( 1 ) Agreed with Franks (1964) that a replication should indeed replicate (as nearly as possible). ( 2 ) Agreed that the Ludvigson (1964) srudy was not a "precise replication" of earlier studies on drugs and eyelid conditioning (e.g., Franks & Trouton, 1958) though it was similar. ( 3 ) The Ludvigson study was not intended to replicate the conditions of the earlier work. If there was any implication that it was so designed it was an unfortunate consequence of the limitations of brief reporting. The author's intention was to point out that his results did not agree with those of the earlier studies in spite of the fact that the several studies used procedures which, albeit different in certain respects, had all been considered operational definitions of "conditioning." The empirical import of the srudy was evidence bearing not upon the reliability of drug effects under a given set of conditions (which a replication would provide), but rather upon the generality of drug effects across experimental conditions. Thus the study actually provides Franks with empirical support for his apparent concern wirh the narrowness of such generality. ( 4 ) The rationale, in brief, for the Ludvigson procedures follows: According to drive ( D ) , theory (e.g., Spence, 1958) D is assumed to be an increasing function of the strength of an emotional response ( r . ) that is aroused by aversive stimulation, such as that employed in eyelid conditioning. Any drugs which alter r, should affect conditioning performance, and in the same manner (e.g., divergent acquisition functions) as other drive variables. Work on the conditioned avoidance and emotional responses suggests that at leasr some of the tranquilizers lower emotional responsiveness while central stimulants, which tend to have behavioral effects opposite to those of the tranquilizers, enhance such responsiveness. The Ludvigson study employed procedures comparable to those of many studies from the same laboratory which have provided much of the empirical base of drive theory. The fact that in the Ludvigson study the drugs did not significantly influence conditioning, can only be taken as a failure to confirm the predictions from Spence's emotionally-based drive theory, assuming the dosages employed were sufficient to alter emotionality. Incidentally, this is a good example of an Iowa study which does not support Iowan theory (Franks, 1964) . ( 5 ) Although the author is in sympathy wirh the suggestion of a trip to England, he feels confident that he can fail to replicate his own findings in this country.