Severe acute respiratory syndrome (SARS) coronavirus (CoV) infection, in addition to infections originating in the laboratory, was detected in four patients with no possible laboratory exposure during December 2003 and January 2004. These four epidemiologically unrelated patients likely acquired the infection through exposure to infected animals in live-game-animal markets in Guangdong. A number of possible agents were detected in the course of investigating patients with suspected SARS, including human metapneumovirus and chlamydia. In the absence of an epidemiological history of exposure, clinical findings are not pathognomic of SARS. Therefore, a positive virological laboratory finding of SARS CoV infection is required for confirmation of the diagnosis. Evidence of SARS CoV replication in the lungs and intestine is provided by detection of virus particles through electron microscopy, by virus isolation, and by the detection of viral antigens and nucleic acid through immunohistology and in situ hybridization. Experimental infections in relevant animal models are important for understanding pathogenesis and evaluating therapeutic and vaccine strategies. The clinical management of patients with SARS includes respiratory support with intensive care when appropriate, prevention of nosocomial transmission, and the possible use of antivirals and immunomodulators. Aspects of its pathogenesis and transmission allowed the human disease outbreak to be quickly interrupted. The global response to SARS demonstrated that a rapid mobilization and coordination of relevant expertise is possible, when faced with a global emerging disease threat. It also highlighted the need for improved international regulations governing the reporting of and response to unusual infectious-disease syndromes.