Abstract
Typhlodromus (Anthoseius) recki feeds on pest mites on tomato plants and its introduction into crops via companion plants, Mentha suaveolens and Phlomis fruticosa, has been recently investigated. This study aims at assessing the predator arrestment behavior, through lab choice tests to determine the effects of (i) prey (Aculops lycopersici and Tetranychus urticae) vs Typha angustifolia pollen deposited on companion plant or Solanum nigrum, (ii) T. urticae vs A. lycopersici on S. nigrum and (ii) M. suaveolens vs P. fruticosa using pollen as food source. The predator position, the number of eggs laid and the number of T. urticae eggs consumed were assessed during 4 days. The predator was more observed on leaf discs with A. lycopersici vs pollen when the pollen was deposited on S. nigrum and M. suaveolens, but was more found on leaf discs with pollen on P. fruticosa. It was equally found on leaf discs with T. urticae vs pollen for all the plants tested. No difference in predator arrestment behavior was noted between A. lycopersici and T. urticae, but a higher proportion was observed on leaf discs with A. lycopersici at the beginning of the experiment than at the end. Finally, the predator was more observed on leaf discs of P. fruticosa than of M. suaveolens. The highest fecundity was observed in systems with pollen and A. lycopersci + pollen, and the lowest with T. urticae + pollen; T. urticae consumption was not different with pollen or A. lycopersici. The predator less arrested on pollen vs A. lycopersici that provides a high fecundity; however, surprisingly it was more attracted to pollen vs a prey T. urticae, questioning the cues perceived by T. (A.) recki. The strong preference for P. fruticosa without prey vs S. nigrum with prey or M. suaveolens with pollen, suggests for the second time for Phytoseiidae, direct effects of plant (may be via odors) and questions the ecological advantages of such a behavior. For practical issues, M. suaveolens would be a better bank plant than P. fruticosa; an arrestment on patches of A. lycopersici within tomato plant is expected even if pollen is present and in case of T. urticae infestation a low dispersal to this prey will probably occur, questioning the interest to use T. (A.) recki in combination with other Phytoseiidae species, specialized in T. urticae predation, as Phytoseiulus persimilis.