Four pigeons were trained to match reinforcement probability of two discs illumined with different degrees of brightness. After Ss had reached the prescribed experimental criterion, the dimmer of the two discs was now made the brighter of the two. In this way, transposition of the probability matching response was investigated. All animals showed this transposition. It was concluded that the probability matching response in pigeons was a fairly stable response, and that it could be transposed to other levels of discrimination, especially when degrees of brightness were used as the discrimination problem.