This thesis is the work of the author and is concerned with the development of a neo-Wigmorean approach to evidential reasoning in police investigation. The thesis evolved out of dissatisfaction with cardinal aspects of traditional approaches to police investigation, practice and training. Five main weaknesses were identified: Firstly, a lack of a theoretical foundation for police training and practice in the investigation of crime and evidence management; secondly, evidence was treated on the basis of its source rather than it's inherent capacity for generating questions; thirdly, the role of inductive elimination was underused and misunderstood; fourthly, concentration on single, isolated cases rather than on the investigation of multiple cases and, fifthly, the credentials of evidence were often assumed rather than considered, assessed and reasoned within the context of argumentation. Inspiration from three sources were used to develop the work: Firstly, John Henry Wigmore provided new insights into the nature of evidential reasoning and formal methods for the construction of arguments; secondly, developments in biochemistry provided new insights into natural methods of storing and using information; thirdly, the science of complexity provided new insights into the complex nature of collections of data that could be developed into complex systems of information and evidence. This thesis is an application of a general methodology supported by new diagnostic and analytical techniques. The methodology was embodied in a software system called Forensic Led Intelligence System: FLINTS. My standpoint is that of a forensic investigator with an interest in how evidential reasoning can improve the operation we call investigation. New areas of evidential reasoning are in progress and these are discussed including a new application in software designed by the author: MAVERICK. There are three main themes; Firstly, how a broadened conception of evidential reasoning supported by new diagnostic and analytical techniques can improve the investigation and discovery process. Secondly, an explanation of how a greater understanding of the roles and effects of different styles of reasoning can assist the user; and thirdly; a range of concepts and tools are presented for the combination, comparison, construction and presentation of evidence in imaginative ways. Taken together these are intended to provide examples of a new approach to the science of evidential reasoning. Originality will be in four key areas; 1. Extending and developing Wigmorean techniques to police investigation and evidence management. 2. Developing existing approaches in single case analysis and introducing an intellectual model for multi case analysis. 3. Introducing a new model for police training in investigative evidential reasoning. 4. Introducing a new software system to manage evidence in multi case approaches using forensic scientific evidence. FLINTS.