepidoptera: Pyralidae) in Benin, West Africa

Mussidia nigrivenella Ragonot is a pest of maize cobs in West Africa. It significantly reduces maize yields and grain quality, with quantitative losses of 2–25% at harvest, and up to 10–15% indirect losses due to an increase in storage pest infestation levels. Infestation by M. nigrivenella also significantly increased the susceptibility of maize to Aspergillus flavus infection and subsequent aflatoxin contamination. Surveys conducted in different agro-ecological zones of Benin on cultivated and wild host plants during 1994–1997 revealed one egg parasitoid, three larval parasitoids and one pupal parasitoid attacking M. nigrivenella. Egg parasitism was scarce on all host plants sampled and in all four agro-ecological zones. Parasitism by larval and pupal parasitoids was usually less than 10%, and varied with host plant species. Both larval and pupal parasitoids were rare or absent in cultivated maize fields. The solitary chalcidid pupal parasitoid, Antrocephalus crassipes Masi, was the predominant species, contributing approximately 53% of the observed mortality. Logistic regression analysis indicated that this parasitoid was more prevalent on fruits of Gardenia spp. (Rubiaceae) than on the other host plant species including maize used by M. nigrivenella, and was most abundant between February and September. The differences in parasitoid diversity and parasitism between Benin and other regions suggest that there are opportunities for biological control through introduction of exotic parasitoids or using the ‘new association’ approach, which uses natural enemies of closely related host species that occupy similar ecological niches to the target pest.

The Effect of Endophytic Fusarium verticillioides on Infestation of Two Maize Varieties by Lepidopterous Stemborers and Coleopteran Grain Feeders

A series of experiments were conducted to test the effect of the presence of Fusarium verticillioides in the maize plant on subsequent infestation by coleopteran and lepidopteran pests. The effect of percent internodes 1 to 5 infected with F. verticillioides, time after planting, and maize variety on attacks of stem and ears by lepidopterous and coleopteran pests was assessed in field experiments in early and late season 1998 and early season 1999 in Benin Republic. Artificial inoculation of the first internode with fungal-treated toothpicks was compared with a hot-water-fungicide seed treatment and a control. In 1998, two varieties that differed in husk tightness, the improved DMRLSR-W and the local Gbogbe, were used. Percentage of node 1 to 5 and plants infected was highest with the inoculation treatment but tended to be similar in the seed treatment and the control. The infection rate tended to increase with time and, within sampling date, decreased with node level. Ear infection was strongly correlated with percent infected nodes, indicating that F. verticillioides in the stem predisposed kernel infection. F. verticillioides incidence was higher in Gbogbe than in DMRLSR-W. Stem and ear infestations by the pyralid Eldana saccharina, the major pest in the area, tended to be highest in inoculation and lowest in the protection treatment. The same trends were found for the pyralid Chilo spp., the tortricid Cryptophlebia leucotreta, and beetles pooled across species. Significant positive correlations were found between ear/stem F. verticillioides infection and E. saccharina, Cryptophlebia leucotreta, Mussidia nigrivenella, and the noctuid Sesamia calamistis, but the latter three pest species were only significantly correlated with fungal infection of the upper nodes of the plant. Similar to disease incidence, E. saccharina numbers in stem and ear were higher in Gbogbe than DMRLSR-W in late 1998, whereas for the pyralid ear feeder M. nigrivenella, it was reversed. It was suggested that some lepidopterous and coleopteran pests are attracted by and survive longer (or have lower mortality) on plants infected with F. verticillioides.