Before making any definite experiments, as complex as those about to follow, each apparatus should be separately examined and its action corrected ; thus we pass from the simple to the intricate. These preliminaries belong entirely to the domain of practice; equations, theories, have no bearing on them.
It is necessary to have the joints absolutely tight; the valves to close hermetically ; as little friction as possible ; good packing ; very exact, corrected manometers; accurate thermometers; absolutely pure chemicals. There are a thousand details, too many to enumerate, each of which, however, is so important that the least oversight or inadvertency will entirely compromise final success. The 'slightest admission of air, so easily occurring, with four ordinary pumps working at a barometric vacuum, would be an inevitable cause of stoppage, and must be prevented at all hazards.
In generating the carbonic acid by the action of chlorhydric acid on marble, the gas must be dried, the primary substances examined, and the gasometer completely freed of air. For the oxygen, the chlorate of potassa must be absolutely pure and dry, and must be introduced, so that no foreign body can get mixed with it.
Finally, a thousand precautions must be taken, and it may be affirmed that none too many can be adopted. In this respect we owe a gratefully acknowledgment to M. le Dr. Monnier, Professor of Biological Chemistry at the University of Geneva, who has lent us his aid with great kindness, as well as to M. Walter, his assistant, and to M. Burgin, engineer, a member of our society.