56 MORPHOLOGIC VARIATION 



Various other staining methods which will differentiate living 

 from dead cells have been described. As is well known, Gram-posi- 

 tive bacteria become Gram-negative in old cultures, and R. and W. 

 Albert have shown that the loss of the Gram-staining property in 

 yeasts is correlated with the degree of autolysis of the dead cells. 

 Fraser studied the action of various dyes on living and dead yeast 

 cells, and found that these could be differentiated readily by methy- 

 lene blue; his procedure was to suspend the unfixed cells in the dye 

 solution. This method was used by Fulmer and Buchanan to meas- 

 ure the death rate of yeasts acted upon by disinfectants; they 

 pointed out that the ability of the cells to absorb stains occurs 

 ajter the cells have lost the ability to grow when transferred to a 

 new medium. Burke has shown that dead spores may be differen- 

 tiated from living ones by staining with carbol fuchsin, and Koser 

 and Mills also found that the ability of the spores to absorb the 

 dye was acquired later than the period when they lost the power 

 to germinate. Seiffert found that when suspended in Congo red dead 

 cells of bacteria absorbed the dye, living ones not. But the differ- 

 entiation by this technique is not so sharp as with the procedure I 

 have used. 



It has generally been assumed that the staining of the dead cells 

 is due to increased permeability of the membrane to the dyestuff, 

 and this view would seem to be supported by the observation of 

 Fraser that dead yeast cells stain much more slowly with the less 

 diffusible Congo red than with methylene blue. With Fraser's tech- 

 nique, however, it is found that with neutral red the dead yeast cells 

 become stained a uniform deep red, and that in the living cells the 

 volutin granules and vacuoles are also stained. The cell wall of the 

 living cells must therefore be permeable to the stain, and the failure 

 of the protoplasm to become stained must be due to a difference 

 in its adsorption properties from those of the dead cells. As I have 

 pointed out above, granules are also stainable in the living cells 

 of bacteria when Congo red is used, so the membrane must also be 

 permeable to this dyestuff. 



The fact that dead, partially autolysed cells may be detected by 

 the use of Congo red gives us a valuable procedure for measuring the 

 rate of death and autolysis of bacteria which has been used in con- 



