proteid, water being eliminated, this process of polymerism producing large and heavy but still simple molecules. In this process the nitrogen of the non-living proteid leaves the hydrogen with which it was combined in the form of amidogen (NH2), and enters into combination with carbon to form the more unstable substance cyanogen (CN). We thus find uric acid, creatine, guanine, &c., as products of proteid metabolism, while none of such cyanogen-containing bodies are obtainable from non-living proteids.

d. Loew's theory. The researches of Loew and Bokorny have taken the same direction as those of Pflüger, that is, they are attempts to explain the distinction between living and dead protoplasm. Living protoplasm or proteid in the cells of various alge has the property of reducing silver from a weak alkaline solution of silver nitrate; dead proteid has no such effect, and animal protoplasm is so quickly killed by silver nitrate that it also does not give the reaction. The conclusion arrived at is that something of the nature of an aldehyde occurs in living protoplasm. Formic aldehyde is probably formed in plants by the union of carbon and water; if this is united to ammonia, aspartic aldehyde is formed, thus:

and by further polymerisation in the presence of a sulphur compound and hydrogen, we get 6C12H17N3O4+H2S+6H2=C72H112N18SO22 +2H2O, which represents the composition of ordinary albumin. The weak point of the theory is that the aldehyde of aspartic acid is unknown to chemists; no doubt it is a most unstable substance. If such an aldehyde group does exist in living proteid, the instability of proteids is explicable, because molecular movements would be constantly occurring in the aldehyde group.

e. Latham's theory.-Latham 2 considers living proteid to be composed of a chain of cyanalcohols, or cyanhydrins as they are sometimes termed, united to a benzene nucleus.

1 Loc. cit

2 Brit. Med. Journ. vol. i. 1886, p. 629.

Cyanalcohols are substances obtained by the union of an aldehyde with hydrocyanic acid, thus:

Ethyl alcohol is taken as an instance in the above equations, but many other alcohols are considered to form similar cyan-derivatives, and these are united to one another and to benzene to form a proteid.

The theory is a satisfactory one, inasmuch as it includes the hypotheses both of Pflüger and Loew. Latham, moreover, shows exhaustively that the various products of the disintegration of albumin can also be obtained by the condensation and intramolecular changes that these cyanalcohols, which are exceedingly unstable bodies, undergo. Instability and proneness to undergo intramolecular changes are two properties common to living proteids and to cyanalcohols.

In an elaborate and painstaking manner Latham moreover adapts his theory to explain certain morbid processes; he shows how, by a rearrangement of atoms different from that occurring in normal metabolism, excess of sugar may be produced in diabetes, excess of uric acid in gout, and certain ptomaines 1 in other complaints.

We can now leave these theoretical considerations and pass on to consider matters of greater practical interest.

TESTS FOR PROTEIDS

Solubilities. All proteids are insoluble in alcohol and in ether. Some are soluble in water, others insoluble. Many of the latter are soluble in weak saline solutions. Some are insoluble, others soluble, in concentrated saline solutions. It is on these varying solubilities that proteids are classified.

All proteids are soluble with the aid of heat in concentrated mineral and acetic acids and caustic alkalis. Such treatment, however, decomposes as well as dissolves the proteid. Proteids are also soluble in gastric and pancreatic juices, but here again they undergo a change, being converted into a variety of proteids called peptones.

Heat-coagulation.-Many of the proteids which are soluble in water or saline solutions are rendered insoluble when those solutions are heated. The solidifying of white of egg under such circumstances is a familiar instance of heat-coagulation. Heat-coagulation must be very

1 Lancet, vol. ii. 1888, p. 751.

carefully distinguished from ferment coagulation-a process by means of which a ferment converts a previously soluble into an insoluble proteid; as instances of ferment coagulation the formation of fibrin in shed blood under the influence of fibrin-ferment, or of a curd of casein in milk under the influence of rennet, may be taken.

The temperature at which a proteid enters into the condition of a heat-coagulum is fairly constant, and may be employed as one of the means of ascertaining what proteid is present in a given solution. The temperature varies somewhat with the reaction of the solution,' with the quantity and nature of the salts also present, and, under certain circumstances, especially in an alkaline solution with the concentration of the solution.3

2

Unless a solution is very concentrated the contained proteid is not coagulated by heat in an alkaline solution, as it is converted into alkali-albumin; if the quantity of alkali is, however, very small, the temperature of heat-coagulation is raised. A neutral solution becomes alkaline after the separation of a heat-coagulum, and this alkalinity (produced no doubt by an alteration in the salts related to the proteid) may hinder the coagulation of the remaining proteid in the solution.

