72 Newcombe, Thigmotropism of terrestrial roots. 



at 28 "C; but all roots remainod straiglit during tlie 6 hoiirs 

 of tlie expemment. The period, it is true, was ratlier short, 

 but consideration must be given to tlie usual capability of guick 

 response in a temperature near tlie Optimum. 



Tlie foregoing experiments do not prove tliat tlie roots nsed 

 are not tliigmotropic, but ratlier tliat tliey are not tliigmotropic 

 when iindergoing pressure over a very small area. To increase 

 tliis area of pressure, loops of very tliin rice paper and of tliin 

 collodion membrane were substituted for tlie rubber bands used 

 in the last 2 series of experiments. These loops were made 

 from Strips of membrane about 3 mm wide and 5 cm long. 

 To fasten them to the U-shaped wire, a cube of cork 3 mm 

 in diameter was inserted between the free ends of the wire, 

 and the ends of the paper or collodion strip were cauglit be- 

 tween the wire and the block of cork. It is best to liave the 

 membrane thus separated at its attached ends; for if the ends 

 are allowed to toucli wliere attached, it is often difficult to 

 separate the sides of the loop for the Insertion of the root. 

 When the loops are caught over the roots, the wliole breadth 

 of the strip can be made to bear on the root by regulating the 

 height of the pendulum. Moreover the small resilience of these 

 Strips causes them to embrace the root closely, so that probably 

 the loop makes contact over neaiiy half of the cylindrical sur- 

 face of the root. Unfortunately this relatively extensive area 

 of pressure cannot be long maintained witli a growing root; 

 for the root carries the strip down with it, and as soon as the 

 strip of membrane leaves the horizontal position, this area of 

 contact and pressure must of necessity be reduced. In the fol- 

 lowing experiments, the preparations were closely watched dur- 

 ing the daytime, and the pendulum adjusted so as to keep as 

 much of the loop as possible in contact with the root. Ten seed- 

 lings of Zea mays (everta) had these loops hanging against tlieir 

 roots for 11 liours, and one became positive and 9 remained 

 straight. Five seedlings of Pisum sativmn after 12 liours sliow- 

 ed 3 roots with positive curves, wliile the other 2 grew straight. 

 Four seedlings of Fagopyrum esculeiitum after 4 hours sliowed 

 all 4 roots positive. With seedlings of Baplmnus sativus, 2 tests 

 were made: With 14 seedlings in 12 hours, 3 became positive, 

 while 11 remained straight; with 16 seedlings attended day and 

 night for 21 hours, nearly all roots were neutral, only a few 

 indefinite nutations occurring. 



In the foregoing series of experiments there are several 

 good indications of thigmotrojnsm. Now if a collodion loop 

 brings responses because of the greater area of pressure on the 

 root, then 2 loops hung one above the other on a root ought 

 to bring still l^etter resjionses, provided the upper part of the 

 elongating zone is sensitive as well as the lower. In conformity 

 with this notion, the roots of 7 seedlings of Plsum sativum had 

 each 2 loops of collodion hung over them, eacli looji made of 

 a strip about 3 mm wide. The 6 mm of root covered by the 



