Current glasshouse biological control practice relies on regular prophylactic introductions of one or two 'best' species of natural enemy. Whilst this is effective for much of the time, occasional failures occur due to factors such as differences in response to seasonal changes in environmental conditions and/or host plant effects. This study looks at the predatory behaviour of a specialist coccinellid, Stethorus punctillum Weise, and a generalist mite, Amblyseius californicus McGregor (which predate on the two-spotted spider mite, Tetranychus urticae ) in order to assess how they responded to temperatures and relative humidities typical of glasshouse conditions on four edible crop plant species. Activity (distance covered, time spent walking, walking speed, angular velocity, and turning rate) was recorded at 20, 25 and 30 o C and at relative humidity (RH) levels of 33, 65 and 90%, on tomato, pepper, aubergine and cucumber leaves, and analysed using video-computer techniques. The results show that the activity of S. punctillum significantly increased at higher temperature levels. Host plant species also strongly influenced the performance of the predator, with it being most active on pepper and tomato and least active on aubergines. RH had no significant influence. The activity and predation by A. californicus increased at low humidity levels, especially in terms of time spent moving and number of prey killed. Temperature levels had no significant influence, but host plant species strongly influenced the performance of the predator, which was most active on pepper, and least active on aubergines. Further research was conducted with semi-field trials to investigate the efficacy in controlling TSSM with different combination of predators. When contrasting the commercially available predatory mite Phytoseiulus persimilis , used alone, compared with its use in a treatment with a combination of predator species, there was a stronger decrease in TSSM numbers on the crop plants in the latter treatment.