In the preceding chapters the phenomena of immunity which are exhibited within the animal body in which the portals were open for the penetration of the micro-organisms and their poisons have been studied. We had to deal almost exclusively with experimental immunity, the study of which constitutes the basis of our present knowledge concerning the general problem of immunity. In natural immunity, however, things do not follow the same course. The micro-organisms and their toxins are not introduced directly into the tissues and blood by means of a syringe or other instrument. The micro-organisms have to make their own way through the skin and the mucosae, tissues which offer a resistance more or less serious and effective; or they may have to take up their abode in the cavities of the animal organism, in order that they may be able to inundate it with their poisons. We must here review briefly these natural barriers to microbial invasion.

The skin constitutes a protective covering of great importance in connection with the preservation against microbial invasion of the delicate parts of an animal. In many of the lower and higher animals, and even in man himself, the skin becomes the seat of a microbial flora, often very abundant, in which may be found, in addition to certain inoffensive organisms, other minute parasites more or less harmful. The pyogenic cocci, staphylococci and streptococci, are constantly found on the human skin, most frequently hidden in the depths of the canals of the hair follicles. These micro-organisms seize every favourable opportunity to attack the organism, producing such local lesions of the skin as acne, pimples, boils, and erysipelas, or even becoming generalised in the blood and tissues, as in the septicaemias and pyaemias. To the skin, therefore, must be assigned a very important function in the prevention of the invasion of micro-organisms which are found constantly on the surface of the body or which, along with all kinds of dirt, are brought there accidentally.

The skin is able to fulfil this protective function from the fact that, in most animals, it is covered with a not very permeable layer of some considerable thickness. In the majority of the Invertebrata, of all classes, the surface of the body is clothed with a chitinous layer, sometimes very thin and capable of folding and following all the movements of the body; or again it may be impregnated with calcareous salts and very hard, as in the case of the integument of Insects and Crustacea, and the shell of the Mollusca. In all cases this cutaneous sheath constitutes a formidable obstacle to the entry of micro-organisms. Even in animals of very small size the thin cuticle is effective in preventing any invasion by these parasites. Thus the Saprolegniae, fungi so fatal to many aquatic animals, are often quite unable to pass through this cuticular layer. In order to pass this obstacle their germs must take advantage of some fissure or wound, produced by other causes. Daphniae, too, may often be observed to succeed in ridding themselves of the Monospora with its needle-like spores by means of a mechanism which we have already described in chapter VI. The white corpuscles of the blood surround the spores of this parasite and transform them into an innocuous detritus. Sometimes, however, a number of these fine spores manage to perforate the cutaneous investment of the small crustacean; quite a small opening is made in the chitinous wall, which in itself is a source of no danger. As soon, however, as a spore of the Saprolegnia approaches this opening, it immediately begins to thrust a process through the small lesion, and from that moment the fate of the Daphnia is sealed. Incapable of opposing the slightest phagocytic resistance to the filaments of the fungus, it is invaded throughout by the mycelium and soon dies.

The integrity of the skin being so important for the preservation of life, a fairly perfect mechanism has been elaborated for the maintenance of this integrity. All animals, no matter what their position in the animal scale, are liable to lesions and wounds of the surface of their bodies. In the Daphniae I have often^[656] observed wounds produced by the bites of other aquatic animals. The surface of these wounds soon becomes covered with a rich microbial vegetation. The leucocytes are brought up to the injured point and there produce a protective layer; but, at the same time, a rapid proliferation of the neighbouring cells of the epidermis takes place; this closes the wound and separates the skin, so reconstituted, from the micro-organisms. Everything resumes its original order and the leucocytes soon disperse, regaining the blood stream.

These phenomena, which can be readily observed under the microscope in such small and transparent animals as the Daphniae, may serve as the prototype of those of a number of analogous processes in the animal kingdom. The thicker and more solid the cuticular investment, the more fully it guarantees the animal against the penetration of micro-organisms. Cuénot^[657] made the observation that Crustacea, furnished with such a hard envelope as is the carapace of the Decapods, are completely defenceless from the moment parasitic micro-organisms make their way into their bodies. These intruders quietly instal themselves in the tissues, without causing the slightest phagocytic reaction, and thus bring about the inevitable death of the host. The protection of the animal in this case is, so to speak, associated with the resistance offered by the carapace.

