It should

no doubt, this is due to a process similar to that described as occurring in the mesenteric glands, where it is accompanied by no permanent lesion in the intestine; but in other cases it seems to be equally beyond question that the specific virus has passed (a) from the mesenteric and retro-peritoneal glands, and (b) from the peritoneal cavity through the central part of the diaphragm and the broad ligament of the lung, or (c) by a more or less circuitous route along the lymphatics of the parietal pleura to the root of the lung. It is worthy of note that until pleurisy is set up in these cases there is no transmission from pleura to pleura, but that as soon as the slightest adhesion takes place there may be continuation of the process on the two surfaces. be observed, however, that in addition to this affection of the costal pleura the visceral pleura may also be the seat of tubercular nodules, the virus in this case having come probably from the abdominal cavity by the diaphragm and the broad ligament. In all these forms there is abundant evidence of the transmission from point to point of the virus by way of the lymphatic channels, especially where the tissues generally are highly resistant, and where the epithelial cells, though they do not arrest the passage of the bacilli into the lymph spaces, have still sufficient vitality to continue to grow in a more or less regular manner. In such cases the connective tissue resistance is also great, and though the bacilli may still continue to grow and to attack the cells in their immediate neighbourhood, those cells outside the immediate sphere of action of the irritant are stimulated into proliferation and fibrous tissue formation, so that a fibroid capsule is formed around the cellular or caseating centre. When the epithelium itself is attacked caseation rapidly ensues, and absorption from this point may take place for some distance along the lymphatics. In consequence of this method of spreading by the lymphatics of the lung, nodules may be sought in those positions in which coal pigment is found to accumulate, the only difference observable being that in tubercle the nodules are usually somewhat limited in their area of distribution, the pigment, on the other hand, being disseminated over the whole lymphatic area. The changes around the vessels and in the bronchi are marked by no special features; the tubercles are formed in connection with the peri-bronchial and peri-vascular lymphatics, and in some cases, as has been

observed, they appear to be formed in the small lymphoid nodules which may be seen in the walls of the lymphatic vessels.

In the intima of the vessels, as Cornil and Ranvier, Hübner Greenfield, Hamilton, and others have insisted, the changes are extremely well marked, and quite recently attention has been called to the fact that even some of the systemic arteries may be deeply affected with arteritis obliterans in cases of chronic phthisis. How far the changes in the intima are associated with those in the adventitia is as yet not fully decided, but there seems every reason to hold with Arnold, that wherever the lymphatic circulation in the adventitia is disturbed-especially where there is irritation and proliferation of the endothelial cells of the lymph spaces-corresponding changes are met with in the intima, particularly where the process is chronic in character.

(354) The Pathology of Acute Miliary Tuberculosis.

Acute miliary tuberculosis must be looked upon as the result of spreading of the infective material directly by the blood channels. The demonstration of this fact was first accomplished by Weigert, who, in a series of several cases of acute miliary tuberculosis, was able to determine the presence of ulceration of the pulmonary vein. The process being similar to that in or near the wall of a bronchus in the cases mentioned, Ponfick had first supposed that the bacilli might pass from a tubercular thoracic duct into the venous trunks, and thus to the general circulation. It is probable that both observers were correct, and that both forms may occur. Coats further points out that a limited distribution of tubercle by the blood may be due to the passage of bacilli into the minute venous radicles in the glands in which tuberculous changes are occurring. That bacilli are found in the blood has been now frequently demonstrated, and quite recently several cases have been recorded, in which general tuberculosis has come on after hæmorrhages in patients suffering from apical phthisis. This is a matter of all the greater interest when it is borne in mind that all these cases of acute tuberculosis were developed in from seventeen to twenty-five days, just the period given by Koch as that required for the development of tuberculosis when produced experimentally. The importance of this can scarcely be over-estimated from a

surgical point of view, indicating as it does the methods of procedure to be adopted in operating on any tuberculous part. The bacilli, though found in the blood in such cases, do not become active until they come to some part of the circulation at which they can make their way into the surrounding tissues. In some cases bacilli are present in the emboli, or they may be actually distributed in the embolic area, in many cases appearing to make their way from the capillary vessels into the lymph spaces, and only then giving rise to the characteristic series of changes.

(355) The Relations of Scrofula, Lupus, and Tuberculosis.

The recent research of Dr. Alfred Lingard1 shows the precise relationship of these affections. They must all be classed as tuberculous. Scrofulous material injected into a guinea-pig produces a general tuberculosis; lupus material does the same thing, but slower. In fact, given tubercle, scrofula, and lupus, there is a regular gradation of virulence, as tested by subcutaneous inoculation into guinea-pigs; with tubercle, in six or seven days the glands enlarge; with scrofula, the enlargement of the glands just above the inoculated point is not observed for two or three weeks; and with lupus, 28 days is the earliest time within which this has been noticed. Tubercle kills a guinea-pig in an average period of 80 days; scrofula, 206 days; lupus, 331-5 days. It is pretty certain that scrofula and lupus may be considered as the expression of different stages of attenuation of the tubercle bacillus, but by successive removes the bacillus acquires its pristine energy; thus, according to Lingard's experiments, if a guinea-pig be inoculated with scrofulous material, and the period in which it lives. be reckoned as 100 days, products from this guinea-pig, injected into a second, kill in 63 days; inoculation from the second into a third guinea-pig kills in 38 days; and inoculation from the third into a fourth kills in 29 days.

