106 Proceeding-'. 



Abstract. 

 The author gives some account of a collection of rocks from the Kermadec Islands, 

 now in the Canterbury JIuseimi. These are, with one exception, andesites of basic 

 affinity and basalts. Fragments of granite frequently occur on the islands, but the 

 rock has not been found in position. This suggests that the island has been built up 

 on a continental area either a little above sea-level or but slightly submerged. Brief 

 reference is made to the geological evidence for the existence of a former Pacific 

 continent, as demanded by various zoologists in order to explain distribution in the 

 south-west Pacific region. The Kermadecs no doubt formed part of this area, whick 

 probabljr broke up in late Cretaceous or early Tertiary times. 



3. " The Vegetation ol the Kermadec Islands," by Eeginald B. Ohver. 



Abstract. 



The jiaper is descriptive of the plant covering of the Kermadec Islands, and gives a list 

 of the species of pteridophytes and spermaphytes inhabiting the group. It is the result 

 of several months' investigation on Sunday Island by the MTiter during the year 1908. 



The main portion of the paper is devoted to the descriptions of the plant -formations 

 of the Kermadec Islands. These are arranged according to the probable order of their 

 evolution, and fall into five groups. 



The coastal formations include rocks, sand-dunes, ngaio scrub, &c. Such cha- 

 racteristic plants as Cofrosma petiolata and Sccevola gracilis, together with some New 

 Zealand and some tropical forms (e.g., Ipomcea pes-caprce, Canavalia ohtusifolia) are 

 inhabitants of these situations. 



The common bulrush [TypJia angustijolia) is the principal swamp-plant. On in- 

 land cliffs Poa polyphyJla, Asplenium ShuUleworthianicm , and a new species of Veronica 

 are usually to be found. 



Forest which owes its existence to a humid atmosphere is the princijial plant- 

 foiination on Sunday Island. It covers the hills from sea-level to their summits, and 

 is only absent from such places as were destroyed by a volcanic eruption in 1870, and a 

 few clearings m.ade by the settlers. 



Under the heading " Young Formations " are included those which have grown 

 since the eruption of 1870, and the landslip of 1604. These will ultimately become forest. 



" Introduced Formations " are certain meadows where the principal jilant is an 

 imported species. There are three described from the Kermadec Islands — bufi'alo- 

 grass and Ageratum conyzoides on Sunday, and beard-grass on IMacaide}" Island. 



The flora of the Kermadec Islands is most fragmentary, and characteristic of oceanic 

 islands where plants are accidentally carried by ocean-ciu-rents and jiossibly other means. 

 The greater number of species have been received from New Zealand, but Polynesian 

 and Norfolk Island forms constitute the largest part of the vegetation. 



The floras of Lord Howe and Norfolk Islands are considered in this connection, 

 with the result that the ■UTiter believes them to be remnants of the larger one which 

 migi-ated from Malaya by this way to New Zealand, together with a number of Aus- 

 tralian forms which have arrived from time to time across the intervening space of ocean. 



The three groups of islands possess oceanic floras, and properly are included in the 

 New Zealand biological region, and together form a subregion for which is projiosed 

 the name " subtropical islands province." 



From the number of kami logs and pieces of Z)' Vrvillcea and other Algse cast up 

 on the shores of Simday Island, it is evident that the strongest and most frequent ocean- 

 currents reaching the group are from the direction of New Zealand, and this, in the 

 wi'iter's opinion, is sufficient to account for the preponderance of New Zealand forms 

 in the flora of the Kermadecs. 



4. " On a Non-flowering New Zealand Species of Ruhus,''' by Dr. L. 



Cockayne. 



A form of Ruhus from Westland is dealt with related to R. parvus, but differing in 

 its leaves being compound, larger, and somewhat difi'erent in colour and serration. The 

 plant has been in cultivation twelve years, grown under many conditions, and yet has 

 never flowered. The author considei's that possibly it is incapable of flowering, and 

 that it originated as a non-flowering species, either by mutation from R. parvvs or as a 

 hybrid between that species and R. au.ttralis. 



5. " List of Lichens and Bryophytes collected in Stewart Island during 



the Botanical Survey of 1908," by Dr. L. Cockayne. 



Fifteen lichens, thirty-six liverworts, and thirty-four mosses are enumerated, a 

 reference being given in each case to the " Handbook of the New Zealand Flora," if the 

 species occurs in that work. The stations of each species are briefly indicated. 



