The Bordet reaction was not only useful in indicating the specificity and blood rela­tionship for animals but also among plants. Thus Magnus and Friedenthal[47] were able to demonstrate with Bordet’s method the rela­tionship between yeast (Saccharomyces cerevisiæ) and truffle (Tuber brumale).

5. We must not forget, while under the spell of the problem of immunity, that we are interested at the moment in the ques­tion of the nature of the specificity of living organisms. It is only logical to conclude that the fossil forms of invertebrate animals and of algæ and bacteria, which Walcott found in the Cambrian and which may be two hundred million years old, must have had the same specificity at that time as they or their close relatives have today; and this raises the ques­tion: What is the nature of the substances which are responsible for and transmit this specificity? It is obvious that a definite answer to this ques­tion brings us also to the very problem of evolu­tion as well as that of the constitu­tion of living matter.

There can be no doubt that on the basis of our present knowledge proteins are in most or practically all cases the bearers of this specificity. This has been found out not only with the aid of the precipitin reac­tion but also with the anaphylaxis reac­tion, by which, as the reader may know, is meant that when a small dose of a foreign substance is introduced into an animal a hypersensitiveness develops after a number of days or weeks, so that a new injec­tion of the same substance produces serious and in some cases fatal effects. This hypersensitiveness, which was first analysed by Richet,[48] is specific for the substance which has been injected. Now all these specific reac­tions, the precipitin reac­tion as well as the anaphylactic reac­tion, can be called forth by proteins. Thus Richet, in his earliest experi­ments, showed that only the protein-containing part of the extract of actinians, by which he called forth anaphylaxis, was able to produce this phenomenon, and later he showed that it was generally impossible to produce anything resembling anaphylaxis by non-protein substances, e. g., cocain or apomorphin.[49] Wells isolated from egg-white four different proteins (three coagulable proteins and one non-coagulable) which can be distinguished from each other by the anaphylaxis reac­tion, although all come from the same biological object.[50] Michaelis as well as Wells found that the split products of the protein molecule are no longer able to call forth the anaphylaxis reac­tion. Since peptic diges­tion has the effect of annihilating the power of proteins to call forth anaphylaxis, we are forced to the conclusion that the first cleavage products of proteins have already lost the power of calling forth immunity reac­tions.

A pretty experiment by Gay and Robertson[51] should be mentioned in this connec­tion. Robertson had shown

that a substance closely resembling paranucleins both in its properties and its C, H, and N content can be formed from the filtered products of the complete peptic hydrolysis of an approximately four per cent. neutral solu­tion of potassium caseinate by the action of pure pepsin at 36°C.

He considered this a case of a real synthesis of proteins from the products of its hydrolytic cleavage. This interpreta­tion was not generally accepted and received a different interpreta­tion by Bayliss and other workers. Gay and Robertson were able to show that paranuclein when injected into an animal will sensitize guinea-pigs for anaphylactic intoxica­tion for either paranuclein or casein and apparently indiscriminately. The products of complete peptic diges­tion of casein had no such effect, but the synthetic product of this diges­tion obtained by Robertson’s method has the same specific antigenic properties as paranuclein, thus making it appear that Robertson had indeed succeeded in causing a synthesis of paranuclein with the aid of pepsin from the products of diges­tion of casein by pepsin.

There are a few statements in the literature to the effect that the specificity of organisms might be due to other substances than proteins. Thus Bang and Forssmann claimed that the substances (antigens) responsible for the produc­tion of hemolysis were of a lipoid nature, but their statements have not been confirmed, and Fitzgerald and Leathes[52] reached the conclusion that lipoids are non-antigenic. Ford claims to have obtained proof that a glucoside contained in the poisonous mushroom Amanita phalloides can act as an antigen. But aside from this one fact we know that proteins and only proteins can act as antigens and are therefore the bearers of the specificity of living organisms.

Bradley and Sansum[53] found that guinea-pigs sensitized to beef or dog hemoglobin fail to react or react but slightly to hemoglobin of other origin. The hemoglobins tried were dog, beef, cat, rabbit, rat, turtle, pig, horse, calf, goat, sheep, pigeon, chicken, and man.

6. It would be of the greatest importance to show directly that the homologous proteins of different species are different. This has been done for hemoglobins of the blood by Reichert and Brown,[54] who have shown by crystallographic measurements that the hemoglobins of any species are definite substances for that species.

The crystals obtained from different species of a genus are characteristic of that species, but differ from those of other species of the genus in angles or axial ratio, in optical characters, and especially in those characters comprised under the general term of crystal habit, so that one species can usually be distinguished from another by its hemoglobin crystals. But these differences are not such as to preclude the crystals from all species of a genus being placed in an isomorphous series (p. 327).