ABSORPTION Oh' WATER AND MINERALS. 69 



the additional pressure exerted by the liquid is | less f, that is of an 

 atmosphere. Since an atmosphere of pressure is roughly 15 pounds to 

 the square inch, the liquid is exerting a pressure of about 5 pounds to the 

 square inch. The temperatures should be about the same when the read- 

 ings are taken at the beginning and end of the experiment. Boyle's law 

 is not strictly true for damp air, though nearly enough so for rough 

 work. Greater exactness in the readings of the above gauges can be ob- 

 tained by making them in two pieces, a lower short piece attached to the 

 plant, which is then allowed to stand until this piece is partially filled 

 with water and all air-bubbles have escaped ; a film of oil is then placed 

 on the water, the upper piece, thoroughly dried, is then attached by a 

 pressure-tight joint and sealed. The oil keeps the air in the tube fairly 

 dry. 



Another method of measuring root-pressures is Pfeffer's (given in Det- 

 mer, 198). In using all such apparatus it is necessary to keep the tubes 

 as small as conveniently possible, because the volume of water given off 

 is often small even though its pressure may be high ; if the gauge is 

 large, the volume of water given off may not be sufficiently large to push 

 the index surface high enough to register the true pressure under which it 

 is being given off. The open S-shaped tube containing mercury, figured 

 in many books, I find rather impracticable as usually recommended, partly 

 because it is difficult to fill the space between plant and mercury with 

 water without admitting air-bubbles and disturbing the pressure- levels 

 to start with, partly because the lower arm of the tube (that dipping 

 below the top of the S) must be a very inconvenient length in order to 

 register a high pressure (thus it must be over 38 cm. long to register one 

 atmosphere). Moreover, as often figured, the tube is so large in diameter 

 that the plant could not give off quantity enough to push the mercury 

 high enough to register the actual pressure under which the liquid is 

 being given off, but this of course could be overcome by using a tube of 

 narrow bore. A modification of this arrangement, shown in Fig. 1 2 on 

 the right, works fairly well. The manometer tube is sealed (and pressure 

 calculated by Boyle's law), while the vertical corked tube allows all air 

 to be removed between mercury and plant. (See Note 3 of the Addenda.) 



A self-registering apparatus for measuring root-pressure is described 

 by Thomas, in Botanical Gazette, 17, 212. 



The large pressures shown by the preceding experiments 

 to result from osmotic absorption, suggest that elastic cells 

 must be stretched by such pressure and hence acquire turgiclity 

 and therefore considerable mechanical rigidity; and also that 

 the rigidity of delicate plant parts, lacking thick-walled ele- 

 ments, may be due to this cause. 



