The natural arrival of rabbit haemorrhagic disease virus (RHDV) in south-western Australia in September 1996 resulted in a reduction in rabbit numbers of ~65% (~90% morbidity, with ~72% mortality of infected rabbits). As no signs of the disease (clinical or serological) were seen over the next two years, and as rabbit numbers over the last 12-month monitoring period at the site were similar to those observed before the natural 1996 RHDV epizootic (i.e. pre-RHD), RHDV was deliberately reintroduced into this rabbit population in April 1999 (autumn). Seven RHDV-inoculated rabbits were released prior to the main breeding season when <3% of sampled rabbits (n = 118) were seropositive for RHDV antibodies. Following the deliberate release, the overall decline in rabbit numbers (68%) was comparable to that seen during the natural 1996 epizootic. However, on the basis of the observed changes in rabbit numbers, and in their serology, the impact of the deliberate RHDV release appeared to be more variable across the six trapping areas than was seen during the natural 1996 spring epizootic. The reductions in rabbit numbers on these areas 6–8 weeks after RHDV-release ranged from 55% to 90%. The serology of the surviving rabbits on the trapping areas was also variable over this period, with the proportion of seropositive rabbits ranging from 5% to 90%. Overall, only 15% of the surviving rabbit population showed evidence of recent challenge by RHDV, giving a morbidity rate of 73% 8 weeks after the release. However, over 90% of infected rabbits died. This provides further evidence that some rabbits remained un-challenged by RHDV for up to 8 weeks after its release. The variable impact of the April 1999 release may have been partially caused by the observed differences in abundance of insect vectors, and/or an apparent increase in the incidence of non-virulent RHDV in the months preceding the release.