The concept of location—of something being located at, or occupying, a region or a place—has been an important topic of philosophical investigation over the past fifteen years or so. Yet all the theories of location that have been put forward so far are unsatisfactory, because they fail to have the conceptual resources to describe certain basic locational phenomena. I introduce and partly develop a novel theory of location that does better in this respect than its predecessors. Its most distinctive feature is the way in which it handles so-called cases of multilocation: according to the theory, objects capable of being multilocated occupy regions via proxies, and—in a slogan, and very roughly—to be multilocated is to have several proxies located at different regions.