Background: The morbidity and socioeconomic effects of onchocerciasis, a parasitic disease that is primarily endemic in sub-Saharan Africa, have motivated large morbidity and transmission control programmes. Annual community�directed ivermectin treatment has substantially reduced prevalence. Elimination requires intensified efforts, including more efficacious treatments. We compared parasitological efficacy and safety of moxidectin and ivermectin. Methods: This double-blind, parallel group, superiority trial was done in four sites in Ghana, Liberia, and the Democratic Republic of the Congo. We enrolled participants (aged ≥12 years) with at least 10 Onchocerca volvulus microfilariae per mg skin who were not co-infected with Loa loa or lymphatic filariasis microfilaraemic. Participants were randomly allocated, stratified by sex and level of infection, to receive a single oral dose of 8 mg moxidectin or 150 µg/kg ivermectin as overencapsulated oral tablets. The primary efficacy outcome was skin microfilariae density 12 months post treatment. We used a mixed-effects model to test the hypothesis that the primary efficacy outcome in the moxidectin group was 50% or less than that in the ivermectin group. The primary efficacy analysis population were all participants who received the study drug and completed 12-month follow-up (modified intention to treat). This study is registered with ClinicalTrials.gov, number NCT00790998. Findings: Between April 22, 2009, and Jan 23, 2011, we enrolled and allocated 998 participants to moxidectin and 501 participants to ivermectin. 978 received moxidectin and 494 ivermectin, of which 947 and 480 were included in primary efficacy outcome analyses. At 12 months, skin microfilarial density (microfilariae per mg of skin) was lower in the moxidectin group (adjusted geometric mean 0·6 [95% CI 0·3–1·0]) than in the ivermectin group (4·5 [3·5–5·9]; difference 3·9 [3·2–4·9], p<0·0001; treatment difference 86%). Mazzotti (ie, efficacy-related) reactions occurred in 967 (99%) of 978 moxidectin-treated participants and in 478 (97%) of 494 ivermectin-treated participants, including ocular reactions (moxidectin 113 [12%] participants and ivermectin 47 [10%] participants), laboratory reactions (788 [81%] and 415 [84%]), and clinical reactions (944 [97%] and 446 [90%]). No serious adverse events were considered to be related to treatment. Interpretation: Skin microfilarial loads (ie, parasite transmission reservoir) are lower after moxidectin treatment than after ivermectin treatment. Moxidectin would therefore be expected to reduce parasite transmission between treatment rounds more than ivermectin could, thus accelerating progress towards elimination.