270 DYEING 



by Roe, Lillie, and Wilcox. ^^^ Each of these was tried on blood- 

 smears in conjunction with eosin ; so also were toluidine blue and 

 thionine. The Romanowsky effect was given by all the azures, but 

 best of all by azure B ; it was given faintly by toluidine blue but not 

 at all by thionine. Azure B was also found to be superior to A in 

 giving the Romanow^sky effect in the presence of methylene blue 

 and eosin; it should therefore be substituted for the azure I of 

 Giemsa's formula. It will be remembered that azure I is in fact 

 azure B, more or less adulterated. The American authors found no 

 conclusive evidence that any benefit could be obtained by adding 

 azure A as well as azure B to the mixture. 



A different line from Giemsa's was followed by MacNeal, 

 though both were animated by the desire to place Romanow^sky 

 dyeing on a scientific basis. MacNeal placed the emphasis on 

 methylene violet. This substance is insoluble in water unless other 

 thiazine dyes are present, but soluble in alcohols and to some ex- 

 tent in mixtures of alcohols and water. MacNeal ^^^ found that by 

 itself or in the presence of eosin it would not dye chromatin at 

 all, but in the presence of methylene blue it show^ed its meta- 

 chromatic property by giving the usual Romanowsky purple to the 

 chromatin of leucocytes. Strangely enough, it did not colour the 

 chromatin of the malarial parasite. 



MacNeal did not deny that the Romanowsky effect could be 

 produced by the azures, but thought that the metachromatic 

 colouring of the chromatin of leucocytes was best achieved by 

 methylene violet in the presence of methylene blue. He thought 

 that the ordinary method of polychroming methylene blue to pro- 

 duce the Romanowsky effect resulted in a preponderance of 

 methylene violet over azure, and that the former was chiefly 

 responsible for the metachromatic dyeing of chromatin. 



MacNeal used azure A to give a reddish colour to the nucleus of 

 parasitic Protozoa, for this is not given by methylene violet. His 

 excellent tetrachrome mixture ^^^ therefore consists of methylene 

 blue, methylene azure A, methylene violet, and eosin, dissolved 

 in methanol. It is used exactly like Leishman's. One might 

 suppose that it could be improved by the substitution of azure B 

 for azure A, but for some unexplained reason this is not so; 

 on the contrary, the change results in a markedly inferior dye.^^ 



Methylene blue itself, if it could be obtained in a perfectly pure 

 state, would probably be orthochromatic (p. 251). The ordinary 



