H^MATOXYLIN AND SAFRAN1N. 159 



nilin (Liar, Zeit. f. iviss. Mik., ii, 1885, p. 149). See "Mucus-cells," in 

 Part II. 



256. Hsematoxylin and Safranin. This celebrated combina- 

 tion, which was used to such good effect by RABL in his 

 classical researches on nuclei (Morph. Jalirb., x, 1884, p. 215), 

 does not strictly belong to this subdivision, the safranin being 

 the primary stain, though used after the hasmatoxylin. You 

 stain very lightly with haematoxylin, so lightly that the stain 

 would not be of any use by itself (Rabl uses very dilute Dela- 

 field's solution, for twenty-four hours) ; wash out first with 

 water, and then with alcohol acidulated with HC1, then stain 

 for some hours in (Pfitzner's) safranin, and wash out with pure 

 alcohol. Eabl certainly was not far wrong when he wrote, 

 " This method is unequalled by any other." For richness of 

 detail in both nucleus and cytoplasm and tissue this method 

 has indeed hardly been equalled. 



Foi (Festschr. R. Virchow gewidmet, &c., i, 1891, p. 481 ; 

 Zeit. f. wiss. Mik., ix, 2, 1892, p. 227) prefers using the 

 haematoxylin and safranin combined in one solution. You 

 take 



Aq. dest. ..... ca. 100 



Bohmer's haematoxylin ... .25 



Safranin, usual 1 per cent, water and 

 alcohol solution . .... 20 



Stain sections in a few drops of the mixture for from one 

 to three minutes, wash in water, dehydrate, and mount; or 

 before dehydrating treat with a weak alcoholic solution of picric 

 acid, or, for some cases, of orange. This mixture was 

 specially used for staining marrow in an investigation into 

 the development of blood. 



257. Hsematoxylin and Metallic Stains. The conditions 

 under which haematoxylin can usefully be employed for 

 staining impregnated tissues have been discussed under the 

 heads of Gold and Silver. It only remains here to remind 

 the reader that haematoxylin works very well after osmic 

 acid. 



Other Combinations. 



258. The Anilin Double Stains. These very important com- 

 binations are so numerous that only a small proportion of 

 them can be mentioned here. 



' '^ x 



