1. A method for the detection of gastrointestinal blood loss based upon the selective measurement of faecal porphyrins was tested in two studies in healthy volunteers. 2. In the first study subjects (n = 6) received intragastric autologous blood (saline, 2 and 6 ml as a single dose) resulting in a dose dependent increase in faecal porphyrins. 3. In a subsequent placebo controlled cross over study in 12 subjects acetylsalicylic acid (ASA), nabumetone (a new NSAID) or placebo were administered for 5 days with a washout period of 9 days. They were no dietary restrictions. 4. All faeces were collected during the treatment period and both the full faecal homogenate and a random faecal sample were analyzed for deutero- and pemptoporphyrin content by h.p.l.c. Additionally a benzidine reaction was performed. 5. There was a highly significant correlation (r = 0.95) between the values obtained from random samples and the full homogenate. ASA increased the faecal porphyrin excretion (P less than 0.001) compared with placebo in contrast to nabumetone. Complaints of dyspepsia were most common after ASA. 6. Measurement of faecal porphyrins is useful for monitoring NSAID induced upper gastrointestinal blood loss and lacks some of the practical constraints of other methods.