Again, in many of the Vertebrata, the skin has a hard, thick sheath, e.g., the scales of fishes and of reptiles. Man, with his supple and not very thick skin, is less well endowed; this, however, does not prevent him from defending himself against the entry of micro-organisms by the cutaneous path. Sabouraud^[658], a well-known dermatologist, has given a very concise and at the same time very complete sketch of the part played by the skin in the protection of the body against micro-organisms; from this author the following data are borrowed.

The epidermic layer sets up a defence by the production and expulsion of corneal cells. In the normal course of the life of the epidermis, the cells of the deeper layers, coming to the surface, become exfoliated and are thrown off. “There is thus produced, a continual exfoliation of the dead layers, and a continual eviction of such micro-organisms as are living on them. The epidermis is dense and its cells have a hard envelope; the micro-organism is not endowed with motion, or at least not with sufficient to be of service in effecting an entrance. It can only penetrate the epidermis by multiplication in situ, a micro-organism originates alongside another, another in front of it, and in front of this again others. In this way they burrow between the apposed cells just as a root penetrates into the ground; so great is the resistance of the horny cells that we never find any micro-organisms within them, but between them only” (p. 734). The epidermic cells, containing micro-organisms, exfoliate, and the skin is thus ridded of them. Frequently the process, as it goes on constantly and slowly, is invisible; but often, on the other hand, it becomes exaggerated and manifests itself in the form of a cuticular desquamation which leads to the elimination of a large number of micro-organisms. The patient may retain “such pellicles for ten years, and even longer, without presenting anything else but these, and there are many other chronic squamous infections in which the course is uncomplicated by even an erosion or the slightest wound.”

The connective tissue of the human skin is also fully able to defend itself; it is extremely vigorous and represents a real obstructing and resisting tissue. The penetration of parasites sets up in it a thickening of the fibrous tissue; this effects a localisation of the microbial focus. To appreciate the effectiveness of this dermic defence, we have only to compare the slow growth of lupus, a form of cutaneous tuberculosis, with that of tuberculosis of the lungs and other viscera, or the slow evolution of farcy, or cutaneous glanders, with that of the visceral form of the disease.

If we examine more closely the process by which the dermis surrounds the intruders with a fibrous capsule, we readily recognise in it a reaction of the macrophages of the skin. In lupus these phagocytes seize the tubercle bacilli, combining to form giant cells and giving rise to an exaggerated development of the connective tissue fibres. Moreover, when the skin is menaced with a microbial invasion, not only the local macrophages but the leucocytes are mobilised. The migratory white corpuscles travel through the epidermis and the connective tissue layer. In spite of the absence of a lymphatic circulation in the epidermis, the leucocytes penetrate into this layer “and, in a section through the normal epidermis, it is very rare not to find here and there some deformed and flattened leucocyte, surprised just as it was creeping between the cells of the rete mucosa or of the stratum granulosum.” Immediately that the epidermis or the dermis finds itself menaced with a microbial invasion, an accumulation of leucocytes of all kinds is produced at once; this may remain microscopic or it may assume proportions visible to the naked eye. Frequently the subjacent epithelium throws off epidermic scales which are filled with leucocytes; often also the leucocytic foci in the dermis become emptied, the micro-organisms being expelled along with their enemies the phagocytes.

The tissues of the skin proper defend themselves against micro-organisms as well as they are able; but so soon as the danger becomes serious there is sent to their succour a whole army of mobile phagocytes. This example of the defence made by the cutaneous investment may serve as a prototype of that of every other region of the body. Alongside a local action, there is always an intervention of mobile phagocytes; but when this action becomes insufficient, a much more abundant accumulation of leucocytes than is found in ordinary cases is immediately produced.

