The locomotion of females of different strains of the parasitic wasp Trichogramma maidis (Hymenoptera, Trichogrammatidae) was measured under standardized conditions and compared to their efficiency in the field for several years. A manual and an automatic method for recording and analysing the locomotion of insects are described and results for differently reared laboratory strains of T. maidis are presented. The correlation between manual and automatic measurements of the travel speed was significant (r = 0.999). The automatic method was developed by means of an Image Processing System. T. maidis females were individually released in an arena, and the walking tracks were recorded by a video camera. To reconstruct the real path, a mathematical model which gives highly accurate path length estimations -termed arcus interpolation -was used. Furthermore the same programme permits the calculation of parameters like turning rate and angles, the activity pattern over the whole observation period or for selected parts of it. The locomotion of different T. maidis strains in the laboratory was compared with their efficiency in the field. An obvious relation between locomotion and parasitization in the field was apparent. The probability for parasitism of Ostrinia nubilalis Hbn.(Lepidoptera, Pyralidae) egg batches decreased linearly with increasing distance from the release point of Trichogramma wasps. The travel speed of the females of different T. maidis strains was found to be positively related to the potential for parasitism in the field. The travel speed can therefore be used as a parameter to estimate the capacity for host location and the efficiency of a T. maidis strain for inundative biological control programmes.