The purpose of these experiments was to measure the differences in total deposition in the nose arising from differences in the breathing patterns. Monodisperse droplets (from 1.7 to 10 mum) were deposited in an artificial nasal cavity (cast) using different human breathing patterns as well as under constant flow rate conditions. The human breathing patterns were recorded from a volunteer and then reproduced by a breathing simulator. The nasal cast was made from Magnetic Resonance images of the same volunteer, which were digitised and milled into consecutive plates to form a cavity. The results for small particles (1.7 and 3 mum) show much higher deposition at high constant flow rates than at fast human breathing. The difference becomes less pronounced with increasing particle size, but is still significant at high flow rates. This suggests that it might not be sufficient to take the average flow rate of the breathing pattern for comparison with constant flow. Therefore, the breathing patterns were partitioned into small segments and deposition was calculated for each segment. Adding deposition of each segment gives a theoretical predicted total deposition caused by the particular breathing pattern. However, the theoretical deposition is higher than the measured deposition at high flow rates, and again this is more pronounced for small particles. An explanation was given for this behaviour. (C) 2002 Elsevier Science Ltd. All rights reserved.