The association between the prevalence of both byssinosis and chronic bronchitis and the level of airborne dust was examined in workers in preparing departments in flax mills in Northern Ireland. A weak association between the dust level and byssinosis was found but not between the dust level and chronic bronchitis. It is suggested that the disparity of these associations may have arisen because the diagnosis of byssinosis, as in most recent published studies, was based on relatively acute and reversible symptoms which are more likely to be closely related to the current dust levels than the slowly developing symptoms on which the diagnosis of chronic bronchitis was based. Furthermore an association between chronic bronchitis and the dust levels is likely to be obscured to some extent by the effects of other non-industrial respiratory irritants such as tobacco smoke. Some discussed. An association between the concentration of airborne dust in the working environment and the prevalence of respiratory symptoms in textile workers has been found by various authors (Roach and Schilling, I960; Mair, Smith, Wilson, and Lockhart, I960; Bouhuys, Van Duyn, and Van Lennep, I96I). Most writers have noted that it is workers in the early preparing textile processes who appear to be at the greatest risk of developing byssinosis, and it is these workers who are usually exposed to the highest levels of airborne dust. Within cotton carding rooms the proximity of workers to carding engines has been shown to be associated with the degree to which they suffer from ill health in general (Home Office, 1932) and byssinosis in particular (Schilling, Hughes, Dingwall-Fordyce, and Gilson, I955). Roach and Schilling (I960) attempted to derive a quantitative relationship between the prevalence of byssinosis and the level of airborne dust in five selected Lancashire cotton mills. They correlated the prevalence of byssinosis in 309 workers in I0 groups defined by workroom , or type of room, with the mean level of airborne dust in these rooms at the times of the study. Batawi, Schilling, Vali6, and