192 SUBANTARCTIC ISLAND.S OF NEW ZEALAND. [Ecological Bnlomj. 



4. Special Ecology of the Plants. 



(A.) GENERAL. 



So far as tlie various vegetative features of the New Zealand subantarctic 

 province go, there is not one, unless we except the great leaves of PleurophijUuni, 

 which is not to be found in one or other of the floristic provinces. The most 

 striking peculiarities — the peat-forming and -utilising habit, prostrate tree-trunks, 

 dense foliage of the forest-roof, divaricatingly branched shrubs, cushion and rosette 

 plants — all these are to be met with in the mountains of New Zealand. In fact, as 

 Drude has pointed out, but from a study of floristic rather than ecological relation- 

 ships, the higli mountain flora, and, to add to his statement, vegetation, is sul?- 

 antarctic, and distinct from the ordinary forest vegetation of New Zealand.* 



To the same category belong many of the lowland bog formations of the southern 

 floristic province of New Zealand. | This subantarctic vegetation is in many respects 

 a reflection of the climate, and wherever that climate exists, no matter what the 

 altitude, subantarctic life-forms occur. The low summer temperature is antagonistic 

 to the unmodified subtropical vegetation ; the wet climate, which otherwise would be 

 favourable, is opposed by the wind factor, which, together with the peaty, sour soil, 

 favours xerophytic structure. 



(B.) LIFE-FORMS. 

 (a.) Trees. 



The plants whicli may finally possess a trunk, and become tall enough to be 

 designated trees, are : Metrosideros lucida (Myrtaceae) ; Olearin Lyallii, Senecio 

 Steivartiae {Compositae) ; Nothopanax simplex (Araliaceae) ; DrncophyUum. longifolium 

 {Epacridaceae). None of these, as shown further on, occur exclusively as trees, 

 all varying excessively according to their environment. 



The most important " adaptation " is the semi-prostrate habit of growth, 

 shown especially in M. lucida, 0. Lyallii, and »S'. Steivartiae. The trunks of these, 

 equalling in every way those of erect trees, generally lie for more than half their 

 length either prone upon the ground or but slightly raised above it, and horizontal 

 in direction. Such trunks are by no means straight, but usually curved, arching 

 or twisted in an irregular manner. In the case of 0. Lyallii the trunks may be 

 from 7 m. to 9 m. in length, and 50 cm. or more in diameter near the base ; those 

 of S. Stewartiae are similar ; and those of M. lucida are generally longer, thicker, 

 and more irregular in shape, arch to a greater degree, and extend more or less 

 horizontally to the surface of the ground perhaps even more frequently than they 

 lie prostrate (fig, 4). 



The prostrate trunks of 0. Lyallii, and probably of *S. Stewartiae, are firmly 

 anchored to the ground by means of adventitious roots, any part touching the sub- 

 stratum being able to put forth roots. In the case of Metrosideros lucida, roots are 

 also produced from the trunk, but these do not function as in the above case. Such 



*" Manuel dc Geograpliie botanique," p. 504; 1897. 



fSee Cockayne, "Report on a Botanical Survey of Stewart Island " ; 1909. 



