Kaufman and Rock (1962) and Rock and Kaufman (1962) concluded that the moon illusion is a function of and attributable to apparent distance. They also reported a large framing effect as an exception. Analysis of the effect suggests two components which can account for the illusion independently of apparent distance. These are apparent size contrasts of visual images of discriminable features or objects of the earth with the moon's image and size constancy of the features or objects plus the interactions of the two. Apparent distances to horizons are always a consequence of the necessary conditions for the illusion. They are related to the illusion but are not a determinant of it.