(556) Propagation of Tubercle through the Agency of Milk derived from a Diseased Cow.

There have been numerous researches upon this important

1 Supplement, containing Report of Medical Officers, to Eighteenth Annual Report of the Local Government Board.

subject. Johme1 maintained that the milk of tuberculous animals should not be given to sucklings, and should not be consumed by any one when the teats were affected. The flesh of tuberculous animals may be eaten if the tuberculosis is not general, but the internal organs affected and the lymphatic glands are to be put on one side. In general tuberculosis the flesh also was not to be eaten. Lydtin 2 made experiments with the raw and boiled milk of tuberculous cows with and without udder affection on guineapigs, and obtained only once with raw milk a positive result, hence he thinks that the danger of infection through milk cannot be great, and through boiling can be quite avoided.

Martin has experimented with Paris milk, derived from thirteen different dairy farms, by injection into the peritoneum of guineapigs. In one case general tuberculosis supervened. Stein obtained a positive result in several instances by injecting the milk of tuberculous cows into the abdominal cavity of guinea-pigs, although in one only of his cases was the udder affected. The Commission 5 appointed in Victoria to inquire as to the extent of tuberculosis gave an opinion that the meat of animals strongly affected with tuberculosis should be forbidden, but in less severe cases could be consumed. On the other hand, tuberculous cows should be rigorously excluded from supplying milk for consumption.

Quittel declared the use of flesh actually tuberculous, not fully boiled or cooked, injurious, but the meat of a tuberculous animal only injurious if the animal has become thin, and more than one organ is affected.

Bang' found the milk in cases of tubercle of the udder always to contain the specific bacillus, and it constantly produced tuberculosis. The milk of tuberculous cows with healthy udders sometimes contained tubercle bacilli, and then could produce tuberculosis; sometimes it was free from bacilli, and then inoculations were fruitless. In bacilli holding milk submitted to the centrifugal apparatus, for the most part bacilli could be recognised in the

1 Deutsche Zeitschrift f. Thier. Medicin. Bd. IX. 2 Archiv. f. Thierheilkunde. 1884. 3 Revue de Médecin. 1884, No. 2. Experimentelle Beiträge sur Infectiositat perlsüchtiger Kühe. Dissertation. Berlin, 5 Report, 1884-5.

1885.

6 Deutsche Vierteljahr. f. offentl. Gesundheit. Bd. XIX. 7 Tidsk. f. Landakonomi.

1886.

scum at the periphery of the apparatus, the cream also contained single bacilli. Cream raised by standing was infectious, and the same must be said of the butter. Heating bacilli holding milk to 70 deg. C. destroyed mostly the infection, but not always. Boiling was constantly successful in destroying the infection.

Galtier concluded from his researches that not only the raw or coagulated milk from tuberculous cows could infect, but also the cheese or whey prepared from it. He also calls attention to the fact that fowls or pigs could be infected from feeding on such substances.

At the International Medical Congress in Copenhagen, Bang gave the results of an examination of no fewer than twenty-seven cases of tubercular mammitis, and he was able to demonstrate the presence of tubercle bacilli in the milk or in the sediment, and with this milk or sediment he was able to produce tuberculosis both by inoculation and by ingestion. Nocard was able to demonstrate the presence of the specific bacillus in milk in eleven

cases.

Dr. Woodhead has made a most careful and systematic examination of over 600 cows in the Edinburgh dairies with Professor MacFadyean. Of the whole, they found thirty-seven beasts in which there was mammitis, but only six, or 16 per cent., in the milk of which they could demonstrate the presence of tubercle bacilli, and then only in small numbers. In one of the six cases, and subsequently in five other cases, they made sure of the existence of the bacilli in enormous numbers in the udder by microscopic examination. They find that new tubercular tissue is disseminated in patches of various sizes throughout a portion of the gland, and that all the more minute elements of tubercle may be distinguished-the small round cells in which the nuclei are comparatively large, and the epithelioid cells, between or amongst which is a fairly well developed reticulum. The giant cells are very numerous, but are not so well defined as one sees them in the human subject; they are scattered throughout the new tissue. The tendency to caseation of tubercle in the udder is not nearly so well marked as in other parts of the body, but it does undoubtedly occur at points. The new growth of tuberculous tissue gradually invades the lobules of the gland, passing in along the lines of the 1 Compt. Rend. T. CIV.