Existing knowledge of point defects and their interactions in uncolored alkali halide crystals is briefly reviewed. Attention is drawn to recent experimental and theoretical work which indicates that vacancy pairs arenothighly mobile defects. This has important implications for some hypotheses on the kinetics of color center reactions at low temperatures.This knowledge of defects in uncolored crystals is then applied to a discussion of the reactions which may take place betweenFcenters and impurity ions in additively‐colored doped crystals and the theoretical possibilities are compared with experimental information onZcenters. Consideration of the thermodynamic equilibria in these reactions seems to eliminate theZ1andZ2centers of Pick as possible causes of either of theZ1andZ2absorption bands in KCl. (TheZ2model of Pick is identical with theZ1model of Seitz and is a divalent impurity cation which has trapped an electron.) Two possible centers for theZ2band presen tthemselves; (a) an associate of anFcenter with a divalent cation and (b) an associate of anFcenter with an impurity‐vacancy pair (theZ2model of Seitz). The structure of theZ1center is not resolved by these considerations. Attention is drawn to some paradoxical experimental results onZ1andZ2centers and further experiments to resolve outstanding questions are suggested.