Time-weighted average (TWA) personal total and respirable dust exposures were determined gravimetrically for 48 subjects in 4 cabinet-making plants. TWA personal formaldehyde exposures also were obtained, with the use of 3M 3750 passive monitors. Selective area sampling for formaldehyde was undertaken using two methods. The results obtained with the passive monitors were compared to the standard chromotropic acid impinger method. Considerable variation was noted in the dust exposures. Cabinet-makers exposed to softwoods were found to have a mean exposure of approximately one half of the current applicable ACGIH TWA-TLV, while hard-wood exposure was twice the applicable TWA-TLV. The highest dust exposures were recorded for those workers sanding, the mean total dust being 2.91 mg/m3 (S.E. 0.70) and respirable dust 0.63 mg/m3 (S.E. 0.20). Sanding operations also were found to produce a higher proportion of respirable dust (22%) than other woodworking operations (6%-14%). Workers in assembly areas also were found to have higher dust exposures, likely reflecting the fact that conventional dust collection devices for stationary woodworking equipment are not appropriate for hand held tools and hand sanding. The importance of making respirable dust measurements is discussed. The poor correlation between paired total and respirable dust concentrations indicates that both measurements should be made. Some potential limitations to respirable wood dust sampling using 10 mm nylon cyclones are noted, however. Area dust concentrations were found to be significantly lower than personal exposures, emphasizing the importance of personal sampling data. Formaldehyde vapor exposures were very low, with a mean of 0.06 ppm (S.E. 0.01).