Like the skin, the mucous membranes are invested with an epithelial layer, which serves as a barrier to the entry of micro-organisms. But whilst the surface of the normal skin is dry or barely moistened by the secretory products of the cutaneous glands, the mucous membranes are always humid, a condition favourable to the multiplication of micro-organisms. Hence the mucous membranes which are most exposed to contact with the air and with external objects, always contain a larger or smaller number of organisms, amongst which the pathogenic species, notably staphylococci, pneumococci and streptococci, are the most common. The part played by the animal organism in getting rid of these micro-organisms becomes more complicated than in the case of the defence made by the skin.

The first of the mucous membranes to be exposed to contamination by micro-organisms is the conjunctiva of the eye. At the moment of birth it is in contact with the vaginal mucous membrane and acquires from it some of its micro-organisms, both innocuous and pathogenic. Tears fulfil the function of averting the danger resulting from this proximity and from the presence of micro-organisms in the conjunctival sac in general. Ophthalmologists have shown that these tears transport the organisms into the nasal cavity by means of the lachrymal canal. To determine this point Bach^[659] introduced a number of Kiel water bacilli along with pyogenic staphylococci into the conjunctival sac of various individuals. Seedings made with the tears showed a very rapid disappearance of the two organisms, which passed into the nose where their presence could be demonstrated by making plate cultures of the nasal mucus. Enormous numbers of the Kiel bacilli, introduced into the conjunctival sac, were all transferred to the nasal cavity, on the average, by the end of half-an-hour. The pyogenic staphylococci persisted on the surface of the conjunctiva for a longer period, but they also passed in large numbers through the lachrymal canal into the nose.

Certain observers, notably Bernheim^[660], thought that the tears, in addition to their purely mechanical defensive action, were capable of destroying the micro-organisms by their microbicidal power. Bach^[661] submitted this question to a minute examination and came to the conclusion that several species of bacteria, introduced in vitro into the tears of healthy persons or of those who were suffering from conjunctivitis or certain other ocular diseases, disappeared somewhat rapidly. Comparative experiments with tears previously heated to 58° and even to 70° C., in most cases gave the same results, that is to say, they caused a rapid disappearance of the organisms introduced. From these facts the author concluded that it is probably to the salts contained in the tears that their bactericidal action is due. Control experiments made with physiological saline solution and with various mixtures of mineral salts met with in the tears have been found by Bach to cause a like disappearance of the same species of organisms. Well water, and even distilled water, gave the same result. In all these cases it is evident that, in the tears, there is no bactericidal cytase comparable with that found in the serums and other body fluids which may contain this phagocytic diastase. The experiments with heated tears demonstrate this clearly. On the other hand, these same experiments lead one to suppose that the diminution and even the disappearance of the micro-organisms in the tears, is due to a large extent, and perhaps completely, to an agglutinative action of the salts, a fact which has been demonstrated by several observers.

In all these cases it is indisputable that the mechanical part played by the tears is the most important of the defences offered by the conjunctiva of the eye against the microbial invasion. That this defence is not always sufficient is proved by the frequency of conjunctivitis, as well as by the ease with which certain micro-organisms, inoculated into the conjunctival sac, set up a general infection. This is specially the case with the coccobacillus of human plague. When it is introduced into the conjunctival sac of susceptible animals (rat, guinea-pig, &c.), it passes thence into the nasal cavity and soon produces a generalised and fatal infection. The conjunctival membrane, even when perfectly intact, readily absorbs certain poisons. Everyone knows the rapidity with which atropin, when introduced into the conjunctival sac, causes dilatation of the pupil. But the mucous membrane may serve also as the port of entry for toxins of microbial origin. Several observers, and especially Morax and Elmassian^[662], have demonstrated that the diphtheria poison placed upon an unbroken conjunctival membrane, where the epithelial layer is uninjured, sets up local lesions which progress very slowly but which terminate in the formation of actual false membranes. Nevertheless, it must be admitted that the intact epithelial layer of the conjunctiva exerts a certain defensive action against the penetration of toxins, although a very slight lesion of this layer will allow of the ready absorption of the diphtheria poison and the formation of false membranes.

