A study of the relative sensitivities of forced expiratory volume in one second (FEV1), maximal mid-expiratory flow (MMF), and closing volume (CV) in the detection of subjects with byssinosis was carried out in a North Carolina cotton mill. Altogether 35 workers participated in the study. Of these, nine showed a decline in FEV1 of 10% or more during the first work shift that followed the weekend break. Twelve subjects showed a decrease in MMF of 15% or more. In contrast only six workers exhibited a 10% increase in closing capacity, while ten showed a 10% increase in CV. Recent evidence of the magnitude of variability in closing volume manoeuvres suggests that our chosen level of change was too low, A 40% change in CV would have identified only five subjects. CV is a more complex manoeuvre for the subject being tested and for the technician to perform, is more time consuming, and is subject to greater variation. To have any advantage over spirometry, CV would have to be appreciably more sensitive. Our study suggests that it is not. However, the MMF may prove to be more sensitive than the FEV1 in the detection of byssinosis.