The phenomenon of serial killers is increasing world wide. During the past decade the most memorable cases of serial homicide in South Africa have been those of the "Norwood serial killer", the "Station Strangler", the "Cleveland serial killer", "The Gaunteng serial killer" and the "St Charles serial killer". The enigma surrounding serial killers is the so-called "lack of motive'' for the murders. No extrinsic motive such as robbery, financial gain, revenge or passion exists as there usually would be in the case of other murders. The available literature indicates that several theories have been proposed to explain the origin of serial homicide, but none can explain sufficiently to the author why one person with a certain type of background and exposed to a certain environment becomes a serial killer, whilst another, with a similar background and circumstances does not. The theories that are discussed are the socio-cultural factors, Ressler's motivational model, systemic factors, demonic possession, neurological factors, psychogenic factors and fantasy. In this thesis the author attempts to answer the question "What is the origin of serial homicide" by applying selected aspects of the psychoanalytic theories of Freud and Melanie Klein on two case studies of South African serial killers. The author formulates twenty nine statements of her own, based on the theories of Freud and Klein which she applies to the two case studies. The cases of the "Station Strangler" in Mitchell's Plain, Cape province, and the "St Charles serial killer" in Donnybrook, KwaZulu Natal are selected. Both of these serial killers are organized in their modus operandi, but the "Station Strangler" is a ego-dystonic killer and the "St Charles serial killer" a ego-syntonic serial killer. The method of research is a qualitative case study method. The author divides the selected aspects of Freud and Klein's theories into categories and analyses the two cases presented according to these categories. In the final chapter she draws maps to illustrate the psychoanalytical developmental paths of serial killers in general and for each of the two cases. Multiple data sources are used for example interviews with the serial killers and their families, police dockets, court proceedings and psychiatric records. Multiple data sources, explanation building and replication by way of multiple case studies are employed to ensure construct validity, internal validity and external validity respectively. Reliability is supported by entering the data in a South African Police Service data base. In conclusion the author compares the differences and similarities between the two case studies, and discusses the statements which are supported by this research. She also recommends that those statements which are not supported, be discussed in consecutive case studies. The statements are generalized to all serial killers. / Thesis (PhD)--University of Pretoria, 1996. / gm2013 / Psychology / unrestricted