272 DYEING 



for the Romanowsky effect, for the dyes act in almost exactly the 

 same way if dissolved together in water without elimination of the 

 inorganic ions ; and anyhow chloride ions are present in the usual 

 mixtures, since the basic dyes are present in excess. 



The evidence suggests, then, that the formation of a neutral dye 

 does not in itself confer important special properties; yet it is a fact, 

 established more than half a century ago by Giemsa,^^^' ^^'^ that 

 azure chloride makes the nucleus of the malarial parasite only 

 feebly violet, while azure/eosin makes it the desired brilliant red. 

 Similarly, polychrome methylene blue alone does not give the 

 Romanowsky effect with the nuclei of parasitic Protozoa; but if 

 the smear be treated previously with eosin, the characteristic red 

 colouring results. ^^^ Since eosin alone will not dye the nuclei of 

 these organisms, there is obviously a problem to be solved. 



The dyeing of the chromatin of leucocytes is not the same 

 process as the dyeing of the nucleus (or part of the nucleus) of 

 Protozoa. It is to be noticed first of all that the colour is different. 

 The nuclei of Protozoa are dyed a much purer red than the 

 chromatin of leucocytes, which is usually coloured purple. 

 Further, the chromatin of leucocytes is similarly coloured whether 

 eosin be present or not, while Romanowsky dyeing of the proto- 

 zoan nucleus is dependent on the use of eosin. When the granules 

 of Mastzellen are coloured purple by a Romanowsky dye, we are 

 concerned simply with what Pappenheim ^^^ meant to call the 

 substantive metachromatic basiphilia of acidic objects (though he 

 accidentally said 'subjective' when he meant substantive). The 

 same applies to the chromatin of leucocytes. Pappenheim regarded 

 the red coloration of the nuclei of Protozoa as an example of 

 adjective metachromatic neutrophilia: that is to say, he considered 

 that the nuclei of Protozoa were 'neutrophil' in his sense and that 

 the eosin played the part of a mordant in the process of dyeing. 



The idea that eosin might act as an acidic mordant for a basic 

 dye (unlike the usual basic mordants for acid dyes) seems to have 

 suggested itself to Nocht ^^^ long before. Nocht had reason to 

 suppose that resorcinol might replace eosin in Romanowsky dye- 

 ing. This idea was further developed by Unna,^^^ who found that 

 potassium tribromophenolate, a colourless substance related to 

 one part of the eosin molecule, could replace eosin in Romanowsky 

 dyeing. The nuclei of Protozoa w^ere dyed in a colour closely 

 resembling that given when eosin was used. (Both Unna ^^^ and 

 Giemsa ^^^ considered that the colour of eosin itself slightly in- 



