He also showed that the development of artificial immunity is attended by the appearance of phagocytosis; also, when an anti-serum is injected into an animal, the phagocytes which formerly were indifferent might move towards and destroy the bacteria. In the light of all the facts, however, especially those with regard to anti-bacterial sera, the presence of phagocytosis cannot be regarded as the essence of immunity, but rather the evidence of its existence. The increased ingestion of bacteria in active immunity would seem to depend upon the presence of immune opsonins in the serum. These, as already explained, are true anti-substances. Thus the apparent increased activity of the leucocytes is due to a preliminary effect of the opsonins on the bacteria. We have no distinct proof that there occurs in active immunity any education of the phagocytes, in Metchnikoff's sense, that is, any increase of the inherent ingestive or digestive activity of these cells. There is some evidence that in certain cases anti-substances may act upon the leucocytes, and to these the name of "stimulins" has been given. We cannot, however, say that these play an important part in immunity, and even if it were so, the essential factor would be the development of the substances which act in this way.

While in immunity there probably occurs no marked change in the leucocytes themselves, it must be admitted that the increased destruction of bacteria by these cells is of the highest importance. This, as already pointed out, depends upon the increase of opsonins, though it is also to be noted that in many infective conditions there is another factor present, namely a leucocytosis, that is, an increase of the leucocytes in the blood, and the defensive powers of the body are thereby increased. Evidence has been brought forward within recent years that the leucocytes may constitute an important source of the antagonistic substances which appear in the serum. Much of such evidence possesses considerable weight, and seeing that these cells possess active digestive powers it is by no means improbable that substances with corresponding properties may be set free by them. To ascribe such powers to them exclusively is, however, not justifiable. Probably the lining endothelium of the blood-vessels as well as other tissues of the body participate in the production of anti-substances.

The subject of artificial immunity has occupied a large proportion Natural immunity. of bacteriological literature within recent years, and our endeavour has been mainly to indicate the general laws which are in process of evolution. When the facts of natural immunity are examined, we find that no single explanation is possible. Natural immunity against toxins must be taken into account, and, if Ehrlich's view with regard to toxic action be correct, this may depend upon either the absence of chemical affinity of the living molecules of the tissues for the toxic molecule, or upon insensitiveness to the action of the toxophorous group. It has been shown with regard to the former, for example, that the nervous system of the fowl, which possesses immunity against tetanus toxin, has little combining affinity for it. The non-sensitiveness of a cell to a toxic body when brought into immediate relationship cannot, however, be explained further than by saying that the disintegrative changes which underlie symptoms of poisoning are not brought about.

Then as regards natural powers of destroying bacteria, phagocytosis aided by chemiotaxis plays a part, and it can be understood that an animal whose phagocytes are attracted by a particular bacterium will have an advantage over one in which this action is absent. Variations in chemiotaxis towards different organisms probably depend in natural conditions, as well as in active immunity, upon the opsonic content of the serum. Whether bacteria will be destroyed or not after they have been ingested by the leucocytes will depend upon the digestive powers of the latter, and these probably vary in different species of animals. The blood serum has a direct bactericidal action on certain bacteria, as tested outside the body, and this also varies in different animals. Observations made on this property with respect to the anthrax bacillus at first gave the hope that it might explain variations in natural immunity. Thus the serum of the white rat, which is immune to anthrax, kills the bacillus; whereas the serum of the guinea-pig, which is susceptible, has no such effect. Further observations, however, showed that this does not hold as a general law. The serum of the susceptible rabbit, for example, is bactericidal to this organism, whilst the serum of the immune dog is not.

In the case of the latter animal the serum contains an opsonin which leads to phagocytosis of the bacillus, and the latter is then destroyed by the leucocytes. It is quite evident that bactericidal action as tested in vitro outside the body does not correspond to the degree of immunity possessed by the animal under natural conditions. We may say, however, that there are several factors concerned in natural immunity, of which the most important may be said to be the three following, viz. variations in the bactericidal action of the serum in vivo, variations in the chemiotactic or opsonic properties of the serum in vivo, and variations in the digestive properties of the leucocytes of the particular animal. It is thus evident that the explanation of natural immunity in any given instance may be a matter of difficulty and much complexity.

Authorities

Bacteriological literature has become so extensive that it is impossible to give here references to original articles, even the more important. A number of these, giving an account of classical researches, were translated from French and German, and published by the New Sydenham Society under the title Microparasites in Disease: Selected Essays, in 1886. The following list contains some of the more important books published within recent years. Abbott, Principles of Bacteriology (7th ed., London, 1905); Crookshank, Bacteriology and Infective Diseases (with bibliography, 4th ed., London, 1896); Duclaux, Traité de microbiologie (Paris, 1899-1900); Eyre, Bacteriological Technique (Philadelphia and London, 1902); Flügge, Die Mikroorganismen (3rd ed., Leipzig, 1896); Fischer, Vorlesungen über Bakterien (2nd ed., Jena, 1902); Günther, Einführung in das Studium der Bakteriologie (6th ed., Leipzig, 1906); Hewlett, Manual of Bacteriology (2nd ed., London, 1902); Hueppe, Principles of Bacteriology (translation, London, 1899); Klein, Micro-organisms and Disease (3rd ed., London, 1896); Kolle and Wassermann, Handbuch der pathogenen Mikroorganismen (Jena, 1904) (supplements are still being published; this is the most important work on the subject); Löffler, Vorlesungen über die geschichtliche Entwickelung der Lehre von der Bacterien (Leipzig, 1887); McFarland, Text-book upon the Pathogenic Bacteria (5th ed., London, 1906); Muir and Ritchie, Manual of Bacteriology (with bibliography, 4th ed., Edin. and Lond., 1908); Park, Pathogenic Micro-organisms (London, 1906); Sternberg, Manual of Bacteriology (with full bibliography, 2nd ed., New York, 1896); Woodhead, Bacteria and their products (with bibliography, London, 1891). The bacteriology of the infective diseases (with bibliography) is fully given in the System of Medicine, edited by Clifford Allbutt, (2nd ed., London, 1907). For references consult Centralbl. für Bakter. u.