The cornea likewise, so long as it is intact, exhibits a marked resistance against the penetration of micro-organisms and of toxins. When it becomes injured in any way its epithelium is repaired with great rapidity, as has been well demonstrated by Ranvier^[663], who has shown that the walls of the wound close by a process of epithelial “soldering” in a purely mechanical fashion, without the intervention of any preliminary proliferation of the epithelial elements. Thanks to this very rapid obliteration the micro-organisms are prevented from penetrating not only into the interior of the cornea, but into the anterior chamber of the eye.

It has already been pointed out that the ocular conjunctiva gets rid of the introduced micro-organisms chiefly by removing them mechanically and sending them through the lachrymal duct into the nasal cavity. This, in turn, defends itself by making use of a similar method. In his experiments on the Kiel red bacillus, inoculated into the conjunctival sac of man, Bach demonstrated that in a very short time these micro-organisms are carried into the nasal cavity. He showed also that they do not remain long in the latter position and that their number decreases hourly.

Twenty-four hours after the introduction of these bacilli into the conjunctiva none, as a general rule, are to be found in the nasal mucus. This expulsion of the micro-organisms likewise takes place by mechanical means, aided by the movements of the vibratile cilia. It is evidently to this process that the mucous membrane owes its relative freedom from micro-organisms. Frequently, when examining the nasal mucus or when making cultures therefrom, one is astonished at the small number of micro-organisms found in the nasal cavities of persons in good health. Thomson and Hewlett^[664] have certainly gone too far when they affirm that the upper regions [i.e. the Schneiderian membrane] of the nasal cavity are, in almost 80% of cases, free from micro-organisms. But it is certain that in these regions we do find a small number only of the bacteria which exist in greater abundance in the lower (cutaneous) passages of the nose.

To explain this paucity of micro-organisms in the nasal cavity, Wurtz and Lermoyez^[665] have assumed the existence of a bactericidal property in the nasal mucus. They affirm that the anthrax bacillus, after contact with this mucus for several hours, loses its virulence for the most susceptible animals, and that several other micro-organisms—the staphylococci, the streptococci, and the Bacillus coli—become attenuated under the same conditions. Others who have studied this question have come to a different conclusion. Thomson and Hewlett found that the nasal mucus is not bactericidal, although it prevents the multiplication of micro-organisms. F. Klemperer,^[666] denies the bactericidal property of the nasal mucus. He could never demonstrate the destruction of micro-organisms by the mucus, and he also observed that bacteria do not multiply at all readily in this medium. These results confirm the hypothesis that the defensive action of the nasal mucous membrane against microbial invasion is mainly effected by the mechanical elimination of the numerous micro-organisms which continually reach it. Amongst these organisms are some which are conspicuous for the ease with which they multiply in the body, taking the nasal cavity as a starting point, e.g. the micro-organisms of influenza, the plague bacillus, which, according to several observers, is very virulent when introduced by the nostrils^[667], and the leprosy bacillus. This last, according to Goldschmidt^[668], Sticker^[669], and Jeanselme^[670] often enters the human body by way of the nose.

It is certain that the olfactory apparatus deprives the inspired air of a large number of the micro-organisms which it carries. These organisms deposited on the mucous membrane are expelled with the nasal mucus. A number of the foreign organisms, carried by the air, may, however, surmount this first barrier and penetrate further into the trachea and bronchi, whence, helped by the movements of the vibratile cilia, they are usually expelled along with the mucus.

