A model is presented for analysis of mark-recapture data of mobile insects which, unlike the Lincoln Index, does not require marked individuals to remain within the sampling area or to mix uniformly with the wild population. The model assumes a single or multiple releases of marked insects from the centre of the sampling area and that captured individuals are not returned to the population. Dispersal rates of marked insects are estimable from serial recaptures and, for catches that are either unaffected by or have been corrected for weather effects, the model also provides estimates of mortality and age-dependent trappability. Application of the model is illustrated using mark-recapture data for adults of the Australian sheep blowfly Lucilia cuprina. 142 absolute densities of L. cuprina proved to be much more difficult. Multiple marking and release of field adults was abandoned for a variety of reasons (Whitten et al., 1977) in favour of single releases of laboratory-reared flies marked with fluorescent dust at emergence (Norris, 1957). The latter procedure is simpler and enables larger numbers of marked flies to be liberated into the sampling area. Thus, of the available methods for estimating population size from mark-recapture data (Seber, 1973;Southwood, 1978), the Lincoln Index (Lincoln, 1930) is the most appropriate but, in common with trapping data for Other blowflies, the mark-recapture data for L. cuprina violates several assumptions of the model (MacLeod, 1958). Spatial distributions of L. cuprina are highly aggregated (Vogt, et al., 1983) and the released flies do not mix uniformly with the wild population prior to reaching the boundaries of the sampling area. Additionally, because carrion-baited traps mimic feeding and oviposition sites, probability of capture is likely to vary in relation to sex and cyclic changes in protein requirements for gonadotrophic development (Roberts and Kitching, 1974;Kitching and Roberts, 1975).Without uniform mixing and equal trappability, changes in ratios of marked to unmarked flies do not allow correction of population estimates for losses of marked flies (c.f., Fletcher et al., 1981;Readshaw, 1982). We therefore developed an alternative model for adjusting catches of marked flies for dispersal, based on their spatial distribution within the sampling area, that does not require the assumption of uniform mixing. Serial catches of marked flies corrected for dispersal losses provide estimates of fly density which, in turn, enable age-dependent changes in survival and trappability to be estimated from the mark-recapture data. Here we present a formal description of the model and illustrate its application to analysis of mark-recapture data for adults of the Australian sheep blowfly L. cuprina.
METHODS
Trapping, Rearing and Release ProceduresMarked adults were liberated at the centre of the sampling area, approximately 2 km due south of the village of Murrumbateman, New South Wales (Fig. 1). Marked and wild flies were sampled using West Australian blowfly traps (Vogt and Ha...