The purpose of this paper is to explain the timing and location of the diffusion of heavy metal music. We use data from an Internet archive to measure the population-adjusted rate of metal band foundings in 150 countries for the 1991-2008 period. We hypothesize that growth in "digital capacity" (Internet and personal computer use) catalyzed the diffusion of metal music. We include time-varying controls for gross national income, political regime, global economic integration, and degree of metal penetration of countries sharing a land or maritime border with each country. We find that digital capacity is positively associated with heavy metal band foundings, but, net of all controls, the effect is much stronger for countries with no history of metal music prior to 1990. Hence, our results indicate that increasing global digital capacity may be a stronger catalyst for between-country than for within-country diffusion of cultural products. Downloaded from Sociological Perspectives 57 (1) processes of global diffusion that are not driven by large institutions or powerful actors, and less still about the diffusion of products that are primarily cultural and not economic in nature. We argue that these lacunae are both understandable and consequential for a full understanding of global diffusion processes. They are understandable in light of the paucity of data on the global spread of (typically hard to measure) cultural phenomena, particularly when large corporate actors advocating for their diffusion are few or relatively impotent. Yet, we argue, rigorous empirical investigations of the mechanisms by which culture spreads within and between countries promise to contribute to a thorough sociological understanding of globalization processes.Our goal in the present study is to provide one such investigation. The central theoretical concern of this article is not the why of decentralized cultural diffusion but the how. Our case is heavy metal music, which we argue below is an appealing example of the decentralized diffusion of culture. In the remainder of this introductory section, we justify the selection of our case. In subsequent sections, we develop a fourfold typology of diffusion systems by cross-classifying the actors sponsoring the diffusion of an innovation 1 with whether that innovation is meant primarily to provide economic returns or simply influence a political or cultural field. In addition to organizing the diverse diffusion literature, we believe this exercise brings into sharp relief the importance of a diffusion mechanism such as the Internet, at least for a relatively marginalized cultural product like metal music. Finally, we describe our data and methods, present findings, and discuss their implications for future research on decentralized cultural diffusion systems.