Studies of juvenile rock lobsters (Panulirus cygnus) in the nursery reefs of Western Australia indicated that density-dependent processes on the reefs were the major determinant of year-class strengths of recruitment to the fishery. Reassessment of early data, plus additional information, indicates that these estimates may have been inaccurate. Alternative methods of estimating juvenile densities were examined; application of length–frequency analysis to calculate the age-classes; visual estimates, and single-night trapping. Length–frequency analysis indicated that either immigration onto the test reefs had occurred, or the data were inaccurate. Visual estimates and single-night trapping were more rapid and used less manpower and resources than mark–recapture methods, but the estimates were of unknown accuracy. No truly satisfactory method of making reliable estimates of either the density of natural mortality rate of juveniles in the nursery reefs is yet available. Nevertheless, the data on the levels of larval settlement and subsequent catches, suggest that density dependent processes probably have a significant effect on the levels of recruitment to the fishery, although the upper level to this recruitment has not been reached over the past 20 yrs.