GEOTROPISM. 125 



the lower, should be smoothly burnt a small hole to receive the tip 

 of the root (so that the latter may not nutate from the center of the 

 cork). Place some soaked horse-beans in moist sphagnum with 

 hypocotyl pointing directly downwards, and leave until the roots are 

 nearly 2 cm. long. Pin one of the stoutest firmly to the under 

 side of a cork fitting the test-tube; in the cork vertical grooves are 

 to be filed to admit air. The root should point vertically down- 

 wards, and the seeds should be wrapped in moist sphagnum, 

 Place just so much mercury in the test-tube that, when the cork is 

 in place and fastened with wire, the tip of the root rests lightly 

 inside the hole burned in the upper cork. Set the apparatus in a 

 dark and moderately-warm place, and observe frequently. An at- 

 tached scale, opposite which one of the needle-points moves, allows 

 the exact amount of depression of the float into the mercury to be 

 observed. When the root has pushed down the float to a point 

 at which a lateral bending of the root begins, the point should be 

 noted. The root is then to be removed, and gram weights are to 

 be placed upon the float until the latter is depressed to the same 

 point to which the root depressed it. The power of the root to 

 grow vertically downwards will thus be given exactly in grams. 

 The readings should be taken at the same temperature at which the 

 experiment is started. 



The above arrangement gives a very exact and practical method of 

 measuring these pressures, and it can doubtless be adapted to other pur- 

 poses (see page 451. With it I have obtained pressures of over 30 grams 

 with bean roots, before they bend. To overcome the slight difficulty of 

 securing the proper height of mercury in the tube, the arrangement in 

 Fig. 34 on the right is good. By sliding the right-hand tube up and 

 down with the clamp on the rubber tube loosened, the level of the mer- 

 cury can be adjusted very exactly. In the case of very delicate roots 

 water could be used in the test-tube instead of mercury. In case a par- 

 ticular root shows any traumatropic bending due to the burnt cork (as 

 may possibly be the case), a tiny glass cone, made from small glass tubing, 

 may be sunk into the upper end of the cork. 



Another method, which appears effective but not easy to apply, is 

 that given by Sachs, in which the growing roots are made to push 

 against tiny scale-pans connected over a pulley with small weights. A 

 modification and improvement of this is proposed by Stone, in Botanical 

 Gazette, 22, 293. Sachs' experiment, in which the root grows down- 

 wards against the resistance of mercury, the resistance causing an up- 

 ward bending just behind the turn, is well known but not easy to make 

 work entirely satisfactorily. An easier way to apply the root to the 

 mercury is the following : Take a block of wood 8 or 10 cm. long, 4 wide, 

 and 3 thick ; bore with an auger a large hole near one end 15 mm. deep. 



