The present paper analyzes the relation between the measurement of subtle and blatant prejudice proposed by Pettigrew and Meertens in 1995 and the tendency to give socially desirable responses. It also tests whether items that measure subtle prejudice are judged as more socially desirable than those that measure blatant prejudice. Data were obtained from two groups, one of 497 Italian high school students and one of 77 university students. In the first case, the analysis concerns the relation between the prejudice scores and scores on a shortened form of Marlowe and Crowne's Social Desirability Scale. In the second case, we analyzed the social desirability judgments expressed on single items of the Petrigrew and Meertens scales. Analyses indicate that (1) neither Subtle nor Blatant Prejudice scores correlate with the tendency to give socially desirable responses and (2) when the items of the two prejudice scales are placed in order on the social desirability continuum, with very few exceptions the Blatant Prejudice items are situated at the not socially acceptable pole and Subtle Prejudice items at the socially acceptable pole.