This paper argues that Alexander Gordon (1752-1799) had a series of empirical arguments that lead him to the conclusion that puerperal fever is an infectious disease. He thus provides a case study of the shift during the 18th century to an empirically based approach to medicine.

Gordon was the physician for the Aberdeen Public Dispensary and treated many of the women who became ill during the epidemic of childbed fever that extended from 1789-1792. This paper first examines the geographic distribution of cases of childbed fever by plotting each case on the 1789 Milne map of Aberdeen. The geographic distribution led Gordon to reject the miasma theory and to seek an alternative causal explanation. A second important observation was when the illness arose. He observed that the onset of the fever never preceded and always followed childbirth and thus focused upon this time as the causal interval. A third and fourth line of argumentation emerges from the detailed records that he kept as the physician for the Public Dispensary. As a result of these records, he observed a pattern linking specific midwives and physicians with who become ill, and he concluded that it was a disease carried by midwives and physicians from woman to woman. Finally, he observed a correlation between the number of cases of puerperal fever and erysipelas (i.e., a recognized infectious fever) and noted that a surgeon can acquire an inflammation and fever as the result of a scratch during a dissection of a patient who died of puerperal fever. As a result of this empirical line of inquiry, Gordon concluded that childbed fever is an infectious disease carried by midwives and physicians.