ON STAINING WITH COAL-TAR COLOUKS. 195 



colours fail to correspond precisely to the technical categories 

 of chromatin stains and plasma stains. 



For instance, orange is an acid colour ; but used as a re- 

 gressive stain I find it will give a very sharp stain of chro- 

 nmtm : it cannot, therefore, be classed as a mere plasma 

 stain, though it is also a very good plasma stain. Saure- 

 fuchsiii is a very acid colour. It behaves in general as a 

 decided plasma stain. But used as a regressive stain it 

 sometimes, under conditions which I am not able to specify, 

 gives a very vigorous stain of chromatin. Safranin is a 

 basic colour, but by the use of appropriate mordants it can 

 be made to behave as a plasma stain. Methylen blue is a 

 basic colour. But, as is well known, when employed according 

 to the method worked out by Ehrlich for the so-called intra- 

 vitam staining of nerves, it affords a stain that is essentially 

 plasmatic, such staining of nuclei as may occur in this pro- 

 cess being an accidental epiphenomenon. Nigrosin is, ac- 

 cording to Ehrlich, an acid colour, and should therefore be 

 essentially a plasma stain. Yet I find that, used as a 

 regressive stain in the same way as safranin, it gives ;i 

 vigorous chromatin stain, cytoplasm being only faintly 

 coloured. Bordeaux is an acid colour, but it stains chromatin 

 as well as cytoplasm : and many similar cases might be men- 

 tioned. Indeed, it is not too much to assert that there is 

 hardly any colour, either basic or acid, that may not be 

 made to afford either a chromatiii stain or a plasma stain, 

 .according to the way in which it is employed. 



It would seem, therefore, that Ehrlich's generalisation 

 does not hold good as a statement of the behaviour of tar 

 colours when employed for staining sections in the usual way. 

 It is rouyldij true that the basic colours are in general 

 chromatin stains, and the acid colours in general plasma 

 stains ; but the rule is subject to many exceptions. 



263. Progressive and Regressive Coal-tar Stains. Very few 

 tar colours give a precise nuclear or chromatin stain by the 

 progressive, or direct method ( 199). Two of them methyl 

 green and Bismarck brown are pre-eminently chromatin 

 stains. Many of the others for instance, safranin, gentian, 

 and especially dahlia may be made to give a nuclear stain 

 with fresh tissues by combining them with acetic acid ; but 



