226 Practical Plant Biology. 



tube was sealed off by melting the narrow steam-exit in a flame. 

 This last operation secured that very little air remained in the 

 tube. Hence the tube now acts as a water-hammer, i.e. when, 

 by inclining it, the water is allowed to rush from end to end, 

 its concussion makes a sharp metallic click. Evidently very 

 little air is left in the tube to act as a cushion and to soften 

 this concussion. If by inclining the tube the long arm is com- 

 pletely filled with water and the tube is gradually raised again 

 into a vertical position, most likely the water will behave just 

 as it did when the air-pump was applied and will fall down 

 into the bend of the tube. If the tube is again gently inclined 

 the long arm will fill up with water, but unless the expulsion 

 of air has been very perfect a little bubble will persist in it. 

 Now the tube is inverted and this bubble rises towards the 

 upturned bend, and it may be chased out into the space in the 

 short arm. When the tube is brought into the vertical position 

 for the second time, it will probably be found that the long arm 

 will remain filled with water, in spite of the inequality of level 

 in the two arms. The water in it is held in the same position 

 as the water was held by atmospheric pressure in the first ex- 

 periment. Here the atmospheric pressure has been removed 

 and the water is held in the long arm by its adhesion to the 

 glass and its own cohesion or its tensile strength. The weight 

 of the lower water is supported by the cohesion of the column 

 above it. A sharp percussion at the bend will overcome its 

 cohesion, and the water, deprived of its support, will fall into 

 the bend of the tube. The conditions for this cohesion to 

 come into play are the elimination of bubbles and the adhesion 

 of the water to a rigid envelope. Evidently if there is a single 

 bubble in the water column, the surface-tension of which is unable 

 to sustain the weight of the water below, this bubble will be 

 enlarged and enlarged until the column is broken across. Similarly 

 if there is any spot where the water does not adhere to the wall, 

 a bubble will be formed there and the column will be broken. It 

 is of interest to notice that if we could enclose the bubble in a tiny 

 cell, the outside of which would be wetted by the water, the com- 

 plete breaking of the column would be prevented. It should 

 further be observed that, although undissolved air in the form of a 

 bubble, unless thus enclosed, breaks a tensile column of water, ex- 

 periment has shown that dissolved air does not destroy its cohesion. 

 We will now turn our attention to the water in the plant. 

 The moistened walls of the cells of the mesophyll lose water 

 into the spaces of the leaf. To make good this loss water is 



