LABORATORY EVALUATION OF FOOD PLACEMENT AND FOOD TYPES ON THE FEEDINGPREFERENCE OF RETICULITERMES-VIRGINICUS (ISOPTERA, RHINOTERMITIDAE)

Two pieces of cardboard or filter paper were placed in a paired or spl it design in a choice test to examine the effect of food placement on termite food choice. When cardboard pieces were presented within an ex perimental unit to Reticulitermes virginicus Banks, one piece was sign ificantly consumed more than the other in experimental units for 24 an d 48 h in the paired design. In the filter paper treatments, significa nt differences in consumption on pieces in the split design were detec ted at 48 and 72 h. Bioassays that result in significant differences w hen food types are the same may lead to spurious conclusions about pre ference when food types that are not equivalent are tested. Although t he paired design may overcome the tendency for termites to feed on the Ist food source encountered, the split design was a more sensitive bi oassay for detecting consumption differences between cardboard and fil ter paper under these conditions. Feeding behavior under our condition s was affected by food placement, indicating that termites did not fee d randomly. Food type also affected feeding behavior. Because termites do not feed randomly regardless of whether the paired or the split de sign is used, replicating control experimental units sufficiently to o vercome significant differences in consumption when food types are the same is paramount.