Inventory of pests and natural enemies on cotton in lowland production zone of Ruzizi Plain, Kivu, Democratic Republic of the Congo

Munyuli Bin Mushambanyi Théodore1,2 1 National Center for Research in Natural Sciences, Biology Department, CRSN-Lwiro, D.S. , Kivu DR CONGO 2 Makerere University, Kampala UGANDA Correspondence author [email protected] Inventory of pests and natural enemies on cotton in lowland production zone of Ruzizi Plain, Kivu, Democratic Republic of the Congo

ABSTRACT velopment of fiber and animal industries, especially by providing cake and other products such as seed oil and fibers (Echekwu, 2001; Miller, 1984; Finley and Cotton has been an important, economic, indus- Wikinson, 1963). Cotton is the only industrial crop cul- trial crop since colonial times. Several genotypes tivated by farmers from Ruzizi Plain. It provides them are cultivated by farmers and some companies. considerableincomes (Fisher, 1917; Jurion, 1941; Knaf, Cotton is very important and had played a valu- 1946; Querton, 1925; Stener, 1927). However, a de- able role in the development of animal industries cline in cotton production has occurred and cultivation (cattle and poultry) in Eastern DRCongo, by pro- in Ruzizi Plain since 1992, due to the spread of civil wars among other factors. Surveys showed pests to be viding foodstuffs (cotton seed) necessary to elabo- a major constraint to cotton production and productiv- rate rations of livestock breeds. However, a de- ity at the farmer level (Munyuli, unpublished data). In cline in cotton production has been observed in order to find appropriate ways to control the pests, it Eastern DRCongo recently. Four years ago, a di- was necessary to know the pest species and their asso- agnostic survey was conducted to explore the situ- ciated natural enemies in order to integrate them into ation. Constraints to cotton production and pro- a sound integrated pest/disease management system for Eastern DR Congo. ductivity observed were mainly biotic. In order to find appropriate way to control the pests, it was Experimental procedure necessary to know the pest species and their as- sociated natural enemies, in order to integrate Pests and their natural enemies were surveyed them into a sound integrated pest management during the rainy and dry seasons of consecutive sea- system for cotton in Eastern DRCongo. Pests and sons of 2000-2001, in the main cotton production zones their natural enemies were collected in main cot- of the Ruzizi Plain: Kiliba, , Kamnyola, Sange, Kamvimvira, Makobola, Lemera, Kiringye, Bwegera and ton production zones.. Arthropod predators and Mulongwe. Ruzizi plain (100 x 125 km) (28°-29° E, parasitoids identified belong to the Manthidae 2°-3° S; 700-900 m), is located along aside Ruzizi river (Mantis religiosa), Formicidae (Formica spp.), on the North of Lake Tanganyika (Figure 1). The aver- Braconidae (Apanteles spp.), Coccinellidae age population density is 300 inhabitants/km². It is a (Cheilomenes lunata, Cheilomenes sulphura), semi dry area with clay-loamy soil and grasslands of Redureiidae (Rhynocoris albipilosus), Chrysoperla Imperata cylindrica and Digitaria sp. The climate of the area is dry tropical type Aw4 of the climatic classifi- carnea. Several pest species were observed be- cation of Vladamir Koppen. Two main seasons, the long longing to the Noctuidae, Gelechiidae, Pyralidae, rainy season (September-March) followed by the dry Tortricidae, Chrysomelidae, Aphidae, Aleyrodidae, season (April- August) are experienced. The rainfall pat- Pyrgomorphidae and Muridae: Order Rodentia. tern is bimodal, and the area receives an annual aver- One of the highlights of this study was the poten- age rainfall, which varies between 700 and 1100 mm, tial for integrating predators in IPM systems de- with average temperatures of 24 °C and a mean rela- tive humidity of 69% (Munyuli, 2002). Several ethnic velopment for Eastern DRCongo. groups live there, including Babembe, Bashi, Bavira, Bafulero and some migrant pastoral population of Tutsi. Introduction Livestock is the key activity for agriculture. Farmers cul- tivate staple crops (potatoes, beans, cassava etc.) and Cotton together with coffee and tea are impor- industrial crops such as cotton. tant lucrative, economic and industrial crops since co- lonial time (Decaene, 1948; Fisher, 1917; Jurion, Survey implementation 1941; Knaf, 1946; Querton, 1925; Stener, 1927). It Pests and their natural enemies were surveyed continues to generate important incomes to farmers during the two cropping seasons of each year of 2000 in DR Congo, through sales and export of products at and 2001. Ten farms were selected randomly along the local, regional and international markets (Anony- roads in the villages. Agreements were made with own- mous, 1983; Autrique and Perreaux, 1989; Vendenput, ers of the farms so that we could more easily conduct 1981). In 1999 it was estimated by DRCongo gov- the surveys on their farms. Time was taken to explain ernment (Ministry of Agriculture, Annual report, 1999) to farmers the value of the survey; once the farmer was that more than 2.7% of annual income of farmers in convinced, thereafter, he could involve himself in the Kivu and Kasai provinces was from cotton cultivation. survey-work on his garden. Each selected cotton farm Several genotype belonging to Gossypium hirsutum and was divided into sections to facilitate the survey. Plants, Gossypium barbadense (Cauquil, 1986; Ahanchede, leaves, fruits, flowers, stems and roots were inspected. 2000; Coulloud, 1989; Echekwu, 2001) are cultivated Collection techniques included pitfall traps for spiders, by farmers and some companies such as “Est Agrico, earwigs, ground beetles, ants predators and other sur- CotonCongo and Cotonnière du Lac”. Cotton is very face moving arthropods and sweep nets for moths, important and has played a valuable role in the de-