In spite of this double defence it has been recognised that very minute corpuscles and, amongst others, micro-organisms may overcome every one of these obstacles and reach the pulmonary alveoli. Here, under the name of “dust-cells” (“cellules à poussière”)—“Staubzellen” of the German writers—located in the alveoli, are described certain large mononucleated elements which contain granules of foreign origin, usually deposits of soot, of a deep black. This permeability of the normal lung tissue for dust particles and pigmented corpuscles has been closely studied and clearly demonstrated by J. Arnold^[671] and his pupils. Several observers have tried to determine whether micro-organisms, introduced by the respiratory channels, behave like other bodies. Animals were made to inhale, or there were introduced into the trachea, cultures of bacteria pathogenic for the animals experimented upon. The results so obtained have been very contradictory. Morse^[672], Wyssokowitch^[673], and Hildebrandt^[674], never succeeded in inducing anthrax by the introduction of anthrax bacilli into the lungs of normal animals. They concluded, therefore, that the uninjured pulmonary tissue was impermeable by virulent micro-organisms. H. Buchner^[675] with his collaborators and pupils maintaining the opposite view, declare that rabbits that have inhaled anthrax bacilli or their spores always succumb to a fatal attack of anthrax. These contradictory results were attributed to differences in the methods employed, and an attempt was made to perfect the methods of research, especially to prevent the penetration of the anthrax bacilli by lesions of the trachea or by any channel other than that of the pulmonary tissue. Gramatschikoff^[676], under Baumgarten’s direction, undertook a series of experiments in order to determine whether it was possible for the anthrax bacillus to traverse the pulmonary tissue. He introduced through the trachea of rabbits and guinea-pigs an anthrax culture, afterwards washing the respiratory passages with a large quantity of broth or of physiological saline solution. Several of the animals so treated did not succumb to the inoculation, and Gramatschikoff concluded that it was impossible for the anthrax bacillus to make its way through the wall of the normal pulmonary tissue. He was satisfied that some of the injected organisms were destroyed in the lung, although he was unable to see how this bactericidal action was determined. In these experiments a large quantity of fluid was introduced after the bacilli; this might wash away the bacilli and convey them to situations where they could exert no morbific action; moreover the anthrax bacilli used were of doubtful virulence (the injections made to control the virulence in the subcutaneous tissue were in nearly every instance made with quantities of fluid greater than those introduced by the trachea), and Gramatschikoff’s results could not be accepted as deciding the question. On the other hand, H. Buchner’s inhalation experiments made with spores, and the study of the organs of animals so treated, leave no doubt as to the possibility of the invasion of an animal by the respiratory channels by the anthrax bacillus. Furthermore, the “rag-picker’s disease” and the “wool-sorter’s disease,” or pulmonary anthrax, developed in man as a result of the inhalation of dust charged with anthrax spores, demonstrate most clearly that it is possible for the anthrax bacillus to enter the body by the respiratory channels. The pulmonary mycoses, produced by the penetration of the Aspergillus fumigatus in the human subject, offer confirmatory evidence.

In spite of the fact that the pulmonary tissue is not impermeable to pathogenic micro-organisms, it is none the less true that it exhibits a very marked resistance to infection by this channel. It is, however, neither the thickness of the wall, as in the case of the skin and the mucous membranes, nor the mechanical elimination with the help of the vibratile cilia or of the secretions, that constitute the means of defence in the respiratory alveoli. Here the cell elements are charged with the duty of ridding the lungs as much as possible of the micro-organisms which enter. Ribbert^[677] and his Bonn pupils, Fleck^[678] and Laehr^[679], observed this fact long ago. They showed that the spores of Aspergillus flavescens and the staphylococci, injected into the veins or into the trachea, penetrate into the pulmonary alveoli, where they are soon seized by the “epithelial cells” and the leucocytes. Laehr observed that this phenomenon is produced at the end of a few hours, and that the ingested cocci within the phagocytes undergo a progressive degeneration and at last disappear. Tchistovitch^[680], working in my laboratory, studied micro-organisms pathogenic for the rabbit—the anthrax bacillus, the coccobacillus of fowl cholera, and the bacillus of swine erysipelas—ingested by the “dust-cells” of the alveoli. He has added the important observation (already referred to in chapter IV) that these phagocytic elements are not epithelial cells at all, but are really macrophages of lymphatic origin. They are not found in the alveoli of new-born animals, but soon appear there and instal themselves in such a manner that for long one was led to regard them as true epithelial cells of the pulmonary tissue. This tissue, invested with an extremely delicate covering, is incapable of defending itself against the invasion of micro-organisms, but the animal organism comes to its aid by sending a permanent army of macrophages which evict from the alveoli, so far as is possible, both micro-organisms and other foreign bodies. Under these conditions, we can readily understand that similar cells which fulfil the same protective function, are also found in the neighbouring bronchial glands. It has long been recognised that the macrophages of these glands are often crammed with various kinds of granules of foreign origin, which have made their way into the lungs with the inspired air.

