676 G ASTRO-INTESTINAL BACTERIOLOGY 
overgrowth of the gas bacillus even of several years' duration do not 
exhibit signs or symptoms of toxemia in spite of the protracted illness.^ 
It is unfortunate that practically none of the bacteria which incite 
intestinal disturbances or illness produce soluble toxins against which 
antitoxins can be prepared; sera likewise have been unsatisfactory. 
There is little, therefore, that can be accomplished serologically with 
present methods in the treatment of intestinal disturbances of bacterial 
causation. Attempts to permanently eliminate or destroy undesir- 
able bacteria with cathartics and intestinal antiseptics have not been 
productive of results in the past- and prolonged starvation^ per se 
does not lead to intestinal sterility or to a significant reduction in the 
offending bacteria. 
There are two ways, however, in which direct influence may be 
applied to bacteria in the intestinal tract: By a substitution of harm- 
less types of organisms for abnormal types, and by varying the diet 
of the host in such a manner that the intestinal contents at the desired 
level shall contain nutritive substances that may be reasonably ex- 
pected to shift the metabolism of the offending organism, and there- 
fore radically change the character of the products of its metabolism. 
A substitution of bacteria may be accomplished, theoretically at 
least, either by feeding cultures of organisms whose products of growth 
are harmless to the host and more or less inimical to the bacteria it 
is desirable to supplant, or by administering a diet which contains 
appropriate nutritive substances in sufficient amounts to create con- 
ditions favoring the development of normal intestinal bacteria whose 
activities are in opposition to those it is desired to restrict or supplant. 
The effects of a monotonous diet maintained for considerable periods 
of time upon the intestinal flora of a normal individual are clearly 
shown in the normal nursling, where intestinal organisms are largely 
carbohydrophilic and fermentative in character. Feeding experiments 
in normal animals indicate that the development of a nursling intes- 
tinal flora follows the prolonged administration of a nursling diet. 
If the intestinal flora to be modified does not contain sufficient 
numbers of the desired types of bacteria, or if these latter organisms 
are inactive, it may be important to reenforce the weakened or inactive 
residual types with suitable cultures from without. Herter'* was the 
first to recognize the possibility' of introducing desirable types of bac- 
teria into the alimentary canal, and Metchnikofi'^ has extended and 
popularized this form of bacteriotherapy through his extensive studies 
upon the eft'ects of milk soured with the Bulgarian bacillus as a thera- 
peutic measure in excessive intestinal putrefaction. The Bulgarian 
l)acillus" is a large Gram-positive organism, which is non-motile and 
1 Kendall: Jour. Am. Med. Assn., 1926, 86, 737. 
2 See Kendall (Jour. Med. Res., 1911, 25, 117) for brief resume. 
' Even after thirty-one days' starvation, a large number of viable bacteria were found 
in the lower part of the intestinal tract of the one case studied with this possibility in view. 
4 British Med. Jour., 1897, ii, 1898. 
'■> Prolongation of Life. 
6 See Rahe (Jour. Infec. Dis., 1914, 15, 141) for description and differentiation from 
pthgr ^ciduric bacteria. 
