638 The American Naturalist. [August, 
plainly expressed might millitate against the value of his work. 
Either horn of the dilemma is equally bad. Some, however, 
who are desirous of doing good work in this field, or at least 
appear to be conscientious workers in other lines, do not seem to 
be aware of the necessity for extreme care in the preparation of 
culture media and the management of cultures. Asa matter of 
fact, many bacteria are extremely sensitive to slight changes in 
the composition of the media in which they are grown or to 
other conditions within the control of the experimenter, and this 
appears to be true especially of the pathogenic species. Hence 
the many conflicting statements about the same organism. A 
few examples will render my meaning plainer. The careless 
exposure of cultures to bright sunshine may destroy the organ- 
ism. An organism supposed to come from diseased tissues or 
from a culture, and which is being examined in a cover glass 
preparation, may have been derived actually from a conta- 
minated staining fluid. The apparently simple matter of 
slightly unclean test tubes or flasks may lead to error, e. g. 
antiseptic influences may be at work, or preciptates may be 
thrown down and subsequently mistaken for bacterial growth. 
Some kinds of glass are unsuited to delicate bacteriological 
work, the culture fluids being contaminated by substances dis- 
solved out of the walls of the beakers, tubes, and flasks. Tyros, 
of course, are liable to mistake almost anything for bacteria or 
to find them anywhere (See a long paper by Bernheim on (12) 
Die parasitiiren Bacterien der Cerealen, in Miinch. med. Wochen- 
schrift, 1888, pp. 743-745 and 767-770, and comments on the 
same by Buchner and Lehmann, Tbid., 1888, p. 906, and 1889, 
p. 110). Boiling culture media, after it has been compounded, 
in open beakers or in loosely plugged test tubes or flasks may 
unwittingly lead to its concentration. The use at different 
times of different peptones, or grades of gelatine, of unlike per 
cents of gelatine or agar, of varying grades of acidity or alkal- 
inity, of impure chemicals, of different concentrations of the 
nutrient media, and of different methods in its preparation all 
tend to render comparative studies impossible. A large source 
of error in the differentiation of species by means of sugar fer- 
mentation experiments has been the employment of bouillon 
