this spectrum is sometimes spoken of as that of alkaline methæmo globin.

1

3. Carbonic oxide hæmoglobin.-This may be readily prepared by passing a stream of carbonic oxide gas through blood, or through a solution of oxyhæmoglobin. Its colour is a peculiar cherry-red. Its absorption spectrum (see fig. 59, 4) is very much like that of oxyhæmoglobin, but the two bands are slightly nearer the violet end of the spectrum; the centre of the a band being X 572, of the 3 band X 534 to 538 according to concentration (Gamgee).

CO-hæmoglobin forms crystals like those of oxyhæmoglobin; it is remarkable for its stability; it is not affected by reducing agents like ammonium sulphide, and the carbonic oxide gas can only be driven off by passing through it for a long time a stream of air or of a neutral gas, or by a stream of nitric oxide gas which replaces the carbonic oxide and forms nitric oxide hæmoglobin. CO-hæmoglobin also resists putrefaction for a long time (Hoppe-Seyler 2).

Carbonic oxide is given off during the imperfect combustion of carbon, such as occurs in charcoal stoves; this acts as a powerful poison by combining with the hæmoglobin of the blood, and thus interfering with normal respiratory processes. The colour of the blood and its resistance to reducing agents in such cases are characteristic. Hoppe-Seyler has, however, introduced another test which has been modified by Salkowski 3 as follows: The blood in question is diluted twenty times, and to some of this in a test-tube an equal volume of aqueous soda of specific gravity 1:34 is added. In a few seconds CO blood becomes whitish, then red; on standing red flocculi separate and finally rise to the surface of a faintly rose-coloured liquid. In normal blood all that is produced by the addition of the alkali is a dirty-brown colouration (hæmatin). Working under Salkowski's direction Katayama has discovered a new test which may be briefly stated as follows : The addition of acetic acid and ammonium sulphide (with sulphur in solution) to normal blood produces a greenish-grey or reddish greengrey colour; to CO blood, a beautiful clear rose-red is produced. This solution shows a spectrum which is a double spectrum, indicating that there is in solution CO-hæmoglobin and sulphur-methæmoglobin (HoppeSeyler 5), viz. one band between C and D, and two others between D

1 The gas may be generated by adding sulphuric acid to oxalic acid or formic acid in a retort, and then applying heat.

2 Hoppe-Seyler, Zeit. physiol. Chem. ii. 181.

3 E. Salkowski, Ibid. xii. 227.

4 Katayama, Virchow's Archiv, 1888, vol. exiv. p. 53. References will be found in this paper to other tests which have been proposed for CO-hæmoglobin.

5 Physiol. Chem. p. 388. The composition of sulphur methæmoglobin is not known.

and E. The spectrum shown by normal blood after the addition of these reagents is also a double spectrum, viz. of sulphur methæmoglobin and reduced hæmoglobin. In other words, in the case of CO-hæmoglobin the colour of that compound completely masks the olive-green tint of sulphur-methæmoglobin, which spectroscopic observation shows to be

present.

4. Nitric oxide hæmoglobin.-When ammonia is added to blood, and then a stream of nitric oxide passed through it, this compound is formed (Hermann 1); it may be obtained in a crystalline form isomorphous with oxy- and CO-hæmoglobin; it also has a similar spectrum. It is even more stable than CO-hæmoglobin.

Other compounds of hæmoglobin have been described: one with acetylene (C2H2), another with hydrocyanic acid (Hoppe-Seyler). Dr. Gamgee3 has, however, pointed out the unsatisfactory nature of the evidence upon which the existence of such compounds rests.

Recently C. Bohr has advanced the theory that hæmoglobin forms a compound with carbonic acid. The importance of this discovery, if confirmed, is very great, and the question will be discussed in the chapter on Respiration (Chapter XIX).

Estimation of Hæmoglobin

The following methods may be adopted for the quantitative estimation of hæmoglobin :

1. By the amount of iron in the ash.

2. By colorimetric methods.

3. By spectrophotometric methods.

1. A weighed quantity of blood or substance containing hæmoglobin is evaporated to dryness, the residue is carefully incinerated at a dull red heat, the ash exhausted with hydrochloric acid to obtain ferric chloride. This is reduced by the action of metallic zinc to ferrous chloride, and the amount of iron in this determined volumetrically with a standard solution of potassium permanganate (see also p. 25). Dry hæmoglobin contains 0.42 per cent. of iron. If m=percentage amount of iron in the specimen under examination, the percentage of hæmo100 m 0.42

globin in that specimen = =

2. Standard solutions of known strength are prepared from crystals

It is formed by passing a stream of sulphuretted hydrogen through a solution of oxyhæmoglobin. The greenish tint which appears on the surface of corpses a few days after death is due to the development of sulphuretted hydrogen, and the consequent formation of sulphur-methæmoglobin. See also Araki (Zeit. physiol. Chem. xiv. 412).

1 Hermann, Reichert und Du Bois Reymond's Archiv, 1865, p. 469.

2 Hoppe-Seyler, Med. Chem. Untersuch. Heft ii. p. 207.

3 Gamgee, Physiol. Chemistry, pp. 107-8.

of oxyhemoglobin. The blood to be investigated is diluted with water until the colour of the standard solution is reached. Knowing the amount of blood and the amount of dilution, the percentage of hæmoglobin is easily calculated (Hoppe-Seyler).

The tint of the solutions must be ascertained by examining them in vessels with parallel sides and of the same width. A hæmatinometer, as such a vessel is called, is usually constructed so that the sides are 1 centimetre apart. Rajewsky' and Malassez recommend the standard solution to be made up of picrocarminate of ammonia, the tint of which corresponds to that of an oxyhemoglobin solution of known strength.

