690 PHYSIOLOGICAL BASIS OF POLARITY. II 



It was expected and found that the larger the mass of apical sis- 

 ter leaves the greater the mass of roots produced in equal time and 

 under equal conditions. Figs. 1 and 2 are typical experiments of 

 this kind. It was not possible, however, to prove that the two 

 masses of roots formed (in about 5 weeks) are alw^ays proportional 

 to the masses of leaves. This may be due to two sources of error 

 inherent in experiments on root formation. First, the roots we are 

 dealing with are air roots which usually dry out and fall off after 

 a short time, while new air roots begin to form. This alone would 

 make it almost impossible to prove an exact proportionality be- 

 tween mass of leaf and of root production even if it existed. The 

 second difficulty lies in the fact of the small mass of each root, 

 which makes the error in cutting off and in ascertaining the mass 

 of the roots rather large. The fresh as well as the dry weight of 

 the roots was determined in each case. The data given in Table I, 

 however, leave no doubt that the mass of roots increases when the 

 mass of leaves increases, though no mathematical expression of the 

 relation can be deduced, for the reasons stated. When the half stem 

 has no leaf the root formation is greatly retarded and diminished if 

 not suppressed, as is shown in Fig. 3. One set of half stems had an 

 apical leaf, while the other set had no leaf. The stems were lying 

 with their cut surface on moist filter paper. The stems without a 

 leaf formed practically no roots though they formed apical shoots. 

 The other set of half stems with an apical leaf formed an abundant 

 supply of roots but the shoot formation was suppressed by the apical 

 leaf. The drawing was made 26 days after the beginning of the 

 experiment. 



The writer has repeatedly called attention to the fact that not 

 only the root formation but also the geotropic curvature of the stems 

 increases with the mass of the apical leaves,** and this fact was con- 

 firmed in the new experiments. The stems in Figs. 1 and 2 were 

 originally perfectly straight, and the curvature noticeable in the 

 photographs is due to the growth in length of a certain layer of cells 

 of the cortex. These experiments show, in the opinion of the writer, 

 that the phenomenon of geotropic curvature is also a phenomenon of 



^Loeb, J., Bot. Gaz., 1917, Ixiii, 25; Ann. Iiisl. Pasteur, 1918, xxxii, 1. 



