(From the Physiological Laboratory, Oxford.) IN a succeeding paper the results will be given of a number of experiments by the carbon monoxide method of determining the oxygen pressure of the arterial blood. As the method is based on a quantitative interpretation of the law of combination of carbon monoxide with hemoglobin in presence of different partial pressures of oxygen, we found it, necessary to re-investigate this law with a view to its more complete elucidation. The results are contained in the present paper, and throw considerable new light on the mode of combination of heemoglobin with oxygen, as well as with carbon monoxide.From a number of experiments Haldane and Lorrain Smith2 concluded that when a solution containing hemoglobin is saturated with a gas mixture containing oxygen and CO the relative proportions of the haemoglobin wbich enter into combination with the two gases are proportional to the relative partial pressures of the two gases, allowing for the fact that the affinity of CO for haemoglobin is about 300 times greater than that of 2. The dissociation curve of HbCO in presence of pure air mixed with varying percentages of 0o, or of a gas mixture containing a constant percentage of CO and a varying percentage of oxygen, is thus, according to their conclusions, a rectangular hyperbola. Their experiments were nearly all made at about 150 or 370 with dilute solutions of ox-blood, and all were in the absence of CO,. Only three rather rough experiments were made with undiluted blood.At that time it was still generally believed that the dissociation curve of the oxyhtemoglobin in blood is a rectangular hyperbola, and1 The former two authors are responsible for the experimental results; the latter for the mathematical analysis.

No abstract

IT has been known for long that under certain conditions the breathing may becomne periodic. This phenomenon was first observed in disease by Drs Cheyne and Stokes, after whom it was named. It has also been observed in healthy men at high altitudes by Mosso and others: in hibernating animals: in infants: and as a consequence of the action of various toxic substances. No satisfactory account of its mode of causation has hitherto been given, although the recent discovery of Pembrey and Allen' that pathological Cheyne-Stokes breathing is abolished by the administration of either oxygen or air containing excess of CO clearly points in the direction from which an explanation may be expected.In a recent paper by Haldane and Poulton' the fact was recorded that in more than one experiment where the subject had been made hyperpnceic by want of oxygen, apncea followed after a few breaths of normal air, this apncea being succeeded by deep breathing and then by a secondary apnoea. On further examination we found that well-marked Cheyne-Stokes breathing could easily be produced experimentally, and we have taken advantage of this fact in order to investigate the causes of periodic breathing.The first method which we used for producing Cheyne-Stokes breathing was as follows. The subject of the experiment reclined in a perfectly easy position in an arm-chair: he then breathed deeply and frequently for about two minutes, so as to produce a prolonged apncea of about two minutes' duration, as described by Haldane and Poulton.

IT has recently been shown, that under normal conditions, which may vary very widely, the breathing is so regulated by the respiratory centre as to maintain a constant, or nearly constant, level in the partial pressure of CO2 in the alveolar air, and therefore also in the arterial blood. The centre is extremely sensitive to the slightest increase or diminution in CO2-pressure. From experiments in which there was an added percentage of CO in the inspired air Haldane and Priestley found that a rise of as little as 020/, of CO2 in the alveolar air, corresponding to an increase of 1-4 mm. in the CO2-pressure, was sufficient to increase the alveolar ventilation by 1000/o.The respiratory centre must evidently be regarded as a very sensitive governor of the C02-pressure in its own substance, and indirectly in the arterial blood and alveolar air. We may compare its action to that of the governor of an engine driven by steam or water, if we bear in mind that the respiratory centre governs C02-pressure, whereas the governor of the engine controls its rate of revolution.It is well known to engineers that something more than a sensitive governor is needed in order to produce practical constancy in the rate of revolution of an engine-for instance a steam-engine or water-turbine driving a dynamo at a constant rate, so as to obtain a constant voltage in spite of very variable amounts of current being produced by the dynamo. The governor must not merely be sensitive, but it inust not cause or permit temporary irregularities, nor must it "hunt," i.e. cause periodic variations in speed owing to its alternate excessive and 1 Hald ane and P riestley. This Journal, xxxii. p. 225. 1905.

There has been much speculation for a long time as to the causes of the increased breathing during exercise. In 1905, Haldane & Priestley put forward the theory that it was due to a slight increase in the alveolar CO2 tension (pCO2) and therefore of the arterial pCO2, just as similar increases in alveolar PCO2 induced by CO2 inhalation at rest produced hyperpnoea. Further work showed that whatever might be true for mild exercise, as the work became harder the increase in pCO2 was certainly insufficient to account for the hyperpnoea, and it was clear that some additional factors must come into play. The theory that hydrogen-ion concentration rather than CO2 as such was the stimulus to respiration was more satisfactory since it was known that severe exercise was associated with the accumulation of an excess of lactic acid in the body. Since then, this view has been challenged by many physiologists, and at present there is no agreement as to the main causes of the hyperpnoea. In view of this uncertainty we decided to re-investigate the changes in alveolar C02 during heavy exercise, together with some factors which might modify the response of the respiratory centre to CO2. The current views are partly based on the contention of Krogh & Lindhard (1913c, 1917) that the directly measured alveolar gas tensions do not reflect the gas tensions in the arterial blood during exercise. It is therefore important to the argument presented in this paper that the old controversy between the Danish and Oxford schools of physiology should be re-examined in the light of recent evidence on the alveolar air-arterial blood gas relationships. The arguments used to justify the use of the directly measured alveolar pCO2 as a guide to the arterial pCO2 in exercise are long and intricate and are therefore presented in an appendix to this paper. The conclusions reached are summarized in the main text. METHODSAlveolar air. At rest, alveolar air samples were taken by the Haldane-Priestley method. The values presented in this paper were the means of end-inspiratory and end-expiratory samples.Automatic alveolar air sample8. In exercise, a modification of the method described by Lindhard (1911), and modified by Rahn & Otis (1949), was used (Fig. 1). In this method the last few c.c. of