THE CONDUCTION OF WATER. II 71 



fere with the capacity of the wood for transporting water. We shall refer later 

 on to the conditions which bring about a diminution in the air in the vessel. 



Before considering in greater detail the distribution of air and water in 

 the vessels, we must briefly allude to one consequence of the reduced pressure 

 of the air in the vessels which is familiar to every one in the course of everyday 

 life, and which is also of great importance in physiological experiments. 

 If a branch be cut off, without taking any special precautions, and placed in 

 water, it very soon begins to wither, because, during the operation, air has entered 

 the open vessels and may ascend to a variable height in accordance with the 

 amount of negative pressure in the vessels. If, however, the branch be cut off 

 under water, the water is forced into the vessels by atmospheric pressure, and the 

 cutting does not wither. The entrance of water naturally equalizes the negative 

 pressure, but this may reappear, if the cut surface of the twig becomes im- 

 permeable to water owing to slime exuded from the plant accumulating on it, 

 or owing to the varied methods of blocking up of the vessels due to the plant's 

 own activity ( WIELER, 1888), or, finally, owing to the action of Bacteria. Con- 

 tinued transpiration again induces withering, but this we can easily overcome by 

 forming a freshly cut surface above the first made, of course also under water. 



The negative pressure of the air in the vessels, as we have already seen, 

 must render the determination of the distribution of water and air in the vessel 

 very difficult. On undertaking an investigation on selected branches, as a rule 

 it will be found that the results arrived at are perfectly useless, because, in the 

 process of cutting, the basal parts of the branches have become filled with air. 

 It is preferable, therefore, to cut the branch off at two places at the same time, 

 or to isolate a cylinder from a tree trunk with a Pressler's growth borer: by 

 either means the air, oil, or mercury enters both sides at the same time, and hence 

 the air-bubbles originally present, mixed up with water, are forced into the 

 central region of the preparation, where not only the length of the parts with 

 water and air may be measured, but also where the initial rarefaction of air may 

 be easily determined. The rarefaction of the air is, according to SCHWENDENER 

 (1886), generally only about one-third of an atmosphere, rarely values of a 

 quarter to one-fifth of an atmosphere are reached, almost never less. Whilst only 

 one air-bubble can occur under such conditions in the middle of a tracheid, driven 

 there by the pressure of water from both ends, numerous air-bubbles always 

 occur in the tracheae, separated by water columns. The alternating air-bubbles 

 and water columns show not inconsiderable differences in size, still the follow- 

 ing averages, which SCHWENDENER (1886) obtained in case of Fagus sylvatica, 

 may stand for a fair representation of the usual distribution of water and air in 

 the vessel : 



Air-bubbles (mm.) 

 Water column (mm.) 



Total (mm.) . . 0-546 0-476 0.380 0-413 0-486 



On an average a segment of the chain composed of air-bubbles and drops 

 of water is about 0-5 mm. in length. Such a series is known as a JAMIN'S 

 chain, and it is obvious that the movement of water in such a chain must take 

 place under conditions essentially different from those obtaining in a tube 

 completely filled with water. 



SCHWENDENER does not say from what part of the wood the vessels upon 

 which he made his observations were taken, and, according to STRASBURGER'S 

 (1891) results, the amount of air in the vessels in the periphery of the wood, 

 as also in the younger parts generally, is much less than that in the older annual 