Toxic substances can be absorbed by the mucous membrane of the respiratory channels. Roger and Bayeux^[681] have shown that no lesion is required in order that diphtheria poison may invade the mucous membrane of the trachea, and so produce typical false membranes. The lung, we know, is accessible to gaseous toxic substances; moreover, its surface readily absorbs fluid poisons.

The protection of the digestive system is more complex than that of the respiratory passages; this is not remarkable, when we consider the greater complexity of the organs of digestion and the varied conditions which they present with regard to microbial invasion.

The buccal cavity, so exposed to the entry of extraneous micro-organisms along with the food and the external air, has a very rich microbial flora, in which Miller^[682], the author of our most complete work on this subject, has recognised in man more than thirty species. Several representatives of this flora, e.g. the Leptothrix and the Spirochaeta are constantly present, and are very characteristic of the buccal cavity of man. With them are frequently found pneumococci, staphylococci, and streptococci, whose pathogenic power is undoubted. Virulent diphtheria bacilli are also met with in a certain number of quite healthy persons. It is astonishing that, in spite of this state of things, wounds in the mouth heal very rapidly, and operations on the buccal cavity done with insufficient or no aseptic precaution do not, in the great majority of cases, set up infective complications of the slightest importance. After certain buccal operations we are often confronted with a complicated and open fissure; nevertheless the wound thus left exposed is not ordinarily the seat of any infection either local or generalised.

It is often asked, how under these conditions does the mouth defend itself against the vast number of formidable micro-organisms. When the theory of the bactericidal power of the body fluids was dominant, and appeared to explain several important points in the general problem of immunity, the saliva was studied from this “bactericidal” point of view. Sanarelli^[683], as the outcome of patient and laborious researches, came to the conclusion that the human saliva acted as an antiseptic and destroyed a large number of micro-organisms. It is true that he recognised its efficacy only when few bacteria were subjected to its action; but even when the saliva was incapable of killing a large number of micro-organisms, it did not allow them to develop—it was a bad culture medium; moreover, it had the power of attenuating the virulence of certain pathogenic bacteria, such as the pneumococcus, so frequently found in the mouth.

The conclusions of the Italian observer did not, however, meet with general acceptance. Miller^[684] did not believe that the saliva exerted any bactericidal action, raising the objection that the absence of nutritive value in the human saliva for bacteria is explained by the fact that in his experiments Sanarelli employed filtered saliva, which consequently had been deprived of much of its nutritive substances,—epithelial débris, mucus, etc. Hugenschmidt^[685], working in my laboratory, carried out a special research on the influence of the human saliva on micro-organisms, and arrived at conclusions quite at variance with those reached by Sanarelli. In spite of the variety of micro-organisms made use of, he could never satisfy himself that the saliva had any bactericidal property.

He sometimes saw, no doubt, a certain slowness of growth or even the destruction of certain of the micro-organisms sown at the commencement of the experiment, but this was very slight and rather exceptional. In most cases the micro-organisms, introduced into the saliva, grew rapidly, so that their number, in a very short time, became very considerable. Where the saliva brought about any diminution in the number of micro-organisms, this semblance of bactericidal action could be noted not only in the normal saliva, but; also, as in the lachrymal secretion above described, in saliva heated to 60° C. Against certain micro-organisms—the torulae and the staphylococci—the heated saliva acted more vigorously than did the unaltered saliva. It is consequently impossible to draw any parallel between the action of the saliva and that of the cytases.

Since the saliva often contains (according to certain authors even constantly) small quantities of potassium sulphocyanide, it seemed to be worth while to ascertain whether this salt is capable of destroying micro-organisms. The experiments carried out by Hugenschmidt, in order to settle this point, demonstrated that when given in doses comparable to those met with in the saliva, the potassium sulphocyanide exerts no bactericidal action.

