18 
Transactions of the Royal Society of South Africa. 
behaviour of the stomata, both on the tree and under the experimental 
conditions. 
Attempts were made to obtain this information by Lloyd's method of 
fixing strips of the epidermis in absolute alcohol, and measuring the size of 
the stomata and the degree of opening with a micrometer eyepiece. The 
method, however, proved unsatisfactory, and neither type of leaf is suitable 
for observations with the porometer. 
It is evident, therefore, that any interpretation of the experimental 
results with reference to the behaviour of whole trees must necessarily be 
highly speculative. One possible interpretation is that the Silver Tree keeps 
its stomata only partially open ; that it has relatively immobile stomata and 
that if the conditions become too severe, and the tension exceeds a limited 
value, the leaves dry back, as in the cut twig. In the Pine, on the other 
hand, the stomata are more mobile, open wider under favourable conditions, 
and close more effectively in dry. Another possible interpretation is that 
there may be a greater power of internal regulation in Pine leaves. That is 
there may be some way other than stomatal closure, or additional to it, of 
reducing water loss, probably connected with lowered water content. Apart 
from these purely speculative suggestions all that can definitely be said at 
present is that the Pine leaves appear to be more resistant to unfavourable 
conditions than the Silver Tree leaves. 
Further research is needed to determine the effect of cutting on the 
stomata and internal condition of the leaves. In this connection transpira- 
tion measurements in the field are needed, (a) on twigs still attached to the 
trees, and (b) on twigs immediately after detachment. 
Another line of research is the measurement of the actual tensions 
existing in trees, and hence of the resistances to be overcome by leaves 
transpiring on the tree. 
V. Summary and Conclusions. 
1. Farmer's experiments on the water-conducting power of the wood of 
trees have been extended to certain trees growing in the Cape Peninsula, 
and the results obtained are in general accordance with his conclusions. 
2. Pinus pinaster has wood of low water-conducting power, its specific 
conductivity being a little higher than that recorded by Farmer for 
P. sylvestris. 
3. Leucadendron arqenteum has wood of high water-conducting power 
for an evergreen tree, its specific conductivity being rather more than 
double that of the Pine. 
4. A method of estimating the surface area of Pine leaves is described. 
5. Pine twigs, kept with their cut ends in water, show a far higher rate 
of transpiration, both per twig and per unit leaf area, than do similar twigs 
of the Silver Tree. 
