ii6 



THE PLANT 



of maintaining the normal relation of soil and roots, a condition more or less 

 impossible in a pot. In both instances the weighing should be done in the 

 habitat, which was not the case in Hesselmann's researches. 



The slight value of the potometer, which has had a vogue far beyond its 

 merits, is indicated by the following table. These results were obtained 

 from three plants of Helianthus annuiis; III was left undisturbed in the 

 pot where it had been growing, IV was placed in a potometer, after the 

 root had been cut off, and V was an entire plant placed in a potometer. 

 The amount of transpiration is indicated in grams per square decimeter of 

 leaf surface. The plants were kept in diffuse light, except for a period of 

 two hours (8:00 to 10:00 A.M.) on the last day, when they were in full 

 sunshine at a temperature of 75 F. Plant IV wilted so promptly in the 

 sunshine that it was found necessary to conclude the experiment in 

 diffuse light. 



The cut plant, IV, lost more water the first day than either of the others, 

 l)ut the water loss soon decreased, and at the end of the period was almost 

 nil. The total transpiration for III and V is much the same, but the range 

 of variation for periods of 12 hours is from +2 to i gram. This ex- 

 periment is taken as a fair warrant that the use of cut stems in potometers 

 can not give accurate results. It is inconclusive, however, as to the merits of 

 potometric values obtained by means of the entire plant, and further studies 

 are now being made with reference to this point. 



159. Expression of results. From the previous discussion of the relation 

 between them, it follows that an expression of the amount of transpiration 

 likewise constitutes an expression of absorption. It is very desirable also 

 that the latter be based upon root surface and chre.sard, but the difficulty of 

 determining the former accurately and readily is at present too great to 

 make such a basis practicable. In expressing transpiration in exact terms, 

 the fact that plants of the same species or form are somewhat individual in 

 their behavior must be constantly reckoned with. In consequence, experi- 

 ments should be made upon two or three individuals whenever possible, in 

 order to avoid the error arising from this source. 



