Two experiments investigated the effects of familiar size on size and distance estimates in natural outdoor settings. In Experiment 1, the subjects compared perceived size and distance of a normal-sized, undersized, or oversized photographic model of a woman with those of the unfamiliar board of the same size as the model. In Experiment 2, the subjects reported perceived size and distance of a real woman and an unfamiliar board at a viewing distance of 40, 80, or 160 m. The obtained main results were: (1) The size estimates of off-sized models partially approached the normal size, (2) the size estimates of a real woman were constant over the range of distance (size constancy), whereas the size estimates of an unfamiliar board increased with increasing distance (over-constancy), and (3) the undersized and normal-sized model were judged farther than the corresponding boards, but the oversized model was judged at the same distance as the board. These results were compared with the predictions from the size-distance invariance hypothesis and the off sized perception hypothesis.