Objectives. Although the contact hypothesis occupies an important place in the study of intergroup relations, several important questions remain understudied. We contribute to the literature by examining the associations between multiple aspects of contact with Latinos and several types of attitudes toward U.S. Latinos, including attitudes toward immigration restrictions. Methods. We use OLS and logistic regression techniques to analyze the opinions of Anglo (non-Hispanic white) and African-American respondents from the 2000 NORC General Social Survey (GSS), which contained a special module on attitudes regarding ethnicity and diversity issues. Results. Our findings reveal a clear association between close nonkin Latino ties, that is, friendship, and more favorable or empathetic attitudes toward Latinos as well as less restrictive or punitive views concerning immigration policy. Having Latino relatives is also linked with several of these attitudes. Most other dimensions of contact (e.g., attending high schools with significant Latino populations, having Latino acquaintances, Spanish-language fluency, and residing in areas with relatively high Latino populations) have at least some association with the attitudes under study, although for most of these contact measures the significance of the effects is inconsistent across outcomes. Conclusions. The findings reinforce the significance of intergroup friendships and underscore the importance of understanding their social origins, patterning, and consequences for diverse groups. It is also important to probe the differential effects of specific types and arenas of intergroup contact on specific attitudinal and policy outcomes.The contact hypothesis has occupied an important place in the literature on race-ethnic intergroup relations for more than 50 years (Allport, 1958;Williams, 1947). The core idea is simple and straightforward: contact, particularly close and sustained contact, with members of other cultural groups provides direct information about the values, lifestyles, and experiences of members of those groups. Information obtained in this way is likely to be more favorable and accurate than information gained through other, less * Direct correspondence (including requests for data and coding for replication purposes) to Christopher G. Ellison,