SPECIAL METHODS. 1/3 



its concentration before and after. It was thus shown that 

 no marked changes in concentration were produced by the 

 absorption of the starch. It had previously been shown 

 by Sachs that potassium ferrocyanide rises with the same 

 rapidity as the water in a strip of filter-paper. 



29711. For "silvering" Correns used essentially the meth- 

 ods which have been employed for a long time in animal 

 anatomy. He placed the previously well-dried objects first 

 in a 2-5$ solution of silver nitrate, and then, after super- 

 ficial drying, but without washing them, in a .75$ solution of 

 salt (NaCl). The silver chloride thus precipitated in the 

 membranes is best reduced by light, for which a few hours 

 in direct sunlight are sufficient. The objects are then dried 

 and mounted in Canada balsam. There may then be ob- 

 served in the parts rich in water a strong blackening due to 

 the reduced silver, which also forms, in part, small opaque 

 granules. Here again the bast-cells of Nerium and Vinca 

 are to be recommended, but they should first be washed 

 some time with water for the removal of silver-reducing 

 substances. 



According to Correns (III, 296), silver nitrate is accumu- 

 lated to a slight extent. A slight change in the concentra- 

 tion of the solution is produced by starch, and the silver salt 

 remains somewhat (about $) behind the water in filter- 

 paper. 



2970. Differentiations of the membrane due to chemical 

 differences are especially recognizable by the fact that they 

 are to be seen in the membranes whether dry or full of 

 water. Suitable objects for the study of this group of differ- 

 entiations are furnished by the large pith-cells of Podocarpus 

 tlongatus and other species (cf. Zimmermann I, 149). 



The so-called transverse larnellation of the bast-cells is 

 due partly to chemical differences and partly to unequal 

 water-content, according to Correns (II, 298). The chemical 

 difference is shown by the fact that the radial, more strongly 

 refractive lamellae remain unstained in a pretty concentrated 

 solution of methylene blue, while the rest of the substance of 

 the wall is deeply stained by it. The stronger refractive 



