254 Transactions. — Botany. 



trunk is almost gifted with the power of locomotion " 

 (16, p. 494). I have given the above quotations partly be- 

 cause they differ somewhat from my account and partly 

 because this prostrate habit seems to me a matter requiring 

 further investigation as to its causes. That it may at the 

 present time be frequently caused by the wind no one can 

 deny, but there is no reason on that account why the pros- 

 trate habit should not be, in part at any rate, hereditary. 

 The behaviour of seedlings may furnish a clue. In the m- 

 terior of the forest beneath the dense leafy canopy the air 

 is comparatively still, when without a gale is blowing. Of 

 the quite young seedlings growing in the ground which I 

 examined, and of which only a few leaves were developed, 

 most had their steins prostrate on the ground for more than 

 half their length. In one instance, in an older plant. 30 cm. 

 was erect and 15 cm. was prostrate. On the other hand, a 

 number of young seedlings growing closely together on the 

 trunk of a tussock were orthotropous, but in this case density 

 of growth or their relation to gravity may have caused the 

 difference. At any rate, it is an interesting matter requiring 

 further investigation, for it bears directly on the inheritance of 

 acquired characters. 



The floor of the Olearia lyallii forest consists of coarse 

 peat covered with numerous fallen leaves in various stages 

 of decay. For the most part it is quite bare, owing probably 

 to the incursions of the numerous sea-lions. There is no 

 undergrowth of shrubs, consequently the forest is much more 

 open than is the rata forest. Growing on the forest-floor are 

 many rather distant patches of young Olearia plants, while in 

 other places are Asplenium obtusatum and Lomaria dura. The 

 young plants of 0. lyallii early on develop quite large leaves ; 

 e.g., a plant 30 cm. tall had leaves measuring 13 cm. x 8 cm. 



Ecologically and floristically the formation under con- 

 sideration is related to the 0. chathamica formation of Chat- 

 ham Island, the 0. colensoi formation of Stewart Island, and 

 the 0. 02}eri)ia -\- Senecio rotundifolms formation of the west 

 coast sounds, though all three must be designated scrub, and 

 not forest. Like the above formations, and situated like them 

 in a rain-forest climate, the 0. lyallii formation is distinctly 

 xerophytic ; but the mild moist atmosphere has caused produc- 

 tion of excessive leaf-surface, just as in the rata forest there 

 is great lateral growth of the Metrosidcros. Occupying only a 

 narrow zone on the sheltered side of the island, the 0. lyallii 

 gives place to the rata forest, which certainly seems on that 

 account the better able of the two to withstand the wind, 

 while in its turn on Enderby Island rata gives place on the 

 windward side to Cassinia scrub, and this in its turn to tus- 

 sock. 



