1808.] of " Connecting Threads" in the Cell Wall. 505 



vulgare, substitution of the Safranin by Eosin is advantageous, 

 and for many ferns Poirier's orange may bo similarly employed. 

 In some instances I have used Cyanin with favourable results. 



5. Mounting. The sections are placed in dilute glycerine and 

 then transferred for permanent mounting into glycerine jelly. 



Some observations on the rationale of the method may now 

 follow. 



As regards the Picric acid no more need perhaps be said than 

 that the lethal action of this quickly penetrating reagent upon 

 the protoplasm is a well-known matter of experience, and that 

 living tissue exposed to its poisonous influence rapidly succumbs. 

 Besides killing the tissue, it swells the walls. This swelling of 

 the cell wall seems to me to be essentially a phenomenon of 

 hydration, and it is possible that the protoplasm is similarly 

 affected. The action of Sulphuric acid is not unlike Picric acid 

 itself, and possibly Acetic acid also belongs to the same category. 

 Certain of my experiments appear to indicate that both Picric 

 acid and Acetic acid function as mordants both for Safranin and 

 Gentian violet, though in view of the great solubility of both 

 reagents and the comparative readiness with which they may be 

 removed from tissues by washing, the real explanation may be; 

 that cytoplasm or cellulose after having been submitted to these 

 actions has a much greater affinity for the two dyes than before. 

 At the same time it must be remembered that Picric acid gives 

 a precipitate both with Safranin and Gentian violet. 



Kolossow's mixture is a far more deadly poison than Picric 

 acid ; is exceedingly rapid in operation ; and fixes the tissue with 

 little or no change. It also hardens the structures and makes 

 them very favourable for cutting and subsequent manipulation. 



Unlike Picric acid there seems to be no question of hydration 

 here, and the action of both the Osmium and Uranium salt 

 appears to consist in forming a fixed and insoluble compound 

 with the tissues exposed to their influence. 



From the point of view of staining it is the Uranium which 

 plays the more important part in that it and not the Osmic acid 

 acts as a mordant to the Safranin. In the case of Hermann's 

 mixture it is the Platinum salt which performs a similar function. 



The Uranium on the other hand does not mordant Gentian 

 violet, and no satisfactory results can be obtained by staining 

 with Gentian violet direct. Thus the staining has always to be 

 effected through the intervention of Safranin, which with this 

 particular method must be taken as the starting point of all 

 further treatment. 



It must not however be imagined that this fact constitutes an 

 objection to the use of Safranin. Safranin is on the contrary 

 almost an ideal dye for such an investigation as the present. 



