62 Transactions. 



trees may grow to a fair height, still keeping branches near the 

 ground, and it is on these trees that a more flattened form of 

 foliage sometimes occurs. This form does not, however, differ 

 in any important respect from the preceding leaves, so I will 

 describe only the usual type of mature foliage. 



As a general rule the leaf is triangular in section, the base 

 representing the upper surface. This form is more like the early 

 stages of the awl-shaped leaves. It is interesting to note the 

 bulging -out of the upper surface in certain of the mature leaves, 

 showing that even here the leaves are liable to more or less 

 modification. 



The arrangement of the chlorophyll parenchyma is rather 

 different from that of the preceding leaves. The row of cells 

 round the leaf next to the hypoderm has here become modified, 

 and forms closely packed palisade parenchyma. In the pre- 

 ceding forms the parenchyma round the edge was composed 

 of loose and irregular parenchyma cells. On the lower surface 

 occur only irregularly shaped parenchyma cells ; on the upper 

 surface their place is taken by elongated cells, which are rather 

 irregular. This arrangement is very analogous to that of the 

 youngest awl-shaped leaves, where, however, there was only 

 one row of irregular-shaped parenchyma between the bundle 

 and the lower surface. 



In the vascular bundle we do not find any increase in the 

 number of elements of true xylem ; there is rather a decrease. 

 The transfusion elements are, however, much better developed, 

 forming great groups at the sides of the bundle, and extending 

 round also on the ventral side. It seems as if nearly the whole 

 of the xylem had here become modified into this tissue. 



Remarks on Origin of Transfusion Tissue in Kahilatea. 



It will be as well here to add a few separate remarks on 

 the origin of transfusion tissue, as, owing to the differences in 

 form, this tissue is arranged somewhat differently. The posi- 

 tion in this leaf in no way contradicts what was said concern- 

 ing the origin earlier. Ir» this species, as in the preceding ones, 

 there is hardly any development of centripetal xylem in the 

 younger stages. If there had been any the tracheids would 

 most likely have been preserved as transfusion tracheids in the 

 flattened form of leaf, for increasing facilities of conduction out 

 towards the spurious margins. When transfusion tracheids do 

 occur in the younger stages, they occur more often at the true 

 sides of the bundle, forming transitions outwards, as in the pre- 

 vious species. I have, however, found an occasional tracheid on 

 the ventral side of the wood in young plants about two years old 

 {vide plate) ; while in older plants we see transfusion tracheids 