1346 World Cotton Research Conference-3 2003 Cape Town - South Africa dragonflies and butterflies. Rodents were captured us- In the future, it will be necessary to establish the ing traditional traps. Traps were set in farms and feeding behavior of different predators observed, un- checked at a regular interval of two weeks. Arthropods der laboratory and field conditions, in order to deter- captured were kept in alcohol in tubes and brought at mine the population density of a predator species that the laboratory for identification. Pupae, larvae and eggs can control a given population of a pest species. of pests found, were brought to the laboratory, and reared thereon local or artificial diets until parasitoids Acknowledgement emerged, and thereafter parasitoids were collected and saved in alcohol. In order to facilitate the identification I am very grateful to the “Institut Technique of certain pest species, some lepidopterans caught as Agricole de Kasenga”, , Ruzizi Plain, for technical larvae, were reared to adults. The identification of the assistance during fieldwork. I thank Dr Joubert, WCRC- pests and their natural enemies involved one of the 3 Conference chairman for sponsoring my participa- guides (Autrique and Perreaux, 1989; Cauquil, 1986; tion. Coulloud, 1989) and comparison to specimens kept at the Museum of Natural Science, and at the Entomol- References ogy laboratory of the National Centre for Research in Natural Sciences, CRSN-Lwiro, D.S. Bukavu, Kivu, • Autrique, A. and Perreaux, D. (1989). Maladies et DRCongo. When necessary, unidentified natural en- ravageurs des cultures de la région des grands lacs emies were saved in alcohol and sent to CAB Interna- d’Afrique Centrale, AGCD-ISABU, pp 232. tional UK, for proper identification. • Anonymous, (1983). La sélection et l’amélioration du cotonnier en RDCongo. Rapport Technique du Results and Discussion Centre de Recherche de l’INERA à Gandajika, INERA-Kinshasa, RDCongo, pp. 357. A checklist of pests and natural enemies found is • Ahanchédé, A. (2000). Compétition entre presented in Table 1. It can be seen that the entomo- mauvaises herbes et culture cotonnière: influence fauna of various cotton ecosystems of Ruzizi plain, is du nombre de sarclages sur la biomasse et le very diversified. Most of the species belong to the Lepi- rendement. Tropicultura, 1818: 148-151. doptera, Hemiptera, Heteroptera, Dyctyoptera, Co- • Cauquil, J. (1986). Maladies et ravageurs du leoptera, Orthoptera, Homoptera, Diptera and Hy- cotonnier en Afrique au Sud du Sahara, IRCT- menoptera (Table 1). The distribution of the species CIRAD-CFDT, France, 92 pp. varied. Some species were common in all sites, while • Coulloud, (1989). Hétéroptères déprédateurs du the habitat of others was restricted to certain sites. The cotonnier, Coton et Fibers Tropicales, France, relative abundance varied considerable from an area 200pp. to another one. The population density of ladybirds was • Decaene, R. (1948). Méthodes statistiques pour high in all the sites surveyed. l’étude des essais de rendements cotonnier à Bambesa. Bulletin Agricole du Congo Belge, 3939: Pest activities and injury levels on farms varied. 802-818. Each pest species had a period when it could cause • Echekwu, C.A. (2001). Correlations and correlated injury, serious damages and losses in cotton produc- Responses in upland Cotton (Gossypium hirsutum tion. For example, rodents were more active between L.). Tropicultura, 1919: 210-213. October and December and between April and May of • Finley, K.W. and Wilkinson, G.N. (1963). The analy- each year (Table 2). At these periods, farmers reported sis of adaptation in plant breeding program. Aus- experiencing significant damage leading to consider- tralian Journal Agric. Research, 1414: 742-754. able yield loss of cotton seeds due to rodent attacks • Fisher, M. (1917). Culture du cotton au Congo (Table 2). Significant attacks of cotton by pests were belge. Ferme expérimentale de Loukala (sankuru). seen in Kiliba, Luvungi and Lemera. Ladybirds, spi- Bulletin Agricole du Congo Belge, 77: 29-37. ders and ground beetle predator species were abun- • Jurion, F. (1941). Quelques considérations sur dant in most of the sites surveyed (Table 3). Their abun- l’orientation de la sélection cotonnière au Congo dance was apparently enough to be able to control belge. Bulletin Agricole du Congo Belge, 3232: 677- aphid population to a significant extent. 713. • Knaff, E.E.A. (1946). Méthodes culturales The parasitism of aphids (Aphis gossypii) by cotonnières du milieu indigène dans le Bas-uélé. Braconidae was very low and not sufficient to control Bulletin Agricole du Congo Belge, 3737: 816-821. prey populations. From this preliminary survey con- • Miller, A. J. (1984). Selection of subsets of regres- ducted, it was observed that predators could be inte- sion variables( with discussion). Journal of the Royal grated into IPM system development for DRCongo cot- Statistical Society, A147A147: 389-425. ton pest management, because their abundance was • Munyuli Bin Mushambanyi, T. (2002). Contribution relatively high. The percentage of parasitism was very à la promotion de la culture du blé (Triticum eastivum low, therefore it is difficult to integrate them into the L.) au Sud-Kivu, République Démocratique du IPM system for the Ruzizi plain. Congo: Evaluation du potentiel de rendement de 2

