225 
to contact with one another, but in all cases, provided a suf- 
ficient pressure is exerted upon the sensitive surface, a distinct 
curvature is always produced, though such tendrils only coil 
permanently around one another under very special conditions. 
That there is nothing peculiar in this is shewn by the fact, 
that such tendrils, though sensitive to contact with loosely 
hanging smooth and thin threads, hairs, or fine glass filaments, 
do not coil around such objects, but behave very much as if 
the object was a similar tendril. Just as contact with a thin 
and smooth tendril exerts a minimal stimulus for curvature 
but a sub-minimal one for coiling, so also is the case with 
smooth thin freely hanging objects resembling as closely as 
possible the tendril as regards the nature of the stimulus pro- 
duced by contact. If however the thread or one of the tendrils 
is coated with 10%, gelatine and kept moist, no curvature 
whatever is produced on contact. If on the other hand the 
threads are fixed at both ends or are from 2 to 3 times the 
thickness of the tendrils (Passifora, Sicyos) with which they are in 
contact, a rather loose but permanent coiling is readily produced. 
Another point which proves that the tendrils of the same 
plant do not in any way behave differently to one another, 
than they would do to similar foreign objects, is, that tendrils 
of Echinocystis ave sensitive to contact with tendrils of Passt- 
fora but refuse to coil around them, and vice-versa. Round the 
tendrils of Smilax prolifera and Vitis discolor, the tendrils of 
Passiflora gracilis, P. trifaciata and Echinocystis lobata twine 
readily, the tendrils of these plants owing to their greater 
stoutness and roughness, affording an adequate stimulus for both 
curvature and coiling. 
Under spevial conditions tendrils of Passifora and Echinocystis 
may coil around other tendrils of the same plant. This is the 
case, if one tendril be fully adult and about to make the old 
age coiling, when it is brought into contact with a neighbour- 
ing tendril, and also if one tendril has coiled around a support 
and the other comes into contact with the spirally coiled 
attaching portion. Both of these cases are in nature extremely 
