144 PHYSIOLOGY OF NUTRITION 



branch, the vesesls of which are not thus plugged, may remain turgid for a 

 number of days. 1 



Air as well as water is present in the vessels and is very much rarefied at 

 times, as was shown by Hohnel's experiments. To show the presence of water 

 in the vessels, a piece is removed from a young stem by means of a double pair 

 of shears, so arranged that the two cuts are made at the same time. From the 

 piece thus obtained, longitudinal sections are prepared and examined under the 

 microscope, of course without any addition of water. If the two cuts are not 

 made simultaneously no water is observed in the vessels, for, because of nega- 

 tive pressure in the gases of the wood, air rushes into the vessels at the cut 

 surface as soon as the incision is made, driving the water before it into other 

 regions of the plant. The water columns in the vessels are frequently interrupted 

 by air bubbles and these may be demonstrated under the microscope. To 

 accomplish this the parenchymatous tissue is carefully removed from one of the 

 woody bundles of a young stem with but little wood (e.g., Begonia or Dahlia). 

 Thus the bundle is exposed, but is uninjured and is still in connection with the 

 rest of the plant at both ends of the preparation. Study of such preparations 

 shows that the vessels are nearly filled with water and contain but few air bub- 

 bles in moist, cloudy weather, but that they contain less water and consequently 

 a greater amount of air 2 on sunny days. 



All the investigations that have so far been made indicate that the water 

 columns in the vessels are not completely broken by air bubbles. Cross-sec- 

 tions of the vessels show that they are not perfectly cylindrical but are more or 

 less prismatic and many-sided and that this irregularity in form is further in- 

 creased by circular, spiral and other secondary thickenings of the walls. Air 

 bubbles tend to assume a spherical form and the irregularly shaped portions of 

 the vessels are thus not completely filled with air, so that a continuous water 

 column results, the air bubbles being wholly surrounded by water/ 



1 Errera, Leo, Ein Transpirationsversuch. Ber. Deutsch. Bot. Ges. 4: 16-18. 1886. 



2 Capus, Guillaume, Sur l'observation directe du mouvement de l'eau dans les plantes. Compt. rend. 

 Paris 97: 1087-1080. 1883. 



r It is doubtful whether this is true when the transpiration rate is considerable and the soil 

 fairly dry. Wherever a gas bubble occurs in a vessel it should enlarge, under these conditions, 

 until it fills that entire vessel segment from the cross-wall below to the one above. The gas- 

 liquid surface tension in such a case as is postulated in the text would have to be as great as the 

 sum of the gas pressure in the enlarged bubble and the tensile stress exerted upon the water by 

 the transpiration process going on above. The gas pressure in the bubble must be less than a 

 single atmosphere, but the magnitude of the tensile stress is at least more than equivalent 

 to an atmosphere. Thus the sum just mentioned is frequently of the order of several atmos- 

 pheres and is surely of greater magnitude than the gas-liquid surface tension. It follows that 

 the bubble must enlarge until its surface film comes into contact with the surrounding vessel 

 walls at every point; thus reinforced, the surface layer of the liquid can withstand the great 

 attraction exerted by the stressed water-mass, and the gas bubble does not expand farther. 

 When the water has been under stress for a sufficient time there should be no free water 

 between cell walls and gas at any point in the entire plant body; all such surfaces should be 

 cell-wall surfaces, at which the liquid surface is held by the force of imbibition. Indeed, this 

 condition would probably be attained by the action of the gas pressure within the bubble, 

 before any stress developed in the liquid at all. The picture presented in the text at this 



