Abstract

Gill-net mesh selectivity parameters were estimated for C. obscurus and F. macki from catches of an experimental net composed of seven panels of mesh ranging in size from 12.7 cm to 21.6 cm. A maximum-likelihood model was used to fit a gamma distribution to length data from each of the different-sized meshes. Mesh selectivity parameters for the gilled-only data set were θ₁ =126.90 and θ₁ =20253, and for the data set including gilled and non-gilled sharks θ₁ =130.13 and θ₁ =29237. The advantages of the parameters from each of these data sets are discussed. Mesh-selectivity parameters for F. macki were θ₁ =173.70 and θ₂ =26415. The exclusion of data from the two largest mesh sizes, which had much lower catch rates than the other mesh sizes, did not significantly change the parameter estim- ates. Catch rates from the panels in the experimental net that correspond to those used by the commercial gill-net fishery in south-western Australia were compared and no significant differences were found.