The aim of this paper is to critically consider how the concept of «industrial district» was born and evolved in the field of regional science. Despite the claim by Isard that the emergence of a spatial dimension in economics was being hampered by Marshall’s alleged prioritising of time over space (Isard, 1956), the concept of «localization economies» introduced into regional science by Ohlin, Hoover and Isard clearly is itself a legacy from Marshall. This contradiction in the work of Isard and others followed to a large extent from the way in which Marshallian concepts were historically situated in economic thought before Giacomo Becattini’s re-reading of Marshall and his ideas. This re-reading began in the 1960s, focussing on conceptual issues related to industry (Becattini, 1962), and then culminated in the following decades with work on the idea of the «industrial district»: considering it first as a unit of investigation of economic research (Becattini, 1979) and, later, as a way of industrial development (Becattini, 1989). In brief, the originality of Marshall’s economic thought can be found in Book IV of his Principles of Economics. This originality can be seen in the statement that economics is more important as a way of studying man in society than as a way of studying wealth; and continues through the affirmation that man’s character is moulded by his daily work. That is to say, a person’s «place of living» (where individuals live and work) is important not only for the formation of human skills but also for the formation of character. The place as the unit of investigation (i.e. the «functional region») is one of the main elements which distinguishes the identity of regional science from other branches of economics.