230 Transactions. — Botany. 



Unfortunate!}', although we waited some days at Port 

 Pegasus, in Stewart Island, for favourahle weather, the sea at 

 the Snares was ahogether too rough to allow a landing to 

 be attempted ; but, with the exception of Macquarie Island, 

 I visited portions of all the other islands. Interesting as a 

 visit in midwinter is, the days are so short, the number of 

 places to be visited by the steamer so many, and the time 

 which can be spared for the excursion as a whole so limited, 

 that in nearly every instance only a quite cursory examination 

 of the plant-formations and of their members was possible. All 

 the work, moreover, under such circumstances had to be done 

 in feverish haste, for in a given number of all-too-short hours, 

 or even minutes in some instances, notes had to be written, 

 perhaps with numbed fingers or in a driving mist, photoiiraphs 

 hastily taken, plants both for drying and cultivation collected, 

 while in many places the country to be traversed was of 

 extreme difficulty. If under such circumstances errors creep 

 into one's work, or if many important observations are 

 neglected, it is perhaps not to be wondered at. In some 

 instances, chiefly with regard to grasses, it has been impos- 

 sible to identify my specimens. On Campbell Island, owing 

 to the density of the scrub. I found it quite impossible to 

 make any headway burdened with a heavy bag of plants. 

 Night was coming on, and to my intense regret I had to throw 

 away nearly the whole of the plants collected on Lyall's 

 Pyramid. Much of the work in this paper is preliminary and 

 provisional, but it may perhaps lead to a more thorough 

 treatment before long of the vegetation of this, the most 

 interesting region in the New Zealand biological area. The 

 plant-formations are the edaphic formations of Schimper, but 

 their limits as defined by me are far from satisfactory. In 

 some few instances I have distinguished the smaller groups in 

 a formation under the term "plant-association," meaning 

 by this a small collection of two or more plants which usually 

 occui- associated together, and which I agree with Kearney* 

 and Ganongf in considering a distinct ecological conception. 



In what follows, the term " New Zealand biological 

 region " includes the Kermadec Islands, the North and South 

 Islands of New Zealand and their adjacent small islands, 

 Stewart Island, the Chatham Islands, and the Southern 

 Islands , while by " New Zealand " is meant the Nortii and 

 South Islands of New Zealand, Stewart Island, and the 

 adjacent small islands. 



• " Report on a Botanical Survey of the Dismal Swamp Region," 

 Contr. from the Nat. Herb., vol. v.. No. 0, p. 359. Washington, 1901. 



t " The Vegetation of the Bay of Fundy Salt and Diked Marshes," 

 Bot. Gaz., vo . xxxvi., p. 301. 1903. 



