The work was carried out in 15 villages of the humid forest zone of Southern Cameroon (3°27'-4°10'N and 11°32'-11°49'E) (Figure 1). These villages were selected among the 45 Forest Margin Benchmark Area (FMBA: 1.54 millions hectares) reference villages. Serving as a focal point for strategic diagnostic research in the sub-region, the benchmark approach was developed and implemented through the Eco-regional Program for the Humid and Sub-Humid Tropics of Sub-Saharan Africa (EPHTA). The 45 villages are situated along a gradient of natural resource utilization intensification, represented by 3 agro-ecological, cultural and demographic blocks (AECD-Blocks) namely Yaounde, Mbalmayo and Ebolowa (Figure 1). Five out of 15 villages of each block were selected for the study. These were Nkongmesse, Nkolmelok, Nkometou-II, Akak-II and Etoud for the Yaounde block; Awae, Evindissi, Ngat, Mvoutessi and Nkolmetet for the Mbalmayo block and Minsélé, Mekoe, Mengomo, Akok and Obang-II for the Ebolowa blocks. The 3 AECD-Blocks differ in several aspects. Yaounde has a high human density (14–88 habitant/km2); whereas low human population densities are observed in Ebolowa (2–15 habitant/km2) and intermediates (10–48 habitant/km2) in Mbalmayo . On the agro-ecological level, forest degradation is more pronounced around Yaounde than in the Ebolowa, which still has some pockets of primary forest . Mbalmayo constitutes an intermediate block. Farmers in these areas all practice slash and burn agriculture . In the Yaounde block, the length of fallow is about three years shorter than that of Ebolowa (7.5 years) while that of Mbalmayo block is a transition (5.4 years). The main food crops are banana, plantain, cocoyam, cassava, yam and groundnut . Cucurbits, okra, vegetables and spices are secondary crops  while cocoa and coffee are the main cash crops of this area. Their production is mainly in small-scale, through domestic or family exploitation. These areas have an equatorial climate type with two dry seasons and two rainy seasons. The average annual rainfall is 1510 mm, 1643 mm and 1820 mm in Yaounde, Mbalmayo and Ebolowa respectively. From an ethnological viewpoint, the "Etons", "Ewondoses" and the "Bulus" are the main ethnic groups in the Yaounde, Mbalmayo and Ebolowa blocks respectively. With regard to the diet, the Bulus consume more sauces made based on almond of the fruits of Irvingia gabonensis (named 'Ndo'o') while Etons and Ewondoses consumes more 'Okok' (pasta product made containing leaves of Gnetum africanum) and 'Kwem' (pasta product made containing leaves of Manihot esculenta) respectively (Lema Ngono, Personal Communication).
Information was obtained in each village between August and October 1998 using the rapid rural appraisal methods (RRAM) [23-25]. There were 17 questions, divided in three parts; 5 relating to the annual crops and, 5 others relating to the young perennial crops and 7 general questions. The final questionnaire was prepared based on a preliminary one that was tested in Bikok and Abang, two other villages different from those used in our investigations. Based on the results of this pre-test, we decided to work with men and women farmers' groups separately in each village. We selected farmers growing annual or perennials crops. A total of 30 groups (389 farmers: 164 women and 225 men) were interviewed. Each group had 8 to 21 farmers aged 18 to 50 years. Participants were recruited in 3 strata based on the 3 blocks of the 'Benchmark' [23,24]. Interviews of two hours per group were conducted at different hours of the same day for the two groups. Farmers were asked to rank the importance of Zonocerus variegatus among the other insect pest of the humid forest zone of the Southern Cameroon and rate the impact of Z. variegatus on crops in the humid forest zone of Southern Cameroon. Farmers were also asked to state factors, which influence the pest status of Z. variegatus, the methods used against the pest and to state what they need to suppress the pest population. In the questionnaire, we used opened-ended and close-ended questions.
Questions were asked in the local language and/or French. During the survey in each village, a sample of Zonocerus variegatus was shown to the farmers to enable them recognize the insect. For the various rankings, scores (described In ) were used while the farmers answered the other questions directly.
Data were analyzed by the Kruskal-Wallis test using the 'Nonparametric One Way' ('NPAR1WAY WILCOXON ') procedure of the software SAS ('Statistical Analysis Systems' version 8). All probabilities were appreciated at the 5% confidence level.