60 Transactions. 



one on each side of the stem, is necessary so that advantage 

 may be taken of the increased surface. The young lateral 

 branchlets, with the flattened leaves ranged down each side, 

 present somewhat the form of a pinnate leaf. The stem is 

 very slender, and the leaves towards the apex become smaller, 

 the apex itself being occupied by imperfect small leaves. As a 

 general rule these young lateral branches are of limited growth. 



If the flattening had been towards what corresponds to the 

 margin of a flat leaf, the appearance in transverse section 

 would have been just that of a reduced totara-leaf. The bundle 

 would then have occupied a central position, slightly nearer the 

 lower surface than the upper. The protoxylem would have 

 been turned towards an upper flat surface, the resin-canals 

 towards a lower, while at each side of the bundle, towards the 

 margins, would have extended similar elongated elements to 

 those of totara. The actual flattening has, however, taken 

 place in the opposite direction, so that each apparent upper 

 and lower surface of the leaf consists half of the mor- 

 phologically lower surface and half of the morphologically 

 upper surface. In other words, the median line of the 

 dorsal and ventral surfaces has become in each case a margin. 

 This makes the protoxylem face one of the margins, but at the 

 same time it is opposite the upper surface, while the resin-canal 

 has a position similar with regard to the lower surface. 



It may be noted again that the position of the whole bundle, 

 including the resin-canal, remains nearer one margin than the 

 other — that is, nearer the lower than the upper surface. 



The dorsi-ventral arrangement may have taken plac 1 simul- 

 taneously with the flattening. If this did not happen so, and 

 the flattened leaves still remained in spiral arrangement -on the 

 branch, the effect would be rather to decrease than to increase 

 the surface for assimilation. The leaves would then present 

 their margins to the sun, as is the case in many species of Euca- 

 lyptus. 



The plant seems to have gone to an unnecessary amount of 

 trouble to insure the flattened form and dorsi-ventral arrange- 

 ment, but it is impossible to know all the factors at work in 

 producing this result. Perhaps it is to the advantage of the 

 plant in assimilation and transmission of food to have a part 

 of both wood and phloem in direct communication with each 

 flat surface. The arrangement of the leaves in the bud may 

 be one factor in producing the flattened form. I have not yet 

 followed out all the details of the development in the young 

 seedlings. 



Having now found out how the flattening has taken place, 

 and which form is the more primitive, it will perhaps be in- 



