Plant Formations.'] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 483 



island by members of the geological and magnetic survey parties. Mr. Smith visited 

 independently Mount Lyall Pyramid. The portions of the island still unvisited by 

 botanists are therefore the south-west peninsula, including Mount Menhir, Mount 

 Paris, and Mount Yvon Villarceau ; Point Courrejolles, in the north-east; and por- 

 tions of the southern and eastern coast-line. Judging from the general uniformity 

 of the vegetation of the island, it seems unlikely, however, that new species or new 

 plant-associations w'U be found in these localities. 



Absence of Trees on the Island. 



The most striking characteristic of the vegetation of Campbell Island is the 

 absence from it of all trees. The average height of the Dracophyllum scrub is perhaps 

 3-4 m. ; and this is the tallest vegetation on the island. The explanation of the 

 absence of forest from the island is not very far to seek. It is clearly not due to the 

 winter cold, for this at sea-level is not greater than that of Christchurch or Ashburton. 

 Forests in many parts of the world are subjected to much greater extremes of tem- 

 perature than those of Campbell Island. One need only refer to the forests of Siberia 

 and Canada, in which the thermometer often falls below zero Fahrenheit ; or, to come 

 to the Southern Hemisphere, to the forests of Patagonia, which grow in the same 

 latitude as Campbell Island, in a climate which, as far as temperature is concerned, 

 is certainly not less severe than that of the locality under consideration. Indeed, 

 the Gaya Lyallii forests that are found at an altitude of from 2,500 ft. to 3,000 ft. 

 along the Southern Alps must often be subjected to heavier snowfalls and lower tem- 

 peratures than are experienced at the lower levels of Campbell Island. Further, if 

 we take the mean annual temperature of the island at 40° F., we find that it cannot 

 account for the absence of forest, for forests are found in the Northern Hemisphere in 

 regions where the mean annual temperature is considerably below this. In certain 

 forest-clad districts of Siberia the mean annual temperature is below 32° F. The 

 absence of trees, therefore, in this case cannot be due to lack of warmth, but is 

 obviously to be attributed to some other cause. 



This clearly cannot be absence of moisture, for the rainfall approximates to 

 50 in. per year. It can scarcely be doubted that the real explanation is to be found 

 in the violent salt-laden westerly winds that sweep over the island. On Auckland 

 Island Dr. Cockayne found definite proof that in many places trees could grow only 

 in lines in the direction of the prevailing wind. One found a little shelter that enabled 

 it to grow, and the next one grew in the lee of the first, and so on, until a row was 

 formed. Now, although we have no definite evidence on the subject, there is every 

 reason to believe that westerly gales on Campbell Island are, on the whole, rather 

 more severe than on the Aucklands. If this is so, we have an obvious reason for the 

 lack of trees on the former island. Wind causes the death of the plant by producing 

 excessive transpiration ; and the ordinary mesophyte forest is unable to withstand 

 the drying effect of a continual gale. Hence, we should expect such shrubs and 

 leaves as do occur on Campbell Island to be xerophytes, or at least small-leaved ; 

 and this is, of course, the case : the Draco fhyllums are distinctly xerophytic, and the 

 Myrsines and Coprosmas are all small-leaved. The large-leaved herbaceous plants 

 seem to require a different explanation ; but in many cases they grow in somewhat 

 sheltered positions, and their large roots and rhizomes must be capable of supplying 

 immense quantities of moisture to their correspondingly large leaves. Tlie presence 



