This paper gives a linear practice effect solution for the so-called "Case III" of equating, in which two equally numerous random groups, alpha and beta, take two forms, X and Y, of a test, alpha in the order X, Y, and beta in the order Y, X. One must obviously make some sort of assumption as to the character of the practice (fatigue) effect that the first testing has upon the second. The two cases studied in this paper are (a) practice effect independent of initial score; and (b) practice effect a linear function of initial score. Some of the currently used estimates (constant practice effect) are derived, and generalizations of these estimates to the case of linear practice effect are developed. The paper concludes with some illustrative examples.