Powerless as an antiseptic, the saliva fulfils an important function in ridding the mouth of micro-organisms in a mechanical way. The parotid secretion and that of the other salivary glands dilutes the bacteria and carries them from the pharyngeal cavity into the stomach. Hence, in diseases where the salivary secretion is much diminished, the mouth becomes the most important portal of entry for micro-organisms capable of setting up secondary infections. The saliva is further useful in diluting the alimentary detritus and preventing its stagnation and decomposition in the buccal cavity.

In addition to the direct mechanical part played by the saliva, it performs a very important indirect function. This fluid contains microbial products and diastases, and is capable of exciting in the leucocytes a positive chemiotactic activity. Hugenschmidt demonstrated the fact by introducing into animals small capillary glass tubes containing saliva. A certain time after being placed in position, these tubes became filled with considerable masses of immigrated leucocytes. The same result was obtained with guinea-pig’s saliva, enclosed in capillary tubes and introduced into the peritoneal cavity of the same species. Here, again, the leucocytes assembled in the tubes and ingested the micro-organisms found in the saliva. The influence of the saliva on the afflux of the leucocytes must be regarded as an act important for the protection of the buccal cavity, and it is probably due to this attraction of leucocytes that lesions of this region heal so quickly. The leucocytes are very numerous in the glands of the mouth and the tonsils always supply large quantities of them.

We must not lose sight of the fact that the epithelial covering of the bucco-pharyngeal cavity also constitutes an important protective factor. Just as on the surface of the skin, the corneal cells are in a permanent state of desquamation, so the cells in the mouth are being constantly renewed. This desquamation increases especially during mastication, when enormous numbers of cells are thrown off; after every meal there is a partial renewal of the surface of the lining of the buccal cavity. Being covered on their surface, and in their interstices charged with innumerable micro-organisms, the epithelial cells carry away with them all this population from the mouth.

The numerous micro-organisms which persist in the mouth, in spite of all these means for getting rid of them, must also play a certain part in the defence against infections. It is very probable that many of these saprophytes impede the multiplication of certain pathogenic bacteria; but at present it is impossible to define more exactly these phenomena of microbial competition. It is only because we have analogies in other regions of the body that we are able to defend this position.

The saliva, incapable of destroying the micro-organisms themselves, is able to act on their soluble products, as on certain other poisons. In this relation the action of the saliva on snake venom is most familiar. Wehrmann^[686], who has made researches on this subject in Calmette’s laboratory at Lille, has shown that the amylase (ptyalin) of human saliva, mixed with very rapidly fatal doses of venom, quite prevents its toxic action. Von Behring^[687] reminds us on this point that the ancient Psylli (a race of northern Africa), at the beginning of our era, employed their saliva as an antidote against snake bites.

