Histamine release by modified gelatin (Haemaccel) and dextran (Macrodex) has been demonstrated in volunteers by direct and indirect methods. In a pilot study of Haemaccel, histamine release was observed in six of seven volunteers. The highest plasma histamine concentration was 4.8 ng/ml, the lowest 1.7 ng/ml: two of the subjects showed slight allergic reactions. Using Haemaccel batch 2551, 10 out of 12 subjects reacted to the rapid infusion of Haemaccel with increased plasma histamine concentrations, whereas none reacted to Ringer's solution. None of the 10 subjects had an allergic reaction, but an increase in gastric secretion was observed in eight. Changes in the venous basophil granulocyte count were found in both those who reacted and those who did not react to Haemaccel. After the rapid infusion of dextran the highest plasma histamine concentration was 5.0 ng/ml, the lowest 1.3 ng/ml. The withdrawal of blood had no influence on plasma histamine concentration. The incidences of histamine release produced by Haemaccel varied with different batches. Thus, it seems unlikely that immunological mechanisms are principally responsible. Nine instances of allergic and anaphylactoid reactions to plasma substitutes have been reported, seven after Haemaccel infusion, and two after dextran administration. One of the patients who received dextran died. Histamine release was always associated with Haemaccel infusion and corresponded in extent to the clinical symptoms observed, but there was no significant histamine release associated with the reactions to dextran.