144 Transactions. 



terize it. In certain individual specimens, however, especially in sheltered 

 situations, I noted deep-blue flowers. Here and there near the tracks I 

 noted dull weather-worn patches of Asplenium falcatum, with patches of 

 Muehlenbeckia complexa, Hierochloe redolens, Pteridium esculentum, and Poa 

 foliosa, with the imported grass Holcus lanatus. 



A few specimens of stunted Weinmannia racemosa and Rapanea Urvillei 

 were also noted, along with Dracophyllum longifolium and Coprosma areolata. 



At the eastern end of the island, which is the highest part, there appears 

 to have been a fire some time ago. This apparently destroyed the scrub 

 over an area of from 2 to 3 acres. A most interesting result has followed. 

 The whole of the burnt part is now covered with a close and almost pure 

 association of Poa foliosa, the tussocks being close together and about 

 1 metre in height, making progress a difficult matter. The grass is growing 

 very strongly, and the flowering-heads are exceedingly numerous and of 

 immense size. 



Upon climbing through this grass I discovered numerous young plants 

 of Veronica elliptica and Muehlenbeckia australis, with a few of Rapanea 

 Urvillei and Lepidimn oleraceum var. acutidentatum, while the ground was 

 full of the burrows of the petrels. It would appear, therefore, as time 

 goes on, the original plant association of the island will regain possession, 

 and that the Poa foliosa is merely a temporary occupant of the ground. 



The extraordinary spreading of this grass, which elsewhere on the island 

 is only found in isolated " tussocks," suggests the possibility of its having 

 an economical value, which might be worth investigating. At present 

 practically nothing is known of its value as a fodder plant. 



Conclusion. 



The existence of the plant covering depends in the first place upon the 

 high rainfall, the soft peaty soil being very spongy and damp for a con- 

 siderable depth. On the other hand, there is little doubt that the traffic 

 of the birds materially affects the plant-growth. The ground is everywhere 

 drained and aerated by thousands of tunnels made by the nesting birds, 

 while the soil is richly manured by the constant deposits of these petrels. 

 Most of the plants have large masses of surface roots only, and the loose 

 soil enabled even fairly large specimens to be lifted out of the ground with 

 ease. 



The distribution of the species depends chiefly on the wind factor, the 

 only plants that can get a hold on the steep exposed sides being those 

 specially adapted to resist transpiration, notably Olearia angustifolia, 

 Veronica elliptica, and Poa Astoni. 



The extraordinary spread of Poa foliosa I attribute to its abundant 

 seeding habit, the rich soil, and the unimpeded light gained by the destruc- 

 tion of the scrub by fire. Its wind-resisting qualities are evident from the 

 manner in which it grows on exposed subantarctic habitats. 



