Due to difficulty to observing soil predator and prey under field conditions, their interactions with each other remain largely ignored. Those studies which have been published are usually non-quantitative and in the case of feeding studies rely heavily on gut content analyses (e.g. Sunderland, 1975 and Holopainen & Helenius, 1992). The feeding behaviour of predators has often been determined in the laboratory rather than in the field (Zhao et al., 1990 and Chaabane et al., 1993).
Quantitative information on the predator - prey interactions in the soil is necessary if effective pest management systems for soil insect pests are to be develop (Edwards et al., 1988). Several studies have indicated that plant protection measures have considerable effects on the population density of earthworms, carabids, staphylinids and other soil organisms (e.g. Nelemans, 1987, 1988; Loreau, 1988; Luff & Rushton, 1988; IPPG, 1989; Luff et al. 1989; Good & Giller, 1991; Cardwell et al, 1994 and Frambs, 1994).
Predation is process that involves interactions between prey defences and the foraging tactics of predators (Godfray & Pacala, 1992 and Malcolm, 1992) In studies of ground surface predator - prey interctions, it is commonplace to relate feeding and foraging behaviour simply by gut content analyses and pitfall trap data, but in the case of below-ground predation, this is more difficult.
Sunderland (1975) found solid food items in the guts of 7 species of carabids, 2 species of staphylinids and one species of lithobiid centipede, but many of the carabids only ingested the fluid contents of their prey. He also found that 15 species of staphylinids had no trace of solid remains in the guts.
Edwards et al. (1979) studied the effect of predation by carabids and othe polyphagous predators on the population of cereal aphids. Loughridge & Luff (1983) tested one species of carabid (Harpalus rufipes Degeer) in the laboratory and the field to see if it fed on aphids. Sunderland et al. (1987) studied a total of 7781 predators belonging to 105 species, and 81 species were found to have eaten aphids. They also showed that staphylinids (both larvae and adults) digested their aphid meal much more rapidly than did carabids and spiders.
Pitfall trapping is useful technique for the study of diverse assemblages of epigeal organisms. Modifications and adaptation made to pitfall traps have been reviewed by Southwood (1978) and Spence & Niemala, 1994. The interpretation of ecological information obtained by these traps has been extensively described. Many authors have suggested that differential species activity, weather conditions and trap placement influence results to such an extent that ecological inferences are difficult to make. Nevertheless, studies are being made which aim to improve the quantitative estimates of carabid populations using pitfall techniques (Clark et al, 1995).
Proponents of the method favour pitfall trapping because it allows continuous unattended sampling over a 24 hour period, and extended trapping equalises the effects of weather. Pitfall trap also permit several sites to be studied simultaneously and the number of animals caught is independent of the sampler's skill. The interaction between predator and prey in the soil cannot be measured by pitfall trapping alone and the reservations expressed by the opponents of pitfall traps should always be borne in mind.
Direct observation of the soil and the events occurring in it can be made using by rhizotrons. These can be of two types; minirhizotrons which use a series of glass tubes inserted into the soil, down which television monitored endoscopes can be inserted to view the soil profile; and large fixed rhizotrons, which are buried chambers with windows along the sides looking out into the soil profile (Carpenter et al. 1985).
The rhizotron used in the present study has previously been described by Sackville-Hamilton et al. (1991). It was designed to allow observations to be made on root herbivores and predator-prey interactions in the soil at different depths.
The main problem with rhizotron observations is that the population of predatory soil anthropods are patchily distributed, the predation process is quick and the number of predators seen in the rhizotron at any one time is small. These problems are even more acute for the users of minirhizotrons, where the area viewed is small, and the observation periods are short. These problems can be minimised by introducing baits to attract the soil fauna and by using time lapse cinematography, to scan long time periods during which little may be happening.
Predatory arthropods are typically abudant in the turf habitat (Cockfield & Potter, 1985), but their importance in regulating densities of pest arthropods in the soil is poorly documented. The predator population is rich in genera and species of beetle families Carabidae and Staphylinidae. Very many studies, recent examples being Wiedenmann et.al. (1992), Fan et.al. (1993), Clark et.al. (1994), Lys (1995) and Tolonen (1995) stress that the epigeal predator fauna is dominated by carabids.
