The definition of the vegetative state and the description of its features have evolved over the years as individuals have reported surveys of patients and as various medical organizations have produced consensus statements. Criteria used for two widely quoted Japanese epidemiological surveys in the late 1970s now seem very imprecise (Tables 2.1 and 2.2), but they may have lost something in translation. However, it is clear that the definition of Sato et al. (2) allowed inclusion of patients who could obey some commands and who would therefore have been excluded by later definitions. The responses of 250 child neurologists (3) who were asked in 1991 to comment on the relative importance of ten features that had been proposed as an operational definition of vegetative state by Nelson and Bernat (4) showed a marked lack of consensus (Table 2.3). An estimate of the prevalence of the vegetative state in children in California in 1991 was based on a survey of State residents registered as developmentally disabled (5). To identify residents who might be considered to be in a vegetative state 15 items were selected from the adaptive behavioural section of the Client Development Evaluation Report form (Table 2.4). However, this form had been devised for other purposes and there must be some doubt as to how accurately this group of items corresponds with more formal definitions of the vegetative state that have emerged since then.