134 Transactions. 



outpost for the tree to spread from. The head of a large tree, therefore, forms 

 an immense complicated system of branches bearing large leaves which are 

 not usually very dense. Thus in its habit this species is similar to the 

 famous banyan of India. 



The bases of the trunks are irregular, as would naturally result from 

 their origin, especially when formed from the union of several descending 

 roots. The bark is reddish-brown, about 10 mm. thick, and when cut dis- 

 charges copiously a thick white latex. Plank buttresses are formed, usually 

 high ones, beginning 1 m. to 2 m. from the ground and extending out for 

 several metres, twisting and branching the way. The roots or continua- 

 tions of these buttresses are often more or less exposed above the ground 

 for some distance. 



Anatomy of leaf from forest, Transit Hill : Upper surface with cuticle. 

 Upper epidermis of three layers of cells, the outer ones small and flat, then 

 a layer of larger squarish cells, followed by an inner layer of large long 

 cells ; no chloroplasts. A few of the inner large epidermal cells are greatly 

 enlarged and contain cystoliths. Palisade parenchyma of three or four rows 

 of small narrow cells densely packed with chloroplasts. Spongy paren- 

 chyma of very small cells with few chloroplasts. Lower epidermis of one 

 layer of small flat cells with cuticle. 



Habitat : Forms the uppermost tier of vegetation, which is patchy, 

 from sea-level to about 150 m. alt., but odd specimens occur in the Erskine 

 Valley up to 400 m. alt. In the forest on the Northern Hills a young tree 

 was noticed in the fork of a Hemicyclia australasica and sending a root to 

 the ground. 



Looking over the island from the Northern Hills one may easily note 

 the distribution of Ficus. It can always be distinguished by the brown 

 colour of the undersides of the leaves, which show in the wind. It occurs 

 right across the island on the lower ground. It is not found quite near to 

 the water's edge : the wind probably determines its limit in this direction. 

 For a similar reason it does not occur high up on the hills. 



It has been recorded that Ficus trees on Lord Howe Island have been 

 killed by scale insects. I have concluded that the distribution of Ficus on 

 the island is determined by wind. On mentioning this to Mr. Kirby he 

 expressed the opinion that it was the wind that killed the Ficus trees when 

 the settlers first cut down the surrounding vegetation. I have little doubt 

 that this explanation is correct — that wind is the primary cause of the 

 death of so many of the large Ficus trees. Scale insects may possibly have 

 assisted as the vitality of the trees was lowered. 



Distribution : Endemic. 



Elatostema reticulatum Weddell var. grande Bentham. 



Elatostema reticulatum Weddell, Ann. Sci. Nat. (4), i, 188, 1854. 

 E. reticulatum var. grande Bentham, Fl. Austr. vi, 184, 1873. 



Becorded : Moore, Lord Howe Id. Official Visit, 25, 1870 (E. nemorosa) ; 

 Bentham, I.e. ; Hemsley, Ann. Bot. x, 251, 1896 ; Maiden, Proc. Linn. 

 Soc. N.S.W. 23, 134, 1898. 



Habitat : In wet places from sea-level to summit of Mount Gower ; 

 floor of cave under dripping water at base, and in watercourse on summit, 

 of Mount Gower. Common in places inaccessible to pigs. 



Distribution : Australia. 



