174 BOTANICAL MICROTECHNIQUE. 



power is partly destroyed by Schulze's macerating mixture, 

 evidently by the solution of the substance which causes the 

 stronger refraction ; and correspondingly, the membranes so 

 treated are evenly stained throughout by methylene blue. 



2Q7p. Finally, the carbonization or pulverization methods 

 introduced by Wiesner (V) into botanical microscopy may 

 be here described. By these the vegetable cell-wall is- 

 broken up into filamentous and then into spherical bodies, 

 which are called by this author dermatosomes, but whose 

 significance cannot be discussed here. 



Linen fibres may serve as suitable objects for the trial of 

 pulverization methods. These are laid, according to Wies- 

 ner (III, 14), for 24 hours in a \% solution of hydrochloric 

 acid, then freed from adhering fluid and warmed to 50 or 

 60 C., until the substance is completely dry, which can be 

 accomplished in 30 to 50 minutes if small quantities of fibre 

 are used. The fibre may then be broken into an extremely 

 fine powder by gentle pressure. 



With other objects a longer stay in hydrochloric acid, or 

 the use of higher temperatures for drying, is necessary. 

 Wiesner accomplished the pulverization of endosperm-cells 

 of Phytelephas only after the action of hydrochloric acid for 

 months. 



Pfeffer (IX) has lately shown that the carbonization 

 methods lead to the same results with artificially prepared 

 collodion membranes. 



B. The Protoplasm and Cell-sap. 



298. Conclusions concerning the morphological characters 

 of the protoplasm have been sought for not only by direct 

 observation of living material, but also by means of micro- 

 chemical reactions and staining methods. Although it would 

 seem probable a priori that microchemistry would prove 

 an aid of the first importance, it has not yet justified these 

 expectations ; which is due largely to the fact that macro- 

 chemical studies of the structures in question have been 

 carried through with any exactness in very few cases, since 



