Culicoides biting midges are hematophagous ceratopogonid insects which occur throughout most of the inhabited world, transmitting some disease-causing pathogens of humans, and domestic or wild animals (Muirhead-Thomson, 1982; WHO, 1984). Therefore, literature is replete with information on studies on these insects from different parts of the world (Noireau et al., 1990; Conte et al., 2003; Mercer et al., 2003). To the best of our knowledge, most of the previous Culicoides studies from Nigeria were associated with wild or domestic animals (Dipeolu, 1976a, 1976b; Dipeolu and Ogunrinade, 1977; Udonsi, 1988), and none assessed human biting rates.

Perstans filariasis, caused by Mansonella perstans, occurs in Africa and South America (Heyneman, 2004) and has been reported from many parts of Nigeria. In the northern part of Nigeria, Wijeyaratne et al. (1982) reported 14.6% prevalence of M. perstans from the Malumfashi district. Anosike (1994) reported 1.4% prevalence from the north-western zone of Bauchi State, northern Nigeria. In the eastern part of Nigeria, Arene and Atu (1986) reported 47% prevalence of M. perstans from the Niger Delta area. In the Igwun River basin (eastern Nigeria), Udonsi (1988) reported 13.4% prevalence of M. perstans. In the central part of Nigeria, Ufomadu et al. (1991) reported 6.8% prevalence of M. perstans. In the western part of Nigeria, Oyerinde et al. (1988) reported 0.8% prevalence of M. perstans in a hospital-based filariasis survey in metropolitan Lagos. In a relatively recent study, M. perstans infection (3.2% prevalence) was reported for the first time from Ijebu North area of western Nigeria (Agbolade and Akinboye, 2001).

In view of the veterinary and medical importance of Culicoides, the need for a longitudinal study to understand the biting rates and human perception of the biting midges in Ijebu North area of western Nigeria cannot be over-emphasized since this will enhance establishment of meaningful control strategies. Due to the problem of inaccessibility to many villages in the area, the decision to use one of the villages for case study was taken. The present study aimed at determining the monthly biting rates, filarial infection rate and human perception on the behavior and bites of anthropophilic Culicoides in a rural community endemic for M. perstans in Ijebu North area of western Nigeria.

A total of 359 specimens of Culicoides adults identified as C. fulvithorax were recorded in this study. Monthly biting rates of C. fulvithorax, mean rainfall, relative humidity and temperature in the study area are given in Table 1. The monthly biting rate was highest in September 2004 and lowest in December 2003 (P < 0.001). There was a positive correlation between monthly biting rates and mean rainfall (r = 0.94, P < 0.001). Similarly, a positive correlation occurred between monthly biting rates and mean relative humidity (r = 0.85, P < 0.01). However, there was no correlation between monthly biting rates and mean temperature.

Total 359 specimens of C. fulvithorax were dissected, and the third stage of filarial larvae (L₃) were found in 7 (1.95% of infection rate), which were collected in July, September and October 2004. The arithmetic and geometric mean intensity of the filarial L₃ infections were 1.6 L₃/Culicoides and 1.3 L₃/Culicoides, respectively.

Of 854 villagers interviewed, 325 (38.1%), 122 (14.3%), 62 (7.3%), 90 (10.5%), 29 (3.4%), and 18 (2.1%) referred to anthropophilic Culicoides as 'finfin', 'kotonkan', 'amukuru', 'tintin', 'yanyan' and 'yanmuyanmu', respectively. Two hundred and ninety-two (43.2%), 260 (30.4%), 142 (16.6%), 39 (4.6%), 25 (2.9%), and 12 (1.4%) of the interviewees believed that Culicoides adults were found in bushy area, everywhere, on farm, inside houses, on playing field, and by the stream side, respectively. The interviewees who professed that Culicoides were more prevalent in wet season than in dry season were 69.6%, whereas 19.2% of the interviewees professed that they prevailed in dry season. Three hundred and thirty-seven (39.5%) answered that Culicoides midges enter houses, and among them 85 (25.2%) and 215 (63.8%) answered that they enter in the morning and in the evening, respectively. Among total 854 interviewees, 13.2% (113) claimed that Culicoides got annoyed, 89.0% (760) knew that they were biting insects, 86.5% (739) had experienced of been bitten by them, and 363 (42.5%) believed that anthropophilic Culicoides transmit diseases, such as fever (45.2%), headache (1.4%) and rash (2.2%).

