Likelihood of marijuana-related police stop/search or arrest depends on many factors other than simply engaging in marijuana-related activity. Police are assumed to suspect individuals of marijuana-related offenses based on several personal characteristics, including ethnicity, age, gender, age, educational level, and subculture. An individual's likelihood of marijuana-related police contact is hypothesized to depend on how strongly this suspicion, the "police gaze," falls on them, independently of their actual participation in public marijuana use. A diverse, street-recruited, purposive sample of 462 marijuana users in New York City completed questionnaires for this study. Several factors, including racial minority status, neighborhood in which the participant was recruited, gender, unemployed/non-student status, youth, and lower educational level were found to be simultaneously and independently related to likelihood of marijuana-related police contact even controlling for frequency of use, public use, and observance of etiquette intended to make the behavior less of a nuisance. Etiquette was found, moreover, to be differentially effective based on race, location, and gender: Predicted probability of marijuana-related police contact was roughly 50% for African-Americans, males, and users recruited from Harlem or the South Bronx who observed none of the etiquettes and 10% or less if they followed all four. By contrast, predicted probability of marijuana-related police contact for whites, females, and users recruited from non-poverty areas of Manhattan hovered around or below 10% regardless of etiquette observance. The odds of marijuana-related police contact for Latinos were more than three times the odds for whites. Results bear out that centrality to the "police gaze" dramatically influences an individual's likelihood of marijuana-related police stop/search and arrest independently of whether they engaged in any marijuana-related illegal behavior. ..PAT.-Unpublished Manuscript [ABSTRACT FROM AUTHOR]