FITZGERALD AND FRASER 813 



sufYering from diseases other than typhoid fever failed to agglutinate typhoid bacilli. A few 

 months later, further communications by Griinbaum and Durham emphasized additional 

 limitations of the Widal test. However, this method when correlated with clinical findings 

 in doubtful cases of typhoid was a most valuable addition to the physician's diagnostic 

 equipment at a time when typhoid fever was much more widely prevalent and when vaccina- 

 tion against the disease had scarcely been undertaken. 



Pfaundler,' in the year 1898, described his so-called "thread" reaction which is now of 

 historic interest only since it is essentially an agglutination reaction. Bordet^ was the first 

 to show that bacteria play a passive role when they are agglutinated. Vibrios killed by 

 chloroform were shown by Bordet to be still agglutinable. Bacteria killed by heat or formalin 

 behave in a similar fashion, as pointed out by Widal and Sicard, Van de Velde, and others. 

 Very high temperatures do, however, render bacteria less agglutinable. This may be due to 

 physico-chemical changes in bacterial protoplasm. 



It was quickly established that agglutinins usually, but not invariably, appear when 

 animals are immunized. The immunizing agents may be bacteria (vegetative or spore forms), 

 fungi, yeasts, protozoan parasites such as trypanosomes, and the parasites of the sporotri- 

 choses. Considerable variation in capacity to stimu'ate the appearance of agglutinins and, 

 in agglutinability, was early shown to exist among the bacterial species. Generally speaking, 

 motile species were found to be more readily agglutinated than non-motile. While various 

 species of vibrios and members of the typhoid-dysentery-colon group are most readily ag- 

 glutinable, B. pyocyaneiis, B. pestis, B. tetani, B. pertussis, B. mallei, are also, as a rule, quite 

 agglutinable. On the contrary, such species as B. anthracis, C. diphtheriae, pneumococci, 

 streptococci, and various other species of cocci agglutinate less readily. Various types of 

 encapsulated bacilli, such as Friedl'inder's bacillus and Bacillus mucosus capsulatus, are fre- 

 quently inagglutinable or but very slightly agglutinable. 



AGGLUTINOGENS AND AGGLUTININS 



Different strains or races of the same species of bacteria may show great varia- 

 tions in respect to agglutinability when acted upon by the same immune serum. 

 Various explanations of this, usually speculative in character, have been offered. 

 Bacteria when freshly isolated from the human or animal body are often inaggluti- 

 nable, but after cultivation upon artificial media may become agglutinable. This very 

 curious fact is difficult to understand. Bacteria which have been killed are not only 

 capable of being agglutinated, as has been pointed out, but are also antigenic. They 

 will, upon injection into animals, stimulate the appearance of agglutinins, as Widal 

 and Sicard and Levy and Bruns have shown. The former also established the fact 

 that culture filtrates, extracts of bacteria, or fluids in which bacteria have undergone 

 autolysis are antigenic. Arloing showed that tuberculin when injected into animals 

 would produce agglutinins for tubercle bacilli — and mallein, agglutinins for B. mallei. 

 This latter observation was made by Bonome. The antigenic element concerned in 

 the production of agglutinins is sometimes described as agglutinogen. It may or may 

 not be the same antigenic element as that which stimulates the appearance of pre- 

 cipitins. After the discovery of agglutinins for bacteria and other micro-organisms, 

 Bordet in 1898 in a study of hemolytic sera showed that these contained in addition 

 to the cellucidal substance a very active hemagglutinin. Furthermore, it was demon- 



'Pfaundler, M.: Centralbl.f. BaklerioL, 23, 131. 1898. 



* Bordet, J.: op. cit., 10, 193. i8g6. 



