Management of high-density koala (Phascolarctos cinereus) populations is essential because of the browsing damage they inflict on their habitat. We have tested two types of gestagen implant, namely levonorgestrel and etonogestrel, as contraceptives for koalas. Free-ranging female koalas were given either a control, levonorgestrel (70 mg) or etonogestrel (34 or 68 mg) implant before the breeding season. Koalas were monitored every 4–12 weeks for births. Plasma progesterone was measured and a cytological smear of the urogenital sinus was taken. Fertility was high in the control group and the two etonogestrel-treated groups, with approximately 90% of females giving birth. In contrast, no levonorgestrel-treated female produced young during the study. Removal of levonorgestrel implants from six females reversed the contraceptive effect in the next breeding season, whereas the eight females in which the levonorgestrel implants were left in remained infertile for six breeding seasons. Vaginal cytology showed evidence of oestrous cycles during the breeding season in all females from all groups and there was no difference seen in the prevalence of cornified epithelial cells in the oestrous smears. This indirectly suggests that levonorgestrel does not prevent follicular development and oestrous cycling. Plasma progesterone in levonorgestrel-treated females remained low all year, but rose in controls concurrent with the onset of the breeding season. This suggests that levonorgestrel prevents pregnancy by blocking ovulation. Etonogestrel had absolutely no contraceptive effect at the two doses delivered and so is not suitable for controlling koala populations. In contrast, levonorgestrel was effective as a long-term, reversible contraceptive in wild koalas.