In view of the marked differences in structure existing among these amino acids it becomes important to know the relative proportions in which the various amino acid radicles exist in the different proteins. This is studied by hydrolyzing the protein and separating and recovering as completely as possible the amino acids resulting from the hydrolysis. Since the recovery of the amino acids cannot be accomplished without loss, the results obtained are not strictly quantitative and our knowledge of the radicles which make up the protein molecule remains incomplete. It is believed by the investigators who have given most attention to the question that the failure of the recovered amino acids to show a summation of one hundred per cent is more probably due to unavoidable losses in estimating the known amino acids than to the presence of other amino acids not yet identified. The accompanying table shows the percentages of amino acids obtained from four proteins occurring in different food materials.

Percentages Of Amino Acids From Four Different Proteins

From the data given in the table it will be seen that the proportions in which a given amino acid radicle occurs in various proteins may be quite different. The four proteins here shown yield from o.o to 16.5 per cent of glycine; from 0.6 to 9.8 per cent of alanine; from 1.0 to 7.2 per cent of valine, from 6.6 to 19.6 per cent of leucine. Of lysine, zein yields none, gliadin about 1 per cent, gelatin 6 per cent, and casein about 8 per cent. Of tryptophane, zein and gelatin yield none, gliadin about 1 per cent, casein about 1.5 per cent.

* In general each figure given in the table is the highest of the results reported in recent investigations. This is deemed more accurate than to give average results, because of the unavoidable losses referred to above.

The data given for casein, gliadin, and zein are taken chiefly from the work of Osborne and his associates; those for gelatin chiefly from that of Skraup and Behler.

For more detailed comparisons of the percentages of amino acids in different proteins and also in the flesh of four widely separated species, the more extended table further on in this chapter may be consulted. Whether it be essential that the proteins of the food shall furnish all the amino acids which the body proteins contain will naturally depend upon whether the body is able to make individual amino acids or not. Experimental evidence has shown that the animal body can make glycine readily, so that the absence of glycine radicles in the food proteins does not detract from their nutritive value. On the other hand the animal body does not seem able to make tryptophane, and as this is an essential constituent of body tissue the food protein must always furnish tryptophane if it is to meet the needs of animal nutrition. Feeding experiments have also shown that the rate of growth of young animals may be largely influenced by the lysine content of the proteins fed; food proteins in which lysine is lacking or inadequate may suffice for the maintenance of full grown animals but fail to support normal growth in the young of the same species.

Such facts as these make it important that we study the proteins not only as a group but also individually and that we learn as much as possible about the kinds and amounts of amino acid radicles in the individual proteins.

The ultimate composition of the proteins shows a general similarity throughout the group. All contain carbon, hydrogen, oxygen, nitrogen, and sulphur; some also phosphorus or iron.

Composition Of Some Typical Proteins According To Osborne

It will be seen that all these typical proteins contain 51 to 55 per cent carbon, about 7 per cent hydrogen, 20 to 23 per cent oxygen, 15.5 to 18.7 per cent nitrogen, 0.3 to 2.0 per cent sulphur. Other typical proteins thus far studied have shown ultimate composition within these same limits.

Similarity of elementary composition is entirely consistent with the belief that there may be an enormous number of chemical individuals among the proteins of nature.

Fischer has recently illustrated the vast number of isomers which may exist among polypeptids and proteins by pointing out that a synthetic 19-peptid obtained by linking 15 glycine and 4 leucine molecules is only one of 3876 possible isomers, without considering the tautomerism of the pep-tid linking. When more than two kinds of amino acids are involved, the possible number of isomers increases very rapidly. If a protein be imagined made up of 30 molecules of 18 different amino acids, one taken twice, one 3 times, another 3, one 4, one 5 times, and 13 taken once each, there would be 1027 isomers even if there were no tautomerism of the peptid group and if the linking took place only in the simple way as with monamino-mono-carboxylic acids.

It is easy to see that when one considers not only isomerism but the vast number of compounds of slightly different composition which can be obtained by varying the kinds and proportions of the amino acid radicals in the protein molecule, the possible number of different proteins of very similar elementary composition is practically unlimited.

Probable Molecular Weights

From the results of ultimate analysis an approximate indication of the minimum molecular weight may often be obtained by a very simple calculation. Thus, oxyhemoglobin contains only 0.335 per cent of iron,and since there must be at least one iron atom in the molecule, it is obvious from a simple proportion making use of the atomic weight of iron, 0.335: 56:: 100 : x, that the molecular weight of hemoglobin must be in the neigh-borhood of 16,800 or a multiple of this.

To take an example from the simple proteins, zein contains 0.60 per cent of sulphur, of which one third is much more readily split off than the other two thirds, from which it appears that the molecule contains three, or a multiple of three, sulphur atoms. Then by the proportion, 0.60: (32 X 3):: 100: x, it is found that about 16,000 or a multiple thereof is the probable molecular weight of zein.

Estimates of the same order of magnitude are obtained if we base our calculations on the proximate rather than the ultimate analyses of the purified protein preparations. Osborne, Van Slyke, Leavenworth, and Vinograd have recently concluded from a very searching investigation that the lysine content of gliadin must lie between 0.64 and 1.20 per cent. Since the molecular weight of lysine is 146 it follows that the corresponding minimum estimate of the molecular weight of gliadin must fall between 12,000 and 23,000. The experimental facts do not permit the assumption of any lower molecular weight but are not inconsistent with the view that the true molecular weight may be some multiple of this.

Physical Properties

In only a few cases have proteins been obtained in crystalline form. Generally speaking the proteins may be regarded as typically colloidal substances. This does not preclude the belief that in the tissues and fluids of the body the proteins may exist largely in combination with electrolytes. In view of the fact that the behavior of proteins in the tissues is largely dependent upon their colloidal character it is of interest to bear in mind the very high molecular weights of the proteins as mentioned in the last paragraph. Discussions of colloids commonly emphasize the fact that the smallest particles demonstrable under the ultramicroscope must still be of quite a different order of magnitude from that calculated for ordinary molecules. In such a case as that of starch or a typical protein, however, the probable molecular weight is so enormous as to make it a debatable question whether the individual molecules may not constitute colloidal particles when dispersed in water (Bayliss).

The proteins are insoluble in all of the solvents for fats (ether, acetone, chloroform, carbon disulphid, carbon tetrachlo-rid, benzene, and petroleum distillate). They differ in their solubilities in water, salt solutions, and alcohol, and these differences play a considerable part in the present schemes of classification.