30 STAINING REACTIONS OF BACTERIA 



the latter, for example, to certain coal-tar dyes completely reverses their gram reaction 

 although similar treatment is without effect on the gram reaction of the former. More- 

 over, thorough modifications of the gram technique in which time of exposure to dye and 

 mordant is greatly shortened and time of exposure to decolorizer greatly prolonged de- 

 colorize B. anthracis in large part but are without effect on M.freudenreichi (see Plate 

 I, (c), facing p. 36). In specimens of B. anthracis stained by such a modification, beauti- 

 ful partial decolorizations occur in which a gram negative central rod is to be seen shin- 

 ing through among remnants of undissolved gram positive material which cling to the 

 surface of the bacteria in the form of lumps, plaques, or dots (see Plate I). I have 

 also called attention to the fact that the gram negative forms of B. anthracis were 

 much smaller in diameter than the gram positive forms. Filar micrometer measure- 

 ments of about three hundred organisms showed a reduction in diameter of 0.5145 fx 

 or 42 per cent. Evidence was adduced to indicate that the difference between the 

 gram stability of an organism like B. anthracis and that of an organism like M.freu- 

 denreichi depends on the fact that the latter is composed of an easily destructible gram 

 positive cortex and a gram negative medulla, an anatomical structure which is not 

 possessed by an organism like M. freiidenreichi. 



A critical review of all the known facts leads Smith' to sum up our knowledge of 

 the mechanism of the gram stain as follows : Gram positivity is due to some property 

 of the bacterial cell which permits the ingress of the water-soluble dye but does not 

 permit the egress of the alcohol-soluble iodine-dye compound. A gram negative cell 

 is not readily penetrated by the water-soluble dye but is easily penetrated by the 

 alcohol-soluble iodine-dye complex. In both cases permeabihty is conditioned by the 

 physical state of the components of the intact cell membrane. When gram positives 

 become gram negative, what is lost is a particular condition or physical distribution of 

 the constituents of the cell membrane. 



The exact technique which Gram himself advised is not now generally used. Im- 

 mediately after its introduction, modifications of the technique were suggested, identi- 

 cal in general principle with the original method though differing in details. With 

 these every bacteriologist is familiar. In a recent study of this subject involving the 

 examination of nearly fifteen thousand smears, I tried out all the better-known modi- 

 fications and came to the definite conclusion that Burke's modification' is in every 

 way superior to all others. In this method the stain (methyl violet) is alkalinized on 

 the slide by the addition of sodium carbonate, the mordant is used in strong solution, 

 and acetone-ether, a powerful decolorizer, is employed. The colors in the final result 

 contrast sharply — the pink of the counter-stain (safranine O) with the bluish black of 

 the gram positive forms — and one is never in doubt in which group to place a given 

 organism. The results are constant and clear cut; that is to say, they are constant and 

 clear cut if the conditions of experiment which are known to modify gram behavior be 

 kept in mind and controlled. For it must be emphasized that the method of Gram is 

 an exact one, and irregular results are due, not to the defects of the method but to 

 failure to keep all the conditions of the examination constant. For this reason, since 

 all of the factors are now known, the great variability in results which were obtained 

 in the early work — before these factors were understood — has largely disappeared. 

 ' Smith, H.: loc. cit. » Burke, V.: /. Bad., 7, 159. 1922. 



