28 BOTANICAL GAZETTE | JANUARY 
to the unaided eye, and usually may be measured with an ordinary 
millimeter scale. A root of Vicia faba whose total length was 
1.5" had its cortex separated from the axial cylinder in the zone 
of growth, and immediately afterward was laid in water. After 
ten minutes the isolated strips of cortex were 9™™ long, and the 
axial cylinder was 8™™ in length, a difference of 12.5 per cent. 
computed on the length of the axial cylinder. To compare this 
difference with that found in stems we may take the figures given 
in Sachs’s Tezt-book, English edition, pp. 798-9, where the 
greatest difference in length of pith and epidermis of stems of 
Silphium perfoliatum is 13.4 per cent. This would probably have 
been somewhat greater had the pith and epidemis been separated 
from each other. Nevertheless the comparison shows that the 
amount of difference in the length of isolated tissues of roots 
may approach that of stems. It does not follow that the tissue 
tensions in the growing zone of roots represent as much poten- 
tial energy as those of stems. The resistance of the tissue under 
negative tension in roots is much less than that of the tissue under 
the same tension in stems, and for the reason that the vascular 
bundles are so incompletely developed in the growing zone of 
roots. This explains why the axial cylinder of roots is often 
broken by the negative tension to which it is subjected, and also 
why the halves of split roots do not always curve with the cut 
face concave immediately after splitting. This curve always 
follows sooner or later provided the halves continue to grow, 
except when the root is laid in a horizontal position with the cut 
face upward. Often in roots growing rapidly the curve follows 
at once after splitting. 
The root on which the above measurements were made was 
afterward plasmolyzed in I0 per cent. KNO, . The cortex @ | 
shortened 1™™, the axial cylinder, 0.5™™. Insome cases the axial 
cylinder shortens more than the cortex, a fact which the tables 
do not show. The fact that the axial cylinder is shorter than 
the cortex after plasmolysis proves that the difference in length 
is not due merely to turgor, but that there is normally a differ- 
ence in the rate of growth of the two tissues, 
re) 
