The use of flowers by butterflies foraging in cereal field margins.
Agriculture is the single largest category of land-use in lowland Britain, and covers about 80% of the total land surface. Although a man-made landscape, modern farmland exists within a framework of natural and semi-natural habitats. However, the ecology of the majority of organisms which do not directly impinge on the efficiency of crop production is poorly understood.
Farmland is generally regarded as a poor wildlife habitat for butterflies (Thomas, 1983; 1984). Little data exist on non-pest species which live in farmland habitats; for example, the National Butterfly Monitoring Scheme (Pollard et al., 1975; 1986) covers some 96 sites in Britain of which only 8% are on farmland, the majority of the remainder being atypical species-rich habitats such as National Nature Reserves.
Agricultural production has undergone major changes in management practices over the last 150 years, coinciding with a general reduction in butterfly species diversity and abundance in Britain (Heath et al., 1984). These changes include increased mechanisation together with associated habitat destruction, and an increase in the use of inorganic fertilizers and synthetic pesticides.
Recent research (Boatman & Sotherton, 1988; Rands, 1985 1986, in press; Sotherton et al., 1987) has focused on the identification and amelioration of the effects of pesticides on farmland wildlife by reducing pesticide inputs on the outermost 6m of cereal field margins, the "conservation headland" technique (Sotherton et al., 1989), and screening agrochemicals for non-target activity. The permitted pesticide applications under the conservation headland concept were outlined by Boatman & Sotherton (1988), the principal impact being an increase in broadleaved weeds and their associated phytophagous and predatory invertebrate faunas following the cessation of the use of insecticides and broad-spectrum residual herbicides (Sotherton et al., 1985).
Conservation headlands have been demonstrated to have beneficial effects on gamebirds (Rands, 1985, 1986, in press), small mammals (Tew, 1988), rare arable weeds (Wilson, 1988) and butterflies (Rands & Sotherton, 1986). The last study, using a modified Pollard walk (Pollard et al., 1975, 1986), showed that more butterflies could be found in fields with conservation headlands than in fields which had equivalent habitat characteristics, but had their field margins fully sprayed with the same complement of pesticides as the main body of the field.
Observations of butterfly behaviour were made to examine in more detail the reasons underlying the increased numbers of butterflies seen in fields with conservation headlands. Data on the exploitation of flowers during nectaring by five common butterfly species in field margins are presented here.