From the above statistics it is easily seen that nursing, especie" primiparte, is an important factor in the causation of mastitis; further more, that affections of the nipples, which are also especially frequent in primipane, very frequently precede inflammation of the gland. Modern pathology admits that simple congestion, particularly pure passive hyper emia, is only exceptionally the direct cause of inflammation; it demands that there should be a particular irritant for the real inflammatory hyper emia, which plays such an important part in the tissue changes during inflammation. In all probability the same irritant is transmitted through the nipple to the interior of the gland, and consists of bodies which enter the lymphatics through the excoriated spot on the nipple and thence into the tissue, or which gain access to the latter through certain changes within the milk ducts, and from these to the deeper portions of the gland. The second view is less probable, because it can hardly be shown how bodies without active movements (such movements are not to be attributed to bacterium spores) can overcome the force of the current of the secre tion. The view that bacteria, which so change the milk that it acts as an irritant to the surrounding tissue, need only to enter the ducts of the nipple to act from thence as a ferment on the milk, is certainly not proven, although it cannot be dismissed. As regards the parotitis of typhus fever (and the same is true of mumps) Virchow has asserted that the inflamma tory irritant enters the gland from the mouth; this is always the most plausible hypothesis in regard to the causation of these diseases. The anatomical studies of puerperal mastitis, which I here offer, seem to sub stantiate this view, although the appearances can be differently interpreted.
.Anatnmy.—Ilitherto no one has had the opportunity of observing histologically the first onset of a puerperal mastitis; but from clinical ob servations of this process it is known that the gland is never affected at once in tot() (as usually appears to be the case in parotitis), but that in flammatory foci are formed in the gland, which may remain separate, but which as a rule gradually coalesce, and suppurate together. These foci may be considered partly as non-escaping milk, surrounded by inflamed tissue, partly as suppurating inflammatory foci, situated in the connec tive tissue between the acini. Clear histological descriptions of this process have not, so far as I know, been given either before or since my work on this subject. Most frequently the observations of Kolb are cited. He says: "The affected parts seem hard, and usually form nodular tumors, section of which shows them to be distended with milk; the glandular tissue is hyperemic and very succulent. In the acini, small extravasations of blood, the size of a pin's head, may be seen. As a rule, suppuration occurs early, and appears to me, indeed, to be a connective tissue suppuration; at least I could not discover anything in such cases which would indicate an epithelial suppuration. Pus appears at first in the acini, partly fluid, partly not, and as it seems to me most frequently with fibrous intercellular substance, so that we find in the grouped acini heaps of yellowish, fibrinous plugs, analogous to thoSe found in croupous pneumonia. Destruction soon overtakes the finer inter-acinous tissue, the small purulent foci coalesce, forming larger ones, the pus becomes fluid and the true mammary abscess is formed. The cavity of this abscess never has a smooth wall, but the membrane is rough, and not infre quently nodular, and ragged particles of broken-down gland tissue are found projecting from it." This description agrees exactly with what I have seen in such abscesses during life, and in the examination of persons who died of puerperal fever while having mastitis. I can, however, add something more accurate as to the histological processes in such cases. It may be seen in Fig. 9
that the acini are all more or less surrounded and in part completely covered by a small-cell (inflammatory) infiltration; but the epithelium seems to take absolutely no part in the cell-formation. It is difficult to see, with greater enlargement even, what becomes of it, though finally it seems to break down with the walls of the gland lobules, and the infil trated interstitial substance. At a, in the lower part of the figure, the pus, with the tissue detritus at the centre of the abscess, has disappeared from the preparation. It at first seems strange that here, in com parison with Fig. 6, the acini are very small and milkless, and that in the fresh state the soft succulent interstitial tissue is so enormously developed. I cannot say how long after confinement the woman, from whom this preparation was taken, died, or how long she had nursed her child; still it is certain that it was more than a week, or the preparation would have appeared very different. Obviously the secretion of milk had ceased for some time, as the gland is already in a state of puerperal in volution. It is very noticeable that only the immediate surroundings of the gland-lobules are infiltrated and that the very abundant interstitial tissue is not infiltrated. Finally, perhaps, another section of the gland would have shown gland-lobules completely normal.
From this and many other similar representations, it is seen that the inflammatory irritant must proceed from the acini themselves, or from their immediate surroundings. The first view seems at first sight to be the most plausible, and agrees also with what has already been said. But, anatomically, it is not the only possible one, as may be seen from the following experiment. I produced suppuration by drawing through the moderately developed mamma of a non-nursing bitch, a white cotton string about .39 inches wide, and killed the animal a few days later. the inflammatory infiltration had not extended far around the string; it was not diffusely distributed in the inter-lobular connective tissue, as in a phlegmon, but surrounded the gland-lobules as in puerperal mastitis. This shows that there are channels of dissemination for inflammatory processes, which follow the ramifications of the gland; these can only bo the blood and lymph-vessels, which surround the ducts and lobules. I might maintain that it is most probable that the irritative material is dis tributed with the lymph through the gland, and thence acts upon the capillary net-work around the lobules, that the leucocytes emigrate and produce the purulent infiltration of the tissue immediately surrounding the lobules and their acini. As to whether the pus cells are formed more from themselves or from the connective tissue cells of the interstitial tissue, I cannot say. At all events, capillary stasis, thrombosis and necrosis of the tissue result, so far as this is not already the subject of cell infiltration. As regards the " fibrinous purulent" plugs, which Klob found in the small abscesses of purulent lobules, I would prefer to look upon these as necrotic glandular and connective tissue. One may easily observe this change in suppuration of the subcutaneous tissue. I have not been able to convince myself of the formation of fibrin in mastitis;at least in the cases I have carefully examined, and would prefer therefore not to accept too readily the comparison with the finer processes of croupous pneumonia, which I have carefully studied. Further researches are necessary to prove that the process described by me in the tissue in puerperal mastitis is constant. These researches are now of new in terest, as there is such a growing inclination to attribute all such infec tions suppurations to the growth of micrococci.