^ETIOLOGY OF TUBERCULOSIS. 
107 
but after a shorter or longer time makes its way into the broncliige, empties 
itself and so is changed into a cavity. The further increase of the cavity goes 
on in a very irregular manner according as the process of vegetation in the 
tuberculous bacilli makes a halt in single places for a shorter or longer time, 
or continues, and according to this indentations or shrivelling are formed in 
places. Taken in general, the cavity, however large or irregular formed, re¬ 
tains the essential properties of the tuberculous caseous herd ; necrotic masses 
in the interior, joined to these towards the outside nests of epithelioid cells 
with gigantic cells sandwiched in, and in the giant cells often tuberculous 
bacilli. An exception occurs only in so far as the tuberculous bacilli in the 
cavity appear in comparative abundance also in the interior of necrotic masses, 
which in the caseous herds remaining permanently closed is not usually the 
case. Probably this has its foundation in the fact that the masses dead, and 
to a certain extent used up as a breeding-ground for bacilli, are constantly 
being emptied, and the parting of the walls of the cavity give constantly a new 
food material for bacilli. 
In this manner the usual chronic form of phthisis would run its course. 
In this usual course the vegetation of the bacilli is a very slow one and the 
occurrence of the bacilli very sparse, and essentially confined to the giant cells 
in the immediate surrounding of the cavities and to the contents of the same. 
The circumstance is very noteworthy that even in comparatively small tuber¬ 
culous herds the growth and dispersion of the bacilli is not uniform but dis- 
continous. In large herds, and especially in larger cavities, this behavior, 
which has already been touched upon, is always more striking. Widespread 
spaces of the cavity may be wholly free from bacilli and sometimes the bacilli 
may be confined to single spots of very slight extent. From this we may con¬ 
clude that the conditions of life for bacilli in a tuberculous herd are not every¬ 
where equally favorable, and probably also in regard to time may be subject 
to fluctuations. The bacilli must then vanish from the places which no longer 
give them suitable breeding-ground. In this case at one time only a temporary 
freedom from the parasites can take place, when the bacilli from the neighbor¬ 
hood later force themselves in or if spores have been left behind which may 
develop under more favorable conditions. At another time a lasting freedom 
of the diseased spot from bacilli can take place when the just mentioned con¬ 
ditions for the reviving of the bacilli vegetation do not occur. Shrivelling, 
scarring and healing will follow then in such a place. One can think that 
since these things may take place partially in the periphery of the tuberculous 
herd, the same might happen in the whole compass of the herd, and so a com¬ 
plete healing take place. Analogous relations are found in other diseases condi¬ 
tioned upon bacteria which also spread themselves out centrifugally from the 
original spot of infection, but can show in their progress considerable irregulari¬ 
ties sometimes cease to grow at one point, sometimes thrive and spread rapidly, 
as is the case, for example, in erysipelas. 
The development of a single tuberculous herd running its course in the 
lung under the type of chronic phthisis can be complicated in many ways, if 
the tuberculous bacilli in any manner get out of the reach of the original 
herd into other places and there give rise to the development of secondary 
