Abstract
In order to derive quantitative estimates of predation rate from serological gut analysis data, one must have an estimate of the interval during which a meal can be detected after feeding. In practice this has turned out to be ‘Dmax,’ defined as ‘…the time from finishing a meal until that meal could just no longer be detected in any individuals.’ However Dmax substitutes an absolute limit for what is really a continuous variable with significant variation. We examined this problem in a study of the detectability of Helicoverpa zea Boddie (Lepidoptera: Noctuidae) fifth instar remains in the guts of Polistes metricus Say (Hymenoptera: Vespidae). Wasps were maintained on Trichoplusia ni (Hübner) (Lepidoptera: Noctuidae) fifth instars before being fed a single H. zea fifth instar. They were killed and frozen at 0, 24, 48 and 96 h intervals, with those held for more than 24 h fed a single T. ni fifth instar at 24 h intervals in order to simulate continued feeding. Wasp abdomens were assayed by immunodot, using a monoclonal antibody to H. zea arylphorin. There was a logarithmic decay in the proportion of P. metricus positive over time, a single H. zea fifth instar meal having a detectability half‐life of 19.4 h at field temperatures. If prey antigen detectability decays exponentially, then a detectability half‐life is a more appropriate unit of detectability than an absolute detectability period.