TRANSPIRATION. 79 



giving guidance to the watering, it allows a somewhat freer access of 

 oxygen to the roots, and it makes possible the experiments described 

 under Experiment 21. Another method which has advantages is shown 

 in Fig. 14 on the right, where the plant is removed from the pot and 

 placed in a battery-jar with enough extra earth to fill the latter ; the jar 

 is then covered with the rubber precisely as described in the above 

 experiment. In this case the plant should be allowed to stand for a day 

 or two before using, in order that it may become accustomed to the new 

 conditions. (See Note 4 of the Addenda.) 



Another method, with manifest advantages, of preparing the plant is 

 given by Barnes (Plant-life, 394) as follows : " Clean and dry the surface 

 of a pot in which a thrifty single-stemmed plant is growing ; close the 

 hole in the bottom with a cork ; with a brush paint the whole surface 

 with a thick layer of melted paraffin. Cut out a piece of stiff paper 

 which will fit around stem and just cover the soil in pot. Using this as 

 a pattern, cut a cover for the soil from a sheet of lead ; slit the cover from 

 the central hole to circumference, adjust it around plant, and cement 

 all cracks with grafting-wax." Or the pot can be set in a tin vessel 

 which it fits and the lead cover luted to this. Professor Barnes adds (in 

 a letter) : a disk of filter-paper should be added to the bottom of the 

 paraffined pot to prevent adhesion to the table, etc. 



A spring-balance for the weighings is much more convenient (par- 

 ticularly in saving time) than a balance using weights, but, even if of the 

 best kind, is accurate enough only for very rough work (see page 33). 

 An ordinary balance can be used and should be sensitive to i gr. ; the 

 torsion form is excellent. For very exact work a more accurate balance 

 is needed, and one especially fitted for the purpose is described by 

 Hansen in Flora, 84, 355. Very well adapted to this work is the 3423 

 balance of Gerhardt (see page 33). An autographic transpiration 

 machine is described by Copeland in the Botanical Gazette, 26, 343, and 

 another on the general principle of Copeland's, but taking a potted plant, 

 by Corbett in the Twelfth Annual Report of the West Virginia Experi- 

 ment Station. The objection to machines on the principle of these two 

 is the difficulty of keeping the water in the reservoirs at a constant tem- 

 perature, as must be done to secure accurate results. A very exact form 

 of autographic transpiration-recorder is described by Woods in the 

 Botanical Gazette, 20, 473, and a registering balance by Anderson in 

 Minnesota Botanical Studies, No. 16. 



I have made attempts, with but indifferent success, to secure an 

 autographic transpirometer by keeping the plant on the spring-balance, 

 and running a thread from the top of the latter over the multiplying- 

 wheel brought into action with the recording cylinder (see later, page 

 102). This would work if the spring-balance were delicate enough. 



For taking temperatures a thermograph (page 33) is best. For 



