348 BOTANICAL GAZETTE : (May 
this objection the seedlings were in several experiments fastened 
to the supporting bar by themselves and the encasing tubes 
were fastened to the same bar by a separate band just below the 
band holding the seedlings. Such a preparation enabled the 
glass tube to be moved up or down while the seedlings remained 
fixed. Results showed that there was no advantage in this more 
troublesome preparation, and it was discarded as unnecessary. 
All of the following experiments were made with Raphanus 
sativus (Early Long Scarlet), the roots of this plant being 
insensitive to diffused light, and light being necessary at fre- 
quent periods of observation. 
For the first series of experiments the apical 2™ of the root 
were kept below the opening between the two parts of the paired 
tube, and not more than 2™™ of root were exposed to the water 
current. In one set of seven seedlings, at the end of 7 hours’ 
revolution, five roots were positive 5 to 40°, while the other two 
roots were straight. In a second set of six roots, at the end of 
4 hours 40 minutes, five were strongly positive, and the sixth © 
root was straight. 
These results show that the part of the root just 
apical 2™™ is sensitive. 
In the next series of experiments ten seedlings were used 
whose apex was covered for a distance varying from 3 to 8 : 
Above the covered portion of the root, a portion 5 to 3m teed 
was exposed. After 8 hours eight roots were positively bent 
and two roots were straight. Bae 8 
__ Five experiments, including forty-eight seedlings, Wé made, 
in which 10™™ of the apex of the roots were covered, a! : 
root left exposed for 5 to 8™™ above the covered part. 
duration of the experiments varied from 10 to IT mos 
above the - 
the end, thirty-four roots were positive, three were negat 
_ eleven were neutral. a a 
Two experiments, including a total of twenty-three : 
re COVETEUr 
| The duratio? 
the other 17 
lings, were made in which 15™™ of the apex we 
Dod oil above the covered part were exposed. 
_ Of the experiments was in one case 10 hours, in 
