We review the aetiology of scleroderma from an epidemiological perspective examining genetic, environmental and stochastic risk factors. The presence of familial clustering (but with low twin concordance) suggests a genetic contribution, and this has been confirmed with recent candidate gene and genome-wide association screening demonstrating both major histocompatibility complex and non-major histocompatibility complex genetic linkage. In contrast, environmental associations are weak or inconsistent. An examination of the age-adjusted incidence curve of scleroderma is consistent with a stochastic process involving five to eight random events. In pathogenesis, scleroderma is best considered as an autoimmune disorder where genetic and environmental factors are both important variables, but random events are also likely to play a pivotal role. We suggest that these random events might result in acquired somatic mutations or epigenetic alterations involving genes coding for immune receptors, tolerogenic gates or proteins involved in immune regulation, inflammation and/or repair that, over time, might summate to form a requisite cassette (of genetic changes), which allows the initiation and progression of the autoimmune process.