A development over the past 70 to 80 years within scientific research has been the need for very large pieces of apparatus to enable the exploration of new scientific topics, particularly within particle physics and space science. These ‘megascience projects' are generally undertaken as cooperative ventures by countries seeking to pursue scientific experimental opportunities in these fields. Such projects, a subcategory of large/megaprojects that have a minimum budget of one billion US dollars, are characterised by high levels of technological uncertainty, given that their success depends on the development of new, highly-advanced technologies . However, there is a notable lack of research into the leadership of megascience projects - an important consideration when embarking on a substantial project. The leadership literature traditionally categorises leaders into five discrete leadership styles, but there is a gap when it comes to understanding the characteristics and development of leaders of megascience projects. In this thesis, I address this gap in knowledge, focusing on three research questions: (1) What are the characteristics of those who lead megascience projects? (2) Where were their leadership skills developed? (3) And how were their leadership skills developed? A useful concept during the intellectual journey to answer these questions was ‘the heterogeneous engineer', which provided the original conceptual framework for this thesis. I use a combination of archival and interview-based research to answer these research questions in the cases of the Tevatron at Fermilab in the United States, and the Large Hadron Collider (LHC) at CERN on the Franco-Swiss border near Geneva. This archival research notably required access to normally restricted sections of the CERN archives related to the LHC. The thematic analysis conducted for this research yielded various findings that include the primacy of technical competence as a foundation for respect, along with strong management ability, the importance of trustworthiness, and team empowerment. Furthermore, I found that leadership training within megascience projects is experiential in nature, with formal leadership training programmes acting at most in a support role. During the analysis of my data, I concluded that the heterogeneous engineer concept was based on a relative anomaly, making it difficult to use this concept as the foundation for a more generalised leadership theory. One unexpected finding, which represents a relatively original contribution of this thesis, is the tailoring of senior leadership selection to suit a specific project phase, something which appears to partially contradict the current literature. I identify four phases, the characteristics of leaders best equipped for each phase, and the implications for other large projects.