1897] | THE CURVATURE OF ROOTS 351 
of the cells of the normal and curved portions had been dis- 
tributed by the subsequent growth, which is of course modified 
by the tension set up by curvature. Sachs raised the objection 
that Cieselski’s method concealed the true relations of the 
length of the cells of the convex side to the normal, and that 
the excessive growth of the former was not apparent. In an 
effort to evade this error, Sachs compared the length of the 
cells of the curved portion with averages attained from the 
measurement of from twenty to forty cells in regions apically 
and basally to the curvature. 
According to his own account, the apical portion of the root 
was allowed to obtain a length of 2 to 3°, and the basal portion 
had made its full growth. He deemed it desirable to allow the 
curved portion to make the sharpest angle possible, and to 
reach a great thickness. It is evident that his results do not show 
the relative stature of the cells of the two sides at the time of 
curvature, since the subsequent growth processes have inter- 
vened. His figures are therefore strictly comparable to those 
given by myself in table VIII, made from curvatures three to 
five days old. Sachs found that the average length of cells of the 
root of Vicia Faba, apical and basal to curvatures, was 40 to 44 
respectively, with a general average of 42. The length of cells 
of the convex side was 41, and of the concave side 26.3. In a 
Second example the lengths of the apical and basal portions 
were found to be 23.2 and 26.2, with an average of 24.6. 
The average length of cells of the convex side was 28.3, and 
of the concave side 15. In a root of disculus Hippocastanum 
the average lengths of the apical and basal portions were found 
to be 16 and 23, that of the cells of the convex side 27, and 
of the concave side 1 3-3. Ina second example of this species 
the lengths of the cells of the apical and basal portions were 
found to be 19 and 21.2, with an average of 20.1. The average 
length of cells of the convex side was 28.1, and of the concave 
side 9-3. The figures given by Sachs represent divisions of the 
Micrometer of a value of .005™™.. 
The fact that the average length of the cells of the convex 
