TWINING OF TENDRILS. 
869 
verse streaks and measured after they had coiled, show that the growth of the con- 
vex surface is more considerable, that of the concave surface less so than in the 
portions of the same tendril that have remained straight above and below the curved 
part. A tendril of Cucurbita Pepo twined round a support 1*2 mm. thick; after the 
curvature was complete, the increment of the curved part for each millimetre of 
original length was 1-4 mm. on the convex surface, while on the concave surface 
it was only cimm.; the mean increment on both surfaces in the portion that 
remained straight amounted to 0*2 mm. If the growth which takes place in the 
entire tendril at the time of contact with a support is small, a considerable accelera- 
tion occurs on the convex surface, but in general there is no elongation on the 
concave surface, or there may even be a contraction ; in the case of a tendril of 
Cucurbita this contraction amounted to nearly one-third of the original length. 
Similar alterations in the length of the convex and concave surfaces are observ- 
able in the spontaneous coiling of free, as well as in the coiled portion of attached 
tendrils between the base and the point of attachment; and since in these cases 
the amount of growth which takes place in the entire tendril is usually small a short 
time previously, the contraction of the concave surface is, according to de Vries, a 
very common phenomenon. 
The conclusion to be derived from these phenomena and from others not 
described here is that the growth of the surface not in contact is first of all increased 
by the pressure of the support; the support presses the surface that is in contact,, 
and the pressure which the concave surface undergoes arrests its growth, or even 
causes a contraction in it. It seems probable that a relaxation of the parenchyma 
of the surface in contact (by giving off water to the parenchyma of the upper sur- 
face) and a consequent elastic contraction of its cell-walls contribute to this result ; 
at least this seems the only explanation of the contraction of the surface in contact 
in the case of tendrils the growth of which has already become slow. We have 
however as yet no knowledge of the mode in which the slight pressure of a light 
thread or that of the revolving tendril on a support causes this alteration of growth 
not only at the point of contact, but along the entire tendril. 
The only cause of the spontaneous coiling of tendrils when not fixed to a sup- 
port is that the upper surface continues to lengthen for a considerable time after 
the growth of the under surface has ceased. The cells of the growing upper sur- 
face probably withdraw from those of the under surface a portion of their water (as 
the inner layers of the pith from the outer layers, see p. 805), which causes the 
latter to become shorter, and the former to become longer. 
Without entering further into the numerous questions of a purely mechanical 
character connected with the curving of tendrils, it may at least be explained why 
thick tendrils are unable to twine round very slender supports. If two tendrils are 
compared one of which twines round a slender, the other round a thicker support, it 
will be seen that in the former the proportional difference in length of the outer and 
inner sides must be greater than in the latter. If a thick and a slender tendril twining 
round supports of equal thickness are compared, the proportionate difference in 
length of the outer and inner surfaces will be greater in the former than the latter case; 
and if the support is supposed to decrease constantly in thickness, the difference will 
increase more rapidly in the case of the thick than in that of the slender tendril, and 
