ABSTRACT. Many common bird species have declined as a result of agricultural intensification and this could be mitigated by organic farming. We paired sites for habitat and geographical location on organic and nonorganic farms in Ontario, Canada to test a priori predictions of effects on birds overall, 9 guilds and 22 species in relation to candidate models for farming practices (13 variables), local habitat features (12 variables), or habitat features that influence susceptibility to predation. We found that: (1) Overall bird abundance, but not richness, was significantly (p < 0.05) higher on organic sites (mean 43.1 individuals per site) than nonorganic sites (35.8 individuals per site). Significantly more species of birds were observed for five guilds, including primary grassland birds, on organic vs. nonorganic sites. No guild had higher richness or abundance on nonorganic farms; (2) Farming practice models were the best (∆AIC < 4) for abundance of birds overall, primary grassland bird richness, sallier aerial insectivore richness and abundance, and abundance of ground nesters; (3) Habitat models were the best for overall richness, Neotropical migrant abundance, richness and abundance of Ontario-USA-Mexico (short-distance) migrants and resident richness; (4) Predation models were the best for richness of secondary grassland birds and ground feeders; (5) A combination of variables from the model types were best for richness or abundance overall, 13 of 18 guilds (richness and abundance) and 16 of 22 species analyzed. Five of 10 farming practice variables (including herbicide use, organic farm type) and 9 of 13 habitat variables (including hedgerow length, proportion of hay) were significant in best models. Risk modeling indicated that herbicide use could decrease primary grassland birds by one species (35% decline from 3.4 to 2.3 species) per site. Organic farming could benefit species of conservation concern by 49% (an increase from 7.6 to 11.4 grassland birds). An addition of 63 m of hedgerow could increase abundance and richness of short distance migrants by 50% (3.0 to 4.8 and 1.3 to 2.0, respectively). Increasing the proportion of hay on nonorganic farms to 50% could increase abundance of primary grassland bird by 40% (6.7 to 9.4). Our results provide support for alternative farmland designs and agricultural management systems that could enhance select bird species in farmland. RÉSUMÉ. De nombreuses espèces d'oiseaux ont décliné en raison de l'intensification des pratiques agricoles, mais ces déclins pourraient être atténués grâce à l'agriculture biologique. Nous avons apparié des sites (habitat et situation géographique) sur des fermes biologiques et des fermes non biologiques en Ontario, Canada, afin de tester des prévisions a priori d'effets sur l'avifaune dans son ensemble, 9 guildes et 22 espèces d'oiseaux, en relation avec des modèles candidats testant les pratiques agricoles (13 variables), les caractéristiques locales de l'habitat (12 variables) ou les caractéristiques de l'habitat qui influencent...