88 FILTRATION EXPERIMENTS. [CH. IV 



(101) Permeability of cell membranes 1 . 



Negative pressure depends on the fact that although 

 the walls of the xylem elements are extremely permeable 

 to water, they present a great resistance to the passage of 

 air as long as they are wet. It is therefore of importance 

 to show that wet wood does not easily allow air to pass 

 while the same wood dried is easily permeable. 



The only important point about the experiment is to 

 make sure that air emerging under pressure from dry 

 coniferous wood is really passing through the walls of the 

 fcracheids and not escaping from the protoxylem or from 

 flaws and cracks in the branch. We find that the wood of 

 Pinus sylvestris is decidedly better than that of Taxus. 

 We use branches about a centimeter in diameter cut into 

 bits of 2 to 4 cm. in length. The branch should be cut and 

 brought into the laboratory without being placed in 

 water. We have not generally used it until 2 or 3 hours 

 have elapsed, but it is probably unnecessary to wait so 

 long. The most effective way of blocking the vessels of 

 the protoxylem is to gouge a small cavity at the centre of 

 the branch, which is afterwards filled up with modelling 

 wax melted in with a hot wire. If this is done at both 

 ends of the bit of wood the vessels will be sufficiently 

 closed. 



Another plan is to turn cylinders of 1 cm. in diameter 

 from the splint wood of Pinus, or of the silver fir (Abies 

 pectinata), which Strasburger especially recommends. Tho 

 wood should be placed, while fresh, in alcohol, and if the 



1 von Hohnel, Pringsheim's Jahrbiicher, xii. 1879, p. 63 ; Pfeffer, Physio- 

 Icgie, i. p. 86 ; Strasburger, Leitungsbahnen, p. 729. 



