AGGLUTININS, AGGLUTINOIDS AND PROAGGLUTINOIDS 147 
diphtheria toxin show a marked increase in globulin content, with a 
decrease in albumin content, and in general antibodies appear to be 
associated with the serum globulins. Beljaeff^ could find no appreci- 
able change in the refractive index, specific gravity, freezing-point or 
reaction of the serum of an immune animal above that of a normal 
animal. 
The chemical nature of protein antibodies, aside from their appar- 
ently close relation to globulin, has not been determined. There is 
evidence that antitoxin molecules may be larger than toxin molecules, 
however. jNIartin and Cherry- found that toxins could be forced 
through dense porcelain filters impregnated with gelatin, which would 
restrain antitoxin, and Arrhenius and Madsen^ determined that the 
toxin molecule diffused several times as rapidly as the antitoxin mole- 
cule, from which observation they assumed that the antitoxin molecule 
was larger than the toxin molecule. The antibody content of sera is 
gradually decreased by drying. Drying at 0° C. is less destructive of 
antibody content than desiccation at room temperature. The hydro- 
gen-ion concentration of sera increases during drying and it is probable 
that several physico-chemical changes play a part in the reduction 
of antibodies.'* 
AGGLUTININS. AGGLUTINOIDS AND PROAGGLUTINOIDS. 
Gruber and Durham^ appear to have been the first to clearly demon- 
strate specific clumping in broth cultures of typhoid and cholera 
organisms when their respective sera were added to them. Somewhat 
later Widal,*" and independently Griinbaum,^ utilized the principle 
of the specific agglomeration of bacteria by their immune sera for 
the diagnosis of typhoid fever. They found that relatively early in 
the disease, sera of typhoid patients clumped typhoid bacilli from 
broth cultures. Pfaundler^ observed that typhoid bacilli cultivated in 
broth containing low concentrations of specific sera grew out into 
long, tangled filaments, the so-called "thread" reaction. Originally 
this phenomenon was regarded as highly specific, but it has largely 
been supplanted in practice by the macroscopic or microscopic agglu- 
tination test. 
Agglutination in the bacterial sense may be defined as a cliunping 
or agglomeration of bacteria from a uniform suspension in a fluid 
medium, brought about by the addition of specific antibodies— 
agglutinins. It takes place in two stages if motile bacteria are con- 
1 Centralbl. f. Bakteriol., orig., 1903, 33, 293, 369. 
2 Proc. Roy. Soc, 1898, 63, 420. 
' Festskrift Statens Serum Institute, 1902. 
* Karsner and Collins: Jour. Infec. Dis., 1919, 25, 427. Karsner and Koeckert: 
Jour. Am. Med. Assn., 1919, 73, 1207. 
6 Miinch. med. Wchnschr., 1896, 43, 285. 
6 La Semaine Medicale, 1896, No. 51. 
' Lancet, 1896, ii, 806; Miinchen. med. Wchnschr., 1897, 44, 330. 
8 Centralbl. f. Bakteriol., 1898, 23, 9, 71, 131. 
