ON STAINING WITH GOAL-TAB COLOURS. 201 



roethylen blue may be made by dissolving 1 grm. of methylen blue in 100 

 c.c. of the formalin solution. Sections are said to stain in half a minute, 

 and to resist alcohol muc*h more than is the case with those treated by the 

 usual solutions. The formalin solution of safranin (Safranin 0, from 

 Grubler) is said to give a plasma stain behaving in all particulars like eosin. 



271. Choice of a Stain. One might think that it would be 

 quite sufficient for all practical purposes to possess one good 

 red stain and one good blue one, so that, for instance, 

 safranin and gentian violet should be sufficient for the most 

 exacting of laboratories. But I think that for delicate work, 

 at any rate, it is desirable to possess one or two more. We 

 have to take account of the manner in which these colours 

 behave when used in combination with the plasma stains that 

 it may be desired to employ. And there is another point 

 that is not undeserving of attention. Some of the dyes dis- 

 cussed in the following chapter give a stain of a somewhat 

 dead or dull quality, so much so that chromosomes and 

 nucleoli frequently come out quite opaque. Gentian violet is 

 in this case ; whilst dahlia, which is otherwise near to it in 

 hue, is not. Safranin and methyl green, on the other hand, 

 leave the structures beautifully transparent. This is an ad- 

 vantage with thick sections, and sometimes for other reasons ; 

 but this transparency of the elements is unfortunately 

 favourable to the production of diffraction lines, which may 

 be a hindrance to good definition in delicate work. So that 

 the dead colours, such as gentian, have a certain advantage 

 for work with very thin sections and where very fine defini- 

 tion of chromatiii is required; whilst the transparent or 

 semi-transparent colours, such as safranin, should be pre- 

 ferred for thick sections. I would also add that it always 

 seems to me that the blue stains, such as gentian, are less 

 favourable for work with artificial light. They give more 

 or less dichroic images, which are not favourable to good 

 definition. 



To sum up, I would recommend safranin for a red chro- 

 matin stain, and gentian for a blue one, except where 

 special conditions suggest another choice. 



