462 Journal of Applied Microscopy. 



colors, but that is not staining in any high sense of the word. Staining means 

 such use of colors as to differentiate structure for accurate observation. And 

 let me here say, that there are no line-drawn rules of universal application ; it is 

 largely an empirical art, and the best results, as a rule, are wrought out by care- 

 ful and intelligent experimenting, and my aim is to offer a few simple and gen- 

 eral rules which may serve as guides in lines of work within limited means and 

 opportunities. 



Reagents. — There is a long list of reagents in use, all of more or less value, 

 but I mention a few only, with their use, which will be found most effective in the 

 line of work before us ; and I will preface this part of the subject by mentioning 

 a single condition without which success is impossible, and that is cleanliness. 

 No dish, or glass, or instrument should ever be used without cleaning the last 

 thing before using. Water, whether distilled or not, should be filtered, and all 

 staining fluids will bear to be watched, and filtered as often as the least impurity 

 appears. 



For staining cell walls there is nothing superior to Delafield's haematoxylin, 

 but nine-tenths of the staining done with this reagent is so deep as to render the 

 object opaque and comparatively worthless. This fluid should be diluted with 

 the purest water obtainable till the color is a very little deeper than that sought 

 in the object, and allowed to stain slowly, or the section- should have a very brief 

 immersion in full-strength fluid — always rinsing in water. Haematoxylin has a 

 delicate reaction, is permanent, and when properly used nothing is better for 

 many things, if quite as good ; but when unskillfully done, no staining is worse. 



Bismark brown is a very valuable, but, as far as my observation goes, a very 

 much neglected stain. It is somewhat opaque, and therefore not adapted to 

 compact tissue, but where lace-like structure, or large spiral or scalariform ves- 

 sels are to be shown, it has no equal. In combined compact, and loose structure, 

 it has all the striking, and brilliant eftect of double staining. Use the same 

 methods as with haematoxylin. 



The easiest and most effective double staining is done with red and green, 

 and of these colors Grenacher's borax carmine, and methyl, or anilin green, will 

 be found most generally reliable. Borax carmine is more properly an animal 

 stain, but it is of the highest value in botanical work, when it plays its part in 

 double stain. It requires time, and does not color deeply, but it is transparent, and 

 can be made deep enough for all practical purposes if the following method is 

 followed : first place the section in borax carmine for twelve hours or more, as 

 found necessary ; wash rapidly, yet thoroughly, in 50 per cent, alcohol ; place for 

 two or three seconds in a saturated solution of any of the anilin greens, prefera- 

 bly anilin or methyl ; wash as before, and return to borax carmine till the red 

 reappears, changing the borax carmine after the superfluous green is driven out. 

 As soon as the outlines of the colors are distinctly marked, and secondary tints 

 disappear, rinse as before, and place for a few moments in alum cochineal ; this 

 acts as a mordant to the borax carmine without destroying its transparency. 

 This process, like any other, is liable to failure by over-doing or under-doing 

 some part of it, but when successfully followed the result will be all that can be 

 desired. It may seem superfluous to call attention to the chemical relations of 