Powerless to kill the micro-organisms, the saliva carries them off mechanically to the exterior or, more frequently, into the stomach. The acid medium of this great reservoir exerts a very marked effect on these microscopic organisms. It has long been recognised that the gastric juice prevents putrefaction and can arrest it even when it has become very advanced. From this observation an antiseptic action of this juice was inferred. Bacteriological researches, undertaken to determine the nature of this action, have demonstrated that several species of micro-organisms die very shortly after being placed in contact with the gastric juice in vitro. Straus and Wurz^[688] found that even anthrax spores and the tubercle bacillus could be destroyed by gastric juice, after a prolonged sojourn in a sufficient quantity of this fluid. Comparative researches, made with aqueous solutions of hydrochloric acid, have demonstrated that the bactericidal action of the gastric juice depends solely on the amount of this acid that it contains, that is to say, the pepsin plays no part in the process. This juice exerts no true digestive action on the micro-organisms, but it destroys a certain number of them by its hydrochloric acid. This antiseptic action may also be inferred from a series of demonstrations on the exaggerated microbial multiplication in cases where the gastric juice has been poor in hydrochloric acid. Several observers have confirmed this bactericidal action of the gastric juice which is exerted specially against certain species capable of causing grave infective diseases. On the other hand, certain bacteria and other lower fungi are quite resistant to the antiseptic action of this fluid; they adapt themselves very readily to an existence in the stomach. Consequently there exists in this organ, even in animals such as the dog, whose gastric juice contains most hydrochloric acid, a special flora, whose most characteristic feature is the relative insensibility to the acidity of this medium. The Blastomycetes, along with the yeasts and the Torulae, constitute the most frequent representatives of this flora; alongside these may be grouped the Sarcinae and certain acidophile bacilli. Miller^[689] has isolated several of these micro-organisms from the contents of the stomach, and has observed that, mixed with the food, they resist the action of the gastric juice, even that of the dog, whose hydrochloric acid content is greater than in man and many of the other mammals^[690]. But these acidophile micro-organisms have no pathogenic power and consequently are not much to be feared. It is very doubtful whether even the infective bacteria which are easily killed by the gastric juice in vitro, are often destroyed in the stomach. The typhoid coccobacillus, which has shown itself to be so sensitive to the destructive action of the gastric juice of man, of the dog, and of the sheep, is, from the experiments of Straus and Wurz, quite capable of passing through the stomach without being affected. Stern^[691], as the result of his own researches, as well as of those of his pupils, came to the conclusion that this micro-organism is not in the least affected by the gastric juice of a healthy man, containing the normal amount of hydrochloric acid. It was only in cases of hypersecretion and of hyperacidity that the micro-organisms of typhoid fever were destroyed before they reached the small intestine.

The cholera vibrio also can pass through the stomach and its acid juice. After Koch’s demonstration of the great susceptibility of this organism to acids in vitro, it was generally concluded that it must perish in the normal content of the stomach. Many cases have since been recorded in which the cholera vibrio was found, in times of cholera epidemics, in the faeces of healthy persons. In order to get into the large intestine it had to pass through the normal stomach. In experimental cholera in young suckling rabbits, a large number of vibrios were also found in the distinctly acid contents of the stomach, and they were seen to pass into the small intestine without any neutralisation of the acidity of the stomach taking place. This example proves, once again, that the phenomena that occur within the living body cannot be identified with those that go on in the test-tube, in vitro.

Whilst the acidity of the gastric juice exerts a certain influence on micro-organisms, the pepsin which it contains acts unfavourably on their toxins. There are many poisons which are readily absorbed, without being modified, by the mucous membrane of the stomach. Even the venom of snakes can, under certain conditions, produce its toxic effect as it is absorbed through the stomach. Thus, according to the experiments of Wehrmann^[692], pepsin exerts a very feeble action on this poison. On the other hand, this diastase has a marked action on certain bacterial toxins. Gamaleia^[693] pointed out that pepsin destroys the diphtheria toxin. Charrin and Lefèvre^[694] have shown that it also weakens other microbial toxins. According to Nencki and Mmes Sieber and Schoumow-Simanowski^[695], the gastric juice of the dog destroys relatively small quantities of the diphtheria poison. A gramme of the juice is capable of rendering innocuous 50 lethal doses of this toxin, but, in order that this action may be produced, a prolonged contact of the two substances is required. Since the neutralised gastric juice acts in the same way, this effect must be attributed not to the acidity of the gastric juice, but rather to the amount of pepsin it contains. This diastase acts much more powerfully on the tetanus toxin, 1 gramme of gastric juice neutralising 10,000 doses lethal for the guinea-pig. On the other hand, abrin is not modified by the gastric juice according to the researches of Répin^[696], carried out in Roux’s laboratory. Nevertheless, its action when administered by the stomach is feeble, and Ehrlich^[697] has been enabled to vaccinate small animals against this vegetable poison by availing himself of his knowledge of this fact. Répin explains this result as due to the very slight absorption of abrin by the gastrointestinal mucous membrane. This same factor, Répin thinks, may contribute also to the failure of various toxins when ingested. This rule, however, is not an absolute one. Thus, the toxin of the botulinic bacillus of van Ermengem^[698] is not destroyed by the digestive diastases, and it is certainly absorbed by the mucous membrane of the alimentary canal. For this reason, when it is introduced by way of the stomach, it exhibits a very violent toxic activity.