The core question of this thesis is why working time in South African gold mining has been so stable and addresses the significance of this fact. The working or labour time of miners and mineworkers is shown to have been remarkably stable for a century since 1911. By construing the length of the migrant labour contract as a measure of labour time, which systematically lengthens over the same period until it aligns with the annual rhythm of industrial working time, the evidence is provided for the argument that labour time constitutes the hitherto unrecognised foundation for the exploitation of mine labour in the South African gold mines. The phenomena - and importance for value-creation - of both relatively long, stable industrial working hours and the ever-longer migrant labour contracts over a century, are explained in terms of the value labour power creates in the mining labour process, as well as how the sheer expenditure of extended periods of labour time create the necessary skills mining requires. The fortunes of the platinum mining sector largely follow suit. Whereas the revisionist literature focused on the acquisition of a mine labour supply, this thesis argues that the retention of mine labour, by way of extending, intensifying and sustaining labour time in mining production, completes our understanding of its exploitation. It does so by employing a value-theoretic analysis which reveals the genesis of value creation in productive social class-based relationships. It shows how a series of qualitative, socially constructive effects, intra-working class occupational differentiation for example, emanate from the very expenditure of labour time underground when measured as a quantitative amount of labour time. It is argued that the substantive study of labour time has been surprisingly ignored in Marxist theory within which it plays a central role in the labour theory of value. A range of research methodologies have been employed to make this case. An ethnographic participant observation research method was aimed at articulating an agent-sensitive approach. The candidate lived in the hostel compounds and worked underground with mining personnel and has been both subject to the working time regimes on the mines as well as having actively participated, via various forms of research, in dealing with restructuring and changing working time schedules. The thesis goes on to show in close empirical detail, informed by actual experience and adopting a triangulated research methodology, how working time arrangements within which labour time is expended, remains immured in complexity. Why capital and labour, for instance, adopt competing stances regarding the restructuring of working time arrangements is explained. I conclude that workers’ production demands need to be taken seriously when working time is restructured in mining.