The nature and determinants of political tolerance have been a major focus of theory and research in political science for the last 50 years. Typically, empirical research on intolerance has examined peoples' willingness to deny civil liberties to particular groups. In the 1950s and 1960s, it was common to examine attitudes towards communists and other left-wing groups. By the 1970s, research was demonstrating that an exclusive focus on left-wing groups (or any other specific groups) underestimated intolerance for many people. A new methodology - the least-liked group approach - was developed to assess intolerance with dislike toward the group held (relatively) constant. This approach is used in much recent research. Whether all people react to the same group or to a disliked group of their choosing, this group-centered approach ignores the substantial, positive correlations of intolerance across groups. Although there is still a considerable group-specific component, there also is a strong tendency for those who are intolerant of one group to be intolerant of many others. This inter-group consistency in intolerance is important because it provides another perspective on the phenomenon and its determinants. What accounts for these individual differences in intolerance? The answer to this question, I argue in this resarch, is a fundamental conflict between the values of social conformity and personal autonomy. This value conflict arises as people deal with the problem of how social order should be maintained. Does a stable social order emerge from the free interactions of autonomous individuals, or must we require people to conform to a common set of social values and conventions? This dimension of social conformity-autonomy is evident in major cross-national studies of values and should be a central component of peoples' beliefs about politics and the nature of political debate in society. In addition to providing a theoretically based analysis of stable individual differences in intolerance, this approach yields new insights about the nature and impact of perceived threat. Recent research shows that perceived threat is one of the major predictors of intolerance but it is not entirely clear what perceived threat is. It is either measured very broadly (by a series of paired adjectives describing the group) or operationalized as threat or belligerence. I argue that another aspect of perceived threat is threat to the stability of the social and political order. Moreover, this dimension of threat should be particularly salient to those who value social conformity more than autonomy. There, thus, should be a strong statistical interaction between the social conformity-autonomy dimension and perceived threats to the social and political order. The research proposed here tests the predictions of this new perspective in four experimental questionnaire studies, two with college students and two with non-random samples of adults from the New York metropolitan area. The two student studies concentrate on the measurement of the social conformity-autonomy dimension and demonstrations of its effects on a wide range of political beliefs and attitudes. The student studies also pretest experimental materials for the adult studies. The two adult studies further validate measures of social conformity-autonomy in more representative samples and test hypotheses about the effects of threat on intolerance. Using constructed news stories about political groups, the first study contrasts threats to the political order with threats of violence. The second study examines the joint effects of two types of threat to the political order: threats directly from a political group and more general threats from increasing social and cultural diversity. The two experimental studies provide a closer look at the nature of political threat and the role of social conformity-autonomy in moderating the perception and effects of threat.