208 . N. H. CO\VDRY. 



sistant to differentiation as are the mitochondria, but the 

 destructive effect is not so apparent as in the cells of the pea 

 radicle. 



The question of the reactions of plant and animal mitochondria, 

 after fixation, to stains may be dismissed with very few words, 

 When they have been fixed by any method, it is usually possible 

 to color them with any stain, by the Benda method, with fuchsin- 

 methyl-green, with fuchsin-picric acid or with iron hematoxylin, 

 but it is sometimes advisable to select some mordant in order to 

 induce the stain to take. The Benda method is the most difficult 

 and it is sometimes refractory, for no apparent reason, in tissues 

 where the others give excellent results. In other words there is 

 nothing whatsoever specific about the staining reaction. 



A number of supravital stains have been discovered for mito- 

 chondria in animal cells, of varying degrees of specificity, and 

 it is interesting to note that none of them have heretofore been 

 applied to plant mitochondria. The most delicate is the janus 

 green reaction, which we owe primarily to Michaelis ('99, p. 565). 

 The delicacy of this reaction is shown by E. V. Cowdry's ('16, 

 p. 429) observation that janus green B will stain mitochondria 

 specifically in human lymphocytes in a dilution of half a million 

 (i. e., one part of stain to 500,000 parts of salt solution), and that 

 janus green G will not stain them even in a relatively strong 

 solution. 



Janus green B is diethylsafraninazodimethylanilin: 



Janus green G is dimethylsafraninazodimethylanilin 



(CHOiNl /x /,-. / I N = N| / x ,N(CH 3 )o 



N CY 



\/ 