Among the age groups, 76.5% (153), 84.8% (345), 99.1% (216), and 86.1% (25) of 6-11, 12-24, 25-54, and > 55 years, respectively, experienced body reactions to Culicoides bites. These frequency occurrences of body reactions to Culicoides bites were statistically similar among the age groups (P > 0.05). The frequencies of the body reactions are summarized in Fig. 2. Itching was the highest (34.4%) while fever was the lowest (9.1%) (P < 0.001). Among 739 villagers who had experience of body reactions to Culicoides bites, 84.6% (625) usually treated themselves with palm oil, kerosene, pain-relieving ointments or tablets.

The anthropophilic habit of C. fulvithorax in the present study area conformed with the previous report of White (1977). C. fulvithorax had also been associated with domestic animals in Nigeria (Dipeolu, 1976b). In this study, the monthly biting rates of C. fulvithorax, which reached at their peak in September, were generally higher in rainy season than in dry season. Similar observations had been reported on some Nigerian Culicoides species associated with wild animals (Dipeolu, 1976a). The study area has abundance of banana plants, shrubs and decaying plant matters which are known to favour the breeding of Culicoides (Muirhead-Thomson, 1982; Mercer et al., 2003). The greater availability of water from rainfall which moistens soil, rotting banana plants and other plant matters might have contributed to the higher abundance of C. fulvithorax in the rains. It is well known that there is higher occurrence of indiscriminate bush-burning in dry season in Nigeria. This might have contributed to the recorded low abundance of C. fulvithorax in dry season in the study area. Results from the questionnaires administered also revealed that anthropophilic Culicoides are more prevalent in the rainy season in the study area. During some previous visits, anthropophily of C. fulvithorax was observed from some other parts of Ijebu North (Agbolade, personal observation). The presence of anthropophilic C. fulvithorax in both seasons in the study area is noteworthy.

The monthly biting rate of C. fulvithorax increased with relative humidity in this study. This finding corroborated an earlier opinion that relative humidity influenced Culicoides breeding in Nigeria (Dipeolu and Ogunrinade, 1977). However, the monthly biting rate of C. fulvithorax was not correlated with temperature in this study. A previous study showed that temperature and relative humidity, vis-à-vis Culicoides occurrence, are independent variables (Conte et al., 2003).

Infection of some examined C. fulvithorax specimens with filarial L₃ possibly indicates that C. fulvithorax is an important vector of M. perstans in the study area which is endemic for M. perstans (Agbolade and Akinboye, 2001, 2005). However, some other possible Culicoides vector species might have been inadvertently excluded from the catches since only early morning collections were made in this study. Results from questionnaires showed that anthropophilic Culicoides seek blood meal in both morning and evening in the study area and this agrees with established phenomenon (Muirhead-Thomson, 1982). The filarial infection rate of C. fulvithorax in this study is similar to that reported for C. grahamii in the forest region of Congo (Noireau et al., 1990). The occurrence of filarial infection coincided with period of high biting rate in this study. Possible occurrence of M. streptocerca cannot be ruled out in the study area as Culicoides species equally transmit this parasite (Heyneman, 2004). Further studies are needed to establish the filarial species in C. fulvithorax using experimental animals.

The commonest local name of Culicoides in the study area is 'finfin'. However, it is noteworthy that some of the villagers erroneously refer to anthropophilic Culicoides as 'yanmuyanmu', a popular name for mosquitoes in the western part of Nigeria. This indicates serious level of ignorance among the villagers. This is corroborated by the claim of some interviewees that anthropophilic Culicoides make noise when seeking blood meal which has never been recorded as a behavior of Culicoides (Muirhead-Thomson, 1982). Ignorance has been identified as one of the major factors enhancing the endemicity of many parasitic diseases in tropical African countries (Ukoli, 1992). The ignorance of the villagers is further emphasized by the fact that none of the villagers knew Culicoides biting midges as a vector of filarial parasites.

Responses from interviewees showed that most of the villagers have experienced Culicoides bites several times with occurrence of noticeable body reactions among all the age groups. The most frequent body reactions to Culicoides bites in the study area was body itching which was often coupled with swelling. The other reactions were pain and fever. These reactions are attributable to inflammatory response (Ward, 1980). The treatment of victims necessitated by the side reactions of Culicoides bites demonstrated the negative socio-economic impact. This is because the victims had to spend substantial proportion of their meager income on treatment.

This study has shown that the presence of anthropophilic C. fulvithorax in both rainy and dry seasons in the study area constitutes serious threat to the socioeconomic lives of the villagers. Unfortunately, there is no Culicoides control program in Nigeria. Therefore, both health education and application of effective control measures against Culicoides biting midges are urgently required for health promotion in Ijebu North area of western Nigeria.