412 
RICHARD M. HOLM AN 
In water the behavior of the secondary root was, in my experiments, 
very much the same as in moist air, although growth persisted much 
longer than in air and as a result the roots soon became so long that 
they bent downward under their own weight. There was only slight 
active and permanent curvature of the roots in water except when 
they formed a very large angle with the normal position of rest. 
In media offering resistance to the advance of the root, the behavior 
of the secondary root was directly parallel to that of the primary root. 
Roots which had grown for a time in the limiting angle in loose moist 
Fig. 3. Curvatures in loose moist sawdust and in earth of the secondary 
roots of Vicia faba when displaced 90° from the normal position. Ai and Bi 
represent roots growing in loose sawdust; A2 and B2, roots growing in earth. 
The arrows with dotted shaft show the direction of gravity relative to the root 
system at the time the secondary roots were of the length indicated by the cross 
strokes on the roots. It was at that time the cultures were turned through 90°. 
The arrows with solid shaft indicate the direction of gravity after the cultures were 
turned. Roots below the main root are omitted. Roots in and A2 grew for 
the same length of time in both positions; also roots in Bi and B2. 
sawdust curved very gradually toward the normal position when 
displaced upward from the original position, whereas roots growing in 
earth and similarly treated curved promptly and acutely into a position 
approximately the same as that from which they were displaced. 
This difference in behavior is represented in figure 3, which was traced 
