This thesis examines effects of companion animals on human cardiovascular reactivity. An examination of previous research investigating effects of companion animals on human cardiovascular reactivity suggested that previous mixed and mainly non-significant results in this area might be due to failures in methodology. This led to the development of recommendations for future studies. Three studies are presented which examined the effect of presence of an unfamiliar dog on participants' cardiovascular levels during a standardised reactivity study. The consistent finding from these studies was that the presence of an unfamiliar dog had no discernible effects on cardiovascular levels throughout the experiment (baseline and task levels combined) or on reactivity to stressors (difference between task and baseline levels). The fourth study investigated the effect of presence of the participants own pet on cardiovascular levels during a reactivity study. The study also included a condition of human companion presence. The results of the study indicate significant moderation of reactivity from the presence of both a pet dog and a human friend. The design of the study allows elimination of certain explanations such as differential vocal styles, distraction, threat of setting and perception of the experimenter. Whether social support is the mechanism which accounts for stress moderation in either companion condition is debatable. However in the case of pet dogs, it is argued that presence of ones pet during an everyday setting where one encounters stressful events would occur too infrequently to provide regular moderation of the stress response in the manner which has been proposed to lead to health benefits.