178 PLANT PHYSIOLOGY 



Upon certain days alter the conditions in the house to as extreme de- 

 grees as practicable towards heat, coldness, humidity, darkness. Then 

 plot the results in a series of graphs arranged one above another, 

 and interpret the parallel and the divergent fluctuations. 



Transpirographs (Autographic, or Recording, or Registering 

 Transpirometers). Of these several forms have been invented. Some of 

 them, notably those of Vesque and of Krutitzky, register the quantity of 

 water absorbed, and have already been described (page 176). But the more 

 important forms register the actual loss from the potted plant. Pfeffer 

 ("Physiology," 1, 242) speaks of the possibility of using a spring balance, 

 magnifying wheel, and recording drum, but, as I know from trial, no such 

 instrument can be constructed to work with accuracy, since even the test 

 of such balances alter the tension of their springs with use, and also involve 

 irregular friction in the mechanism. A form, taking only a water-culture 

 plant, balanced over a wheel by a submerged float and recording upon a 

 revolving drum, is described by Copeland (Botanical Gazette, 26, 1898, 

 343); another on the same general principle, but taking a potted plant, is 

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

 ment Station. An objection to these forms is their comparative unporta- 

 1 ility and the difficulty of keeping their water at constant temperature, as 

 is essential if the float is not to change its buoyancy. A very different form 

 is the Evaporimfetre of Richard Freres of Paris, described in their cata- 

 logues; it is a balance of adjustable delicacy, recording the movement of 

 its beam directly upon a revolving drum, and using Masure's method earlier 

 mentioned (page 173), of compensating evaporation from pot and soil. But 

 this instrument has serious limitations in principle and practice. A vast 

 advance upon all earlier instruments was made by Anderson in his reg- 

 istering balance, described in Minnesota Botanical Studies, 1894, 177; 

 it is constructed upon a general principle closely followed by my own 

 instrument described below, and has only the demerits that it is costly and 

 must be specially made to order. Another very accurate instrument is that 

 invented by Woods (Botanical Gazette, 20, 1^95, 473); it is a modification 

 of the Marvin self-recording rain gauge. Finally I have constructed, and 

 there is offered among my normal apparatus (Botanical Gazette, 39, 1905, 

 145), a precision transpirograph, applicable to any good balance, the construe 

 tion and use of which is as follows. A cylinder, showing prominently in 

 the accompanying figure 45, contains on a spiral track between its outer and 

 an inner wall some 250 spherical gram weights. These weights are steel 

 bicycle balls of one-fourth inch diameter, which weigh almost exactly one 

 gram each, and vary not over one-thousandth of this weight from one another 

 These feed by gravity, one at a time, into a simple releasing valve, so arranged 

 that when acted on by an electromagnet a slide rises and allows one ball 

 to drop through a tube into a scale-pan, a new ball immediately taking its 

 place in the releaser-slide. Attached to the releaser slide is a bar carrying 

 a pen, so adjusted that every time the slide moves, that is, every time a ball 

 is dropped, the pen makes a vertical fine line with chronographic ink upon 



