314 PATHOGENIC BACTERIA. 



themselves or remain in small groups, the result is quite 

 different, and much more closely resembles that of the 

 action of the living organism. 



Baumgarten, whose researches were made upon minute 

 tubercles of the iris, has shown that the first manifesta- 

 tion of the irritation caused by the bacillus is not the 

 attraction of leucocytes, but the stimulation of the fixed 

 connective-tissue cells of the part affected. These cells 

 increase in number by karyokinesis, and form about the 

 irritating bacterium a minute cellular collection which 

 forms the primitive tubercle. 



The leucocytes are of secondary advent, and are no 

 doubt attracted both by the substance shown by Prudden 

 and Hodenphyl to exist in the bodies of the dead bacilli 

 and by the necrotic changes which already affect the 

 primary cells. For reasons not understood, the amount 

 of chemotaxis varies greatly in different cases. Some- 

 times the tubercles will be sufficiently purulent in type 

 to justify the name "tuberculous abscess;" sometimes 

 there will be a marked absence of leucocytes. 



The important toxic substance produced by the bacillus 

 does not cause the chemotaxis, for when the leucocytes are 

 absent from the tubercles the coagulation-necrosis which 

 is so characteristic persists. 



The group of epithelioid cells and leucocytes constitut- 

 ing the primitive tubercle scarcely reaches visible pro- 

 portions before coagulation-necrosis begins. The proto- 

 plasm of the affected epithelioid cells takes on a hyaline 

 character, and seems abnormally viscid, so that contigu- 

 ous cells have a tendency to unite. The chromatin 

 of their nuclei becomes dissolved in the nuclear juice 

 and gives stained nuclei a pale but homogeneous 

 appearance. Sometimes this nuclear change is only 

 observed very late. As the necrosis advances some 

 of the cells flow together and form large protoplasmic 

 masses — giant-cells — which contain as many nuclei as 

 there were component cells. It may be that these nuclei 

 multiply by karyokinesis after the protoplasmic coa- 



