A number of recent studies have suggested that the "intersection of constraints" model of two dimensional motion perception, put forward by Adelson and Movshon [(1982) Nature, 300, 523-525], is incomplete. Evidence has been mounting that there is a second two-dimensional motion sensitive mechanism which is monocular and which appears to respond directly to the movement of the intersections (or "blobs") in a two-dimensional image. The current study extends these findings by demonstrating that the perceived coherence of a drifting plaid is largely under the control of a monocular mechanism. Prior exposure to a similarly drifting grating or plaid substantially raises the coherence threshold of a test plaid only if the same eye is adapted and tested. The threshold elevation is much more modest if the test plaid is presented to the unadapted eye, suggesting that coherence judgements are primarily based on the activity level of a monocular process--possibly the "blob tracking mechanism". The results of Expt 2 suggest the possibility that this monocular mechanism is inhibited by binocular exposure.