In a previous paper (Krogh and Ussing, 1936) a preliminary study was made of the exchange of hydrogen-atoms between the free water and the organic substances in living organisms by using heavy hydrogen (deuterium) as an indicator and comparing the deuterium percentage in the water formed by combustion of dry tissues with the deuterium in the free water of the body.As no satisfactory agreement was obtained between different determinations we thought that some systematic error was responsible. The technique of drying and combustion of the organic material was therefore tested again. It turned out that the errors were due mainly to incomplete drying of the tissues, the last traces of water being held with great force. 'When the lipoids are not removed by extraction a complete dehydration of the tissues seems to be unattainable and this proved to be the main reason why our first experiments with frogs and rats gave too high results compared with the later experiments on mice, in which lipoid extraction was used.But in some cases the drying was still insufficient especially as regards blood serum and similar protein-solutions. When the water is distilled off the protein is as a rule left as horny scales which will not give off the last water within a reasonable time in a disiccator at 760 mg Hg pressure. If the last traces of water were ordinary water they would have no peroeptible influence upon the result. But in most cases they contain heavy water at a concentration several times higher than the water which is formed by the combustion, and then the influence even of small amounts of free water may be disastrous. An example will show this.