126 



48 



risation. Obviously, species which grow slowly in milk will also decolorise slowly. Sc. 

 pyogenes and most of the betacocci do not decolorise at all. Both spherical and rod-shaped 

 lactic acid bacteria can as a rule decolorise litmus milks more rapidly than they coagulate 

 it, but there are also species — Sc. thermophilus, for instance — wMch do not decolorise 

 the milk until long after it has coagulated. As the power of reduction is in the highest 

 degree dependent upon the vitality of the bacteria at the moment, it is of very little value 

 as a species character. 



As has been shown by the present writer^) and by Beijerinck^), the true lactic- 

 acid bacteria, in contrast to most other bacteria, are totally lacking in 

 catalase, and the fact that broth cultures or surface colonies of the lactic acid bacteria 

 do not develop oxygen with peroxide of hydrogen thus furnishes the principal test reac- 

 tion for these bacteria. The tetracocci, however, are exceptions, which as a matter of 

 fact do split up hydrogen peroxide to a very marked degree; the same applies to the 

 microbacteria, which also in other respects behave somewhat differently. 



Of the lactic acid bacteria, only some of those which do split up 

 hydrogen peroxide are capable of reducing nitrate to nitrite, but when 

 dealing vnth such, this feature should always be tested. The easiest method is to cultivate 

 the bacteria in dextrose broth with 2% KXO^; if nitrite is formed, then a small ladleful 

 will form a dark blue spot in a mixture of zinc-iodine starch and sulphuric acid. 



With the knowledge obtained, through the investigations here described, about the 

 lactic acid bacteria, their further determination will not as a rule occasion any serious 

 difficulty. 



We always commenced, of course, with a microscopic examination of the isolated 

 acid formers, in order to ascertain whether they were Gram positive, and whether they 

 were spherical or rod-shaped, but this is as far as it is possible to get by morphological 

 investigation in the first instance. Not till we have learned the biological qualities of the 

 strains, and grouped them accordingly, can we begin to consider the question of whether 

 certain related strains may have certain morphological qualities in common, and this 

 will also frequently be found — though by no means always — to be the case. The morpho- 

 logical differences between the different species of spherical forms, or between those of 

 the rod forms, or even, indeed, between their various genera, are so slight that they cannot 

 be determined until we know what strains are related together, as only the total impres- 

 sion of a great number of related strains can give a general idea of any value. 



After the Gram test, the next thing to try is the reaction with peroxide of hydrogen, 

 whereby the tetracocci and most of the microbacterium are separated off. The attitude 

 of the strains towards different temperatures is then noted, as also towards different 

 sugars, and different sources of nitrogen. The acids formed, and also any splitting up of 

 casein, are likewise further studied. 



') Del kgl. danske Videnskabernes Selskabs Forhandlinger 190(i. Nr. .'i. In this work it is proved 

 that also the butyric acid bacteria lack catalase. 



Archives Néerlandaise des Sciences Exactes et Naturelles 1907. Série li. Tome XIII, p. 357 