1347 Inventory of pests and natural enemies on cotton in lowland production zone of Ruzizi Plain, Kivu, Democratic Republic of the Congo

génotypes d’origine Burundaise, dans différentes • Stener, (1927). Rapport mycologique sur la zones agro-écologiques locales. Tropicultura, 20 campagne cotonnière de 1927 au Manièma et au (4): 210-216. Kivu, Bulletin Agricole du Congo Belge, 88: 529-551. • Querton, (1925). Rapport sur la propagation • Vandenput, R. (1981). Les principales cultures en cotonnière dans le district du Sankuru. Bulletin Afrique centrale, AGCD, Bruxelles, Belgique, p: 857- Agricole du Congo Belge, B16B16: 347-353. 896., AGCD, Belgique, Bruxelles, 1252 pages.

Figure 1. Map of Demo- cratic Republic of Congo (Ruzizi plain is shown between the two points). An arrow indicates the survey area.

1348 World Cotton Research Conference-3 2003 Cape Town - South Africa T T T T T able 1. able 1. able 1. able 1. able 1. List of pests and natural enemies found during the survey, in Ruzizi Plain, Eastern DR Congo, 2000-2001.

1349 Inventory of pests and natural enemies on cotton in lowland production zone of Ruzizi Plain, Kivu, Democratic Republic of the Congo T T T T T able 1. able 1. be1 Contd. able 1. able 1. able 1.

1350 World Cotton Research Conference-3 2003 Cape Town - South Africa T T T T T able 2. able 2. able 2. able 2. able 2. Relative importance of pests at different growth periods cotton in rural areas Ruzizi plain, Kivu, DR Congo.

1351 Inventory of pests and natural enemies on cotton in lowland production zone of Ruzizi Plain, Kivu, Democratic Republic of the Congo T T T T T able 3. able 3. able 3. able 3. able 3. Relative timing of natural enemies at different sampling periods in Ruzizi plain Kivu, Eastern DR Congo (200-2001).

1352