John Haldane

The Relation of the Action of Carbonic Oxide to Oxygen Tension

THE following investigation took its origin from an enquiry in which I am at present engaged into the nature and action of the suffocative or poisonous gases met with in the air of coal-mines. Among these gases is carbonic oxide. So far as my experience goes, however, carbonic oxide never occurs alone as an impurity, but always in connection with an excess of diluent gases, chiefly nitrogen; so that when carbonic oxide is present in the air of a coal-mine, there is also a more or less considerable reduction in the percentage of oxygen. It is known, however, firstly, that part at least of the action of carbonic oxide is due to its property of entering into combination with the haemoglobin of the blood corpuscles, and so putting them out of action as oxygen carriers; and, secondly, that the proportion of carbonic-oxidehaemoglobin formed in blood brought into contact with a gas-mixture containing carbonic oxide and oxygen, depends not merely on the tension of carbonic oxide in the mixture, but also on the tension of oxygen. The larger the proportion of oxygen in the gas mixture the smaller will be the amount of carbonic-oxide-bkemoglobin formed in the presence of a given proportion of carbonic oxide. Hence it might be expected that in air containing a diminished proportioni of oxygen, carbonic oxide will be more poisonous; and that in air containing an increased proportion the poisonous action will be less. A very simple experiment served to show that this is the case.

The Oxygen Tension of Arterial Blood

THE chief interest of the present investigation centres round the question whether diffuision alone explains the transference of oxygen fromn the air of the pulmonary alveoli into the blood, or whether other important factors are concerned in this process. If the oxygen tension of the blood leavina the alveolar capillaries is always, or even occasionally, higher than that of the alveolar air, diffusion alone evidently cannot explain the transference. The analogy presented by other organs of absorption or excretion would perhaps lead us to believe that absorption of oxygen by the lungs is probably not due to any simple physical process such as diffusion. On the other hand it does not seem at all clear that any physiological advantage would accrue fronm an active absorption of oxygen by the lungs. The red corpuscles of blood, saturated with air of alveolar oxygen tension (about 140/ of an atmosphere in man), would contain very nearly as much oxygen as the haemoglobin is capable of absorbingo, so that any further increase in the oxygen tension of arterial blood would imply only a very trifling increase in the quantity of oxygen taken up by the blood. If, therefore, the exposure of the blood to the alveolar air is sufficiently prolonged under all ordinary circumstances for saturation to occur by diffusion alone, no appreciable advantage would apparently result from an active absorption of oxygen. On this point however we have no experimental data, so that there is thus not much

The colorimetric determination of haemoglobin.

COLORIMETRIC DETERMINATION OF H.AEMOGLOBIN. BY JOHN HALDANE, M.D., F.R.S. (One Figure in Text.) [From the Physiological Laboratory, Oxford.] THE colorimetric method of determining the relative percentages of haemoglobin in different samples of blood has hitherto suffered from the disadvantage that there has been no satisfactory means of preparing a permanent and at the same titne definite standard of colour for making the comparisons. Some time ago Hoppe-Seyler, introduced the plan of using as a standard a solution of pure oxy-hemoglobin ofknown strength. He also drew attention to the fact that although such a solution soon decomposes when left exposed to the air, yet strong solutions of either reduced haemoglobin or CO-ha3moglobin, when kept sealed up, may be preserved indefinitely without showing signs of spectroscopic change, and may be opened and diluted when required for use with a colorimeter. In a later paper2 he described a special colorimeter, for which a dilute solution of CO-haemoglobin is made as required from a stronger standard solution which has been kept sealed up. Dilute solutionis of CO-haemoglobin are stated to be unstable. Onaccount of the difficulty in preparing perfectly pure solutions of haemoglobin Hoppe-Seylers method is not verv satisfactory, besides being too complicated for ordinary work. Solutions artificially coloured of the same tint as dilute oxy-hsemoglobin of a certain strength have been adopted in some heemoglobinometers, the simplest and most convenient of which is the instrument of Gowers3, where a mixture of picric acid and carnilinate of ammonia is employed to give the standard tint. In the instruments of v. Fleischl and Oliver coloured glass of variable depth of tint is substituted for the coloured liqiuid. There are two serious defects connected with the use of artificially coloured solutions and tinted glass. The first is the difficulty in standardising a given coloured solution or piece of coloured glass to correspond with a definite strength of haemoglobin solution. The second is that although with a certain strength of solution or thickness

A Contribution to the Chemistry of Hæmoglobin and its Immediate Derivatives

THE investigations described in the present paper arose from the observation that when ferricyanide of potassium is added to a not too dilute solution of blood in water bubbles of gas are evolved at the same time as niethaemoglobin is formed. On examination the gas proved to be nearly pure oxygen. This fact was exceedingly surprising, since Hiifner and Kulz1 have shown that although methaemoglobin yields no oxygen to a vacuum, it gives up to nitric oxide (a reducing substance) exactly the same amount of oxygen as oxyha3moglobin does, and must therefore contain just the same amount of easily available oxygen. Reduced blood was found to yield no gas. Blood saturated with carbonic oxide yielded just as much gas as blood saturated with air, and this gas was found to be inflammable. The quantities of gas obtained evidently corresponded more or less closely to the quantities present in loose combination in the hb3moglobin; but in order to obtain more definite information as to the nature and exact volume of the gas liberated I made the following experiment with the blood-pump. Fresh ox-blood was saturated with air and diluted with an equal volume of water to dissolve the corpuscles, since ferricyanide does not act on undissolved corpuscles. A receiver of known capacity (41 c.c.), with two air-tight taps, and made to fit into the froth-chamber of the blood-pump, as described by Bohr2, was then filled with the diluted blood, and fixed in position. After the frothchamber had been evacuated two or three c.c. of the blood were allowed to pass inwards, and a corresponding quantity of saturated solution of ferricyanide was allowed to flow into the receiver through the otber

A Method of Detecting and Estimating Carbonic Oxide in Air

IN view of the very poisonous nature of carbonic oxide, and the comparative frequency with which cases of poisoning by this gas occur, much attention has been given to its detection and estimation. No satisfactory method has hitherto been described, however, for determining very small, but still more or less poisonous, percentages of carbonic oxide in air. There is special difficulty when, as almost always happens, marsh-gas or other hydrocarbons are present along with the carbonic oxide. The method now to be described is very simple, and depends on the fact that when a haemoglobin solution is well shaken with air containing carbonic oxide the proportion of the hbemoglobin, which finally combines with the carbonic oxide, varies with the percentage of carbonic oxide present in the air. By determining colorimetrically the proportion of the haemoglobin which has combined with the carbonic oxide it is thus possible to infer the percentage of carbonic oxide present in the air. Hsemoglobin solutions have been used for long as a qualitative test for carbonic oxide in air. Vogel1, who introduced this test, directs that a small quantity of very dilute blood be shaken in a small bottle filled with the air, and that the blood be then treated with ammonium sulphide to reduce the oxyhaemoglobin present, and examined with a spectroscope. If the two well-known absorption bands are still visible carbonic oxide is present in the air. Vogel found that by this method as little as *25 0/0 of carbonic oxide could be detected. He states that