It is generally advisable to have the solution very faintly acid; a weak solution of acetic acid (2 per cent.) may be employed for the purpose of acidification.. Acid-albumin does not form so readily as alkali-albumin, and the presence of a small amount of acid renders easier the separation of the coagulated proteid into flocculi, which can be then removed by filtration. An excess of acid lowers the temperature of coagulation, or it may convert the proteid into acid-albumin and so prevent coagulation altogether.

The simplest method of ascertaining the temperature of heatcoagulation is to place enough of the solution in a test-tube to cover the bulb of a thermometer. The test-tube, the contents of which should be kept constantly stirred by the thermometer, is then placed in a flask containing water and situated over a Bunsen burner. As the temperature rises the point at which flocculi separate should be carefully noted; a few degrees below this point the liquid becomes thick and opalescent. A form of double water-bath consisting of two beakers one within the other is recommended by Gamgee, and Schäfer"

1 Halliburton, Journ. of Physiol. v. 165.

2 Limbourg, Zeit. physiol. Chem. xiii. 450.

3 Haycraft, Brit. Med. Journ. vol. i. 1890, p. 167.

4

4 Physiol. Chem. p. 15.

In my own work I have found certain inconveniences in the use of Gamgee's apparatus, and have therefore used Schäfer's. A description of it will be found in my paper in the Journ. Physiol. vol. v. p. 153.

has invented a very convenient form of running water-bath, the temperature of which can be easily changed (see fig. 44).

Fractional heat-coagulation may be sometimes used for the separation of proteids from one another. Suppose one had a solution of fibrinogen and serum-globulin together, the solution faintly acidified is raised to the temperature of 56° C. and at that point the fibrinogen is precipitated; this is filtered off; the filtrate is once more raised to

C

F

FIG. 44. The liquid of which the heat-coagulation temperature is to be determined is placed in a test-tube in quantity sufficient to cover the thermometer bulb; this is placed in the neck of the flask F. Hot water enters the flask by the tube a and leaves by the tube b. The water comes from the tap T and is warmed by passing through the coil of tubing contained in a copper vessel filled with boiling water; the more slowly the water passes the hotter does it become; the rate of flow of water can be regulated by the tap. In the figure the front of the ves: el C has been removed in order to show the coil of tubing within it.

56° in order to ascertain whether any fibrinogen remains in solution; if so, the heat-coagulum occurring at that temperature is once more filtered off and the filtrate again raised to 56°. When all the fibrinogen is removed, serum-globulin alone remains in solution, which is precipitated on raising the temperature to 75°. Serum-albumin 1 and eggalbumin 2 have also by this means been each differentiated into several proteids.

The proteids which are coagulated by heating their solutions come under two classes—the albumins, which are soluble in water and in weak saline solutions, and the globulins, which are insoluble in water and soluble in weak saline solutions.

The temperatures of coagulation of some of the principal proteids are as follows:

1 Halliburton, Journ. of Physiol. v. 159.

* Corin and Berard, Travaux du laboratoire de Léon Fredericq, Liège. ii. 170.

Indiffusibility. The proteids (with the exception of the peptones) belong to the class of substances called colloids by Thomas Graham. That is, they pass with difficulty, or not at all through animal membranes. In the construction of dialysers vegetable parchment is very largely used. Proteids may thus be separated from diffusible (crystalloid) substances like salts, but the process is a somewhat tedious one. The forms of dialyser used have been already described (p. 13). Take some serum, place it in a dialyser, and renew the water outside frequently. Some thymol crystals should be added to the serum to prevent the occurrence of putrefaction. The salts and extractions pass out through the membrane into the water, and the proteids alone remain within. The albumin is still in solution, but the globulin is precipitated as the salts which held it in solution have diffused out.

It is, however, found impossible even with the most prolonged dialysis to entirely remove all the salts which adhere to a proteid; so close is this adherence that one is inclined to believe that it is rather of the nature of loose chemical union. However carefully a proteid may have been purified, it always leaves on ignition a small quantity of ash, the composition of which varies in different cases, chlorides and phosphates of the alkaline metals and of calcium being the predominant constituents.

The term colloid does not necessarily imply that the substances in question are not crystallisable; for some of the vegetable proteids have been crystallised, and F. Hofmeister states that by carefully evaporating a solution of pure egg-albumin half saturated with ammonium sulphate, he has succeeded in obtaining that substance in a crystalline condition.

Action on polarised light.-All the proteids are lavorotatory. If pure and in solution they may be identified and estimated by means of their specific action on polarised light (see p. 41).

The specific rotations (for the yellow line D) of some of the principal proteids are as follows:

1 Zeit. physiol. Chem. xiv. 165.