Delta–wing theory, which predicts the aerodynamics of aircraft like the Concorde, is the conventional explanation for the way in which a bird's tail operates in flight. Recently, doubt has been cast on the validity of applying a theory devised for supersonic aircraft to the small tails of slow–flying birds. By testing delta–wing models and birds' tails behind bodies with wings, I empirically show that the tails of birds produce lift in a very similar way to conventional delta–wing models. Both Perspex and birds' tail models produce lift similar to that predicted by delta–wing theory when narrowly spread and at low angles of attack. However, when widely spread and at high angles of attack, both tails and Perspex models produce much less lift than predicted, owing to vortex breakdown after which the assumptions of delta–wing theory are violated. These results indicate that birds' tails can be regarded as delta wings but that the theory predicting the forces produced by delta wings can only be applied within acceptable limits (i.e. tails spread less than 60° and at angles of attack of less than 20°).