In Fleischl's hæmometer a wedge of red-tinted glass forms the standard of comparison; the wedge is arranged to slide under a hole in a brass plate, the thickness of the glass under observation can thus be varied and adjusted so as to give a red tint equal to that of the blood under examination, which is always diluted to a certain fixed extent.

Gowers' hæmoglobinometer, like Fleischl's instrument, is designed for clinical use. The apparatus consists of two glass tubes, c and D, of

11 BA

201 CH

B

FIG. 60.-Hæmoglobinometer of Dr. Gowers. (Hawksley.)

the same size. D contains glycerine jelly tinted with carmine to a standard colour, viz. that of normal blood diluted 100 times with distilled water. The finger is pricked and 20 cubic millimetres of blood are measured out by the capillary pipette, B. This is blown out into the tube c, and diluted with distilled water, added drop by drop from the pipette stopper of the bottle, A, until the tint of the diluted blood reaches the standard colour. The tube, c, is graduated into 1 Rajewsky, Pflüger's Archiv, xii. 70. 2 Malassez, Arch. de Physiol. 1877, p. 1.

100 parts. If the tint of the diluted blood is the same as the standard when the tube is filled up to the graduation 100, the quantity of oxyhæmoglobin in the blood is normal. If it has to be diluted more largely, the oxyhæmoglobin is in excess; if to a smaller extent, it is less than normal. If the blood has, for instance, to be diluted up to the graduation 50, the amount of hæmoglobin is only half what it ought to be 50 per cent. of the normal, and so for other percentages.

The instrument only yields approximate results, but is extremely useful in clinical observations.1

3. The spectrophotometric method for the estimation of coloured solutions has been already described (see p. 50).

In connection with the estimation of oxyhemoglobin it may be added that the region of the spectrum selected for photometric measurements is that of the 3 band of absorption. This part of the spectrum in the case of oxyhæmoglobin has been found to be that most easily affected by changes in the concentration of the solution through which the light passes.

Glazebrook's spectrophotometer is, in principle, the same as Hüfner's. Light from each of two sources passes first through a Nicol's prism by which it is polarised, then through a direct vision prism; thus, two adjacent superposed spectra are obtained, and these are observed by an eyepiece in which is an analysing Nicol's prism. This eyepiece can be rotated, and the amount of rotation measured by a pointer attached to the eyepiece moving over a graduated circle. The nicols are then adjusted in the way already described, so that the two spectra appear of equal brightness. One may then proceed to interpose the coloured solution and measure the angles through which it is necessary to rotate the nicols, or, more simply, in the following way, as suggested by Dr. Sheridan Lea." On the path of one beam of light is placed a solution of known concentration, and on the path of the other, one of unknown concentration. If the latter is of greater concentration than the first, it may be diluted down till the effect upon the two spectra is the same; from the amount of dilution necessary to produce this effect its concentration can be calculated. Or the same effect can be produced by varying the thickness of the layer of the fluid under observation; this latter plan is found to be perfectly feasible, and is applicable also when the concentration of the unknown solution is less than that of the known; and Dr. Lea has invented an instrument (absorptiometer) with parallel sides, one of which is movable, and so the thickness of the layer of fluid in it can be varied to known extents. Take as an llustration the following example, which reduces to its simplest elements 1 Gowers, Lancet, vol. ii. 1878, p. 822. Sheridan Lea, Journ. of Physiol. v. 239.

25

the method of spectrophotometry. The spectra were first equalised, and a layer of a standard solution of known concentration, C1, was examined in a layer inch thick. It was found necessary to interpose on the path of the other beam of light a layer 3 inch thick of a solution of unknown concentration, C, in order to make the spectra C 13 once more equal. = from which equation, C the unknown quanC1 6'

tity is easily calculated.

Preyer's method is also a spectrophotometric one. A 0.85 per cent. solution of oxyhæmoglobin (thickness of layer being 1 centimetre) is the most concentrated solution which allows the green light of the spectrum to pass through it (see fig. 58, sign). Take a known amount of the blood to be investigated and dilute it with water till it just allows the faintest shade of green light to pass through it. If b= volume of blood taken, and w=volume of water added, then the per0.85 (w+b) centage of oxyhæmoglobin in the blood= b

I append here a table of the results of analysis which I take from Preyer, Die Blutkrystalle,' p. 117.

100 grams of healthy human blood contain

From data given by Malassez it can be calculated that the amount of hæmoglobin in each human blood corpuscle is approximately 30 billionths of a gramme. Venous and arterial blood contain the same amount of hæmoglobin (Krüger).1 Foetal blood is of lower specific gravity than that of adults, and is especially deficient in hæmoglobin.2

Determination of the Activity of Reduction' of Oxyhemoglobin.3 The time of reduction of oxyhæmoglobin is determined by examining the spectrum of the blood under the thumb nail; the first band can be always, the second sometimes, distinguished by the direct vision spectroscope. If a ligature

1 Zeit. Biol. xxvi. 452. Krüger also states that congestion of a part increases the total solids and hæmoglobin in the blood drawn from it; that the blood of the splenic vein is richer, and of the renal vein is poorer, in solids and pigment than arterial blood. Copeman and Sherrington (Proc. Physiol. Soc. 1890, p. viii) have, by a different method, arrived at similar conclusions.

? Scherrenziss, Inaug. Diss. Dorpat, 1888. 3 Hénocque, Comptes rendus, ciii. 817.