An important hypothesis about why people generally interact with people who are socially or culturally similar to themselves is that the opportunities they have to meet similar others are greater than the opportunities they have to meet dissimilar others. We examine this supply-side perspective on social relationships by empirically linking marriage choices to the type of setting couples had in common before they married. We focus on five meeting settings (work, school, the neighborhood, common family networks, and voluntary associations) and five types of homogamy (with respect to age, education, class destinations, class origins, and religious background). Using data from face-to-face interviews among married and cohabiting couples in the Netherlands, we show that these five contexts account for a sizable portion of the places where partners have met Using loglinear analyses, we subsequently examine whether couples who shared settings are more homogamous than couples who did not share a setting Our results indicate that schools promote most forms of homogamy, while work places only promote homogamy with respect to class destinations. Neighborhoods and common family networks promote religious homogamy, but they are not related to homogamy with respect to class origins. While in some cases, settings have unexpected effects on marriage choice, our findings generally confirm the notion that mating requires meeting, the pool of available interaction partners is shaped by various institutionally organized arrangements and these constrain the type of people with whom we form personal relationships.