EXPERIMENTS ON WATER-CONDUCTION 316 



is devoid of true vessels, sap occupies at least two-thirds, and at most 

 nine-tenths of the tracheal cavities. Hartig's estimates, however, take 

 no account of the fact that different vessels in one and the same woody 

 cylinder may often contain very different amounts of water at the 

 same moment. 



The' author has shown that the rudimentary tracheides in the 

 central strands of Moss-stems likewise contain a watery liquid, which 

 after active transpiration becomes replaced by rarefied air. In isolated 

 cases (Mnium undulatum, Polytrichum) these elements of the central 

 strand contain starch-grains and drops of oil, even when they are fully 

 developed. 



The preceding discussion of the contents of vessels and tracheides 

 naturally leads to a consideration of the experimental and anatomical 

 data from which it may be inferred that the elements in question do 

 not merely contain water, but are actually engaged in the transporta- 

 tion of this substance. Attention may first be directed to the experi- 

 mental observations that have been carried out upon the supply of 

 water to vessels and tracheides, and upon its removal from these organs 

 under normal conditions in the intact plant. It is known, on the one 

 hand, that the water-conducting tubes of the root become filled with 

 water owing to the osmotic action of the living parenchyma, while, on 

 the other hand, it has been shown, that, in leaves, a more or less rapid 

 depletion of the vessels and tracheides is effected by the osmotic 

 " suction " of the transpiring mesophyll cells. The co-existence of 

 these two processes necessarily implies a transportation of water in that 

 portion of the vascular system which intervenes between the absorbing 

 roots and the transpiring leaves. 



From the earliest days of physiology onwards, attempts have been 

 made to obtain an ocular demonstration of the conduction of water. For 

 this purpose it is customary to make use of coloured liquids, or of solu- 

 tions which give rise to coloured precipitates on the addition of suitable 

 reagents, the liquid being in either case applied to the severed ends 

 of branches or to the roots of intact plants. Experiments of this 

 nature were carried out as early as the eighteenth century, and have 

 been frequently repeated, in the original fo*rm and with various modifi- 

 cations, up to the present clay. In every instance the conclusion to be 

 drawn is the same, namely, that the liquids rise most rapidly in the 

 vessels and tracheides. It should, however, be noted that the rate of 

 ascent of the liquid is, in such cases, generally estimated by the coloration 

 of the cell-walls in which the pigment becomes deposited, a circumstance 

 which may easily lead to erroneous conclusions, inasmuch as the water- 

 conducting capacity of a cell is, of course, in no way related to the 

 staining properties of its membranes. A priori, therefore, one would 



