178 PHYSIOLOGY OF NUTRITION. 



water which has penetrated into the wood vessels from outside 

 (even into the widest ones), and is there retained by capillary 

 attraction, holds in equilibrium the final mercurial pressure. It 

 is also easy to see why the ultimate mercurial pressure will be 

 less if we experiment, e.g., with stems of Vitis, than if we use 

 stems of Sambucus, Prunus, or Cratasgus. The vessels in the 

 first plant, as we can easily ascertain by examination of trans- 

 verse sections under the microscope, are much wider than those 

 of the others, and therefore also in the Vitis stem the resistance 

 of the water retained by the vessels by capillary attraction is less 

 than that of the water which has penetrated the Sambucus, 

 Prunus, or Crataegus vessels. 3 



It is also of interest to satisfy ourselves that the length of the 

 vessels or segments of vessels is not nearly so great as we have 

 been accustomed to suppose. From a shoot of Alnus glutinosa, 

 about five years old, we isolate under water a middle piece about 7 

 cm. long. With the upper end of the piece of twig (that, viz., which 

 was originally directed towards the apex of the shoot) we now 

 connect by means of rubber tubing a glass tube 6 cm. long, and 

 connect this in turn with the air-pump. The lower end of the 

 piece of twig we dip into a fluid consisting of 3 parts of water 

 and 1 part of the officinal Liquor ferri oxychlorati.* On exhausting, 

 the brown fluid penetrates into the opened vessels. In the glass 

 tube, however, appears a perfectly colourless fluid. We continue 

 the exhaustion for about quarter of an hour, and then, by means 

 of garden clippers, cut away a portion from the lower end of the 

 piece of twig. If, continuing the injection, the fluid passing into 

 the glass tube still remains water-clear, we cut off another piece, 

 and so on, till finally the brown fluid begins to appear in the 

 tube. This happens in the Alnus twig when it has been reduced 

 to a length of about 5 '5-6 cm. 



The Ferric oxychloride is highly colloidal, and therefore cannot 

 traverse plant membranes, If accordingly, with a long piece of 

 branch, the Ferric compound does not appear in the tube, its pro- 

 gress in the vessels must have been stopped by a membrane (trans- 

 verse wall), and the occurrence of such imperf orate transverse 

 walls is in this way established. At the same time the injection 

 method affords a means of determining the length of the vessels. 

 The length is obviously that of the piece of twig when the iron 



* This fluid is an aqueous solution of Ferric oxychloride. 



