The aim of this dissertation is to investigate the legacy of industrial spaces, the effect of this legacy on the surroundings, and how these spaces then become disconnected and isolated after industrial activity is decommissioned. The research forms part of an NRF research scheme that specifically focuses on building the resilience of cities through innovation in the planning, design and construction of the built environment. The hypothesis on which the dissertation is based states that a process of reintegration of a decommissioned industrial site with the immediate surroundings would enable such a site to become a positive space of transition, and would allow for the reconciliation of society and the ecology that was exploited by the industry. It sees the decommissioning of industrial infrastructure not as a loss or abandonment of obsolete capital, but as the release of energy and potential that can be positively reconstructed. The mechanistic and reductionist world-view that contributed to an unhealthy relationship between people and their ecological surroundings is theoretically explored through the hybridization theories proposed by Bruno Latour (Latour 1993), and the regenerative methodologies put forth by members of Regenesis (Mang, Reed 2012a). The potential of obsolete industrial infrastructure to provide powerful leverage points for changing paradigms from mechanistic to ecological is discussed in the light of its history of developing from craft to large-scale production. Craft becomes an important mechanism for the integration of people with the value and purpose of their work, and also of natural materials and the cultural objects they become. The theories stated above are architecturally applied to an industrial site in Eersterust, Pretoria, which is on the verge of being decommissioned. The site is approached as a constantly evolving and living entity. It is investigated in terms of its patterns and cycles, and these are illustrated as a narrative of all the forces that have impacted on it over millions of years. The narrative provides clues as to possible programmes and site lifecycles, and enables those phenomena that will nurture the biophysical evolution of the site to be given form. The concept of potential sets arises from this investigation, and informs an architecture that aligns itself with both the ecological and cultural forces on site, and represents the hybridization of the two. Potential sets distinguish patterns of ecological, social and industrial phenomena that occur on site over different time frames. These patterns aid the understanding of the ecological purpose of the site and the alignment of the built intervention with this purpose. A building is imagined that will create solutions for public, industrial and ecological spaces, with different levels of engagement between the three. The concept of a solvent enforces the notion of hybridity and allows for new relationships between the public, industrial processes and natural cycles to develop. / Dissertation MArch(Prof)--University of Pretoria, 2014 / Architecture / MArch(Prof) / Unrestricted