THE KERMADEC ISLANDS. 14] 
are white and pink varieties; the swamp-matipo (Suttonia Coxit), 
a sbrub with pretty mauve fruits; the Chatham gentian (Gentiana 
chathamica) ; two spear-grasses, Cozella Dieffenbachu, a genus con- 
fined to the Chathams, and Aciphylla Travers ; a sedge which is 
also subantarctic South American (Carex Darwini var. urolepis) ; 
and the Chatham mingimingi (Cyathodes robusta), which forms rounded 
bushes on dry ground, and in autumn is covered with white or red 
“berries” (fig. 94). 
Science is indebted to Mr. T. F. Cheeseman, in the first place, 
and, in the second place, to Mr. W. R. B. Oliver, who much more 
recently spent a year in the group, for a considerable amount of 
information regarding the Kermadec Islands, the only truly sub- 
tropical portion of the New Zealand Botanical Region. 
From the subantarctic islands to the subtropical Kermadecs 
(lat. 29° 15’ 8.) is a long step, and yet the dominant tree in the 
latter is also a Metrosideros (M. villosa), a near relation, however, of 
the pohutukawa and not of the southern rata. But with this the 
similarity between the two botanical provinces ends, except that both 
are of volcanic origin; and, generally speaking, there is no more 
outward resemblance between the plant-forms than there is between 
the climates. 
As seen from the sea, there is nothing in the appearance of the 
plant-covering of the Kermadecs to recall the tropics. For instance, 
no feathery coconut-palms fringe the shore, nor is even the mangrove 
present. On the contrary, the rather dull hue of the ordinary New 
Zealand foliage, as seen from a distance, is everywhere manifest. 
Sunday Island (distant from the North Cape of New Zealand 
about 625 miles), the largest of the group, is forest-clad, while 
Macauley Island is almost entirely without arborescent growth owing 
to the depredations of the goats placed there a number of years ago, 
and in its place is a close-cropped pasture of beard-grass (Polypogon 
monspeliensis). The whole group is of volcanic origin, as stated 
above, and the small Curtis Islands are still in the solfatara state, 
with hot mud, boiling pools, and sulphur. 
A certain number of tropical plants have reached the Kermadecs, 
but not to the extent that might be expected. Amongst these are 
one or two ferns (e.g., Nephrolepis exaltata, Asplenium caudatum, A. 
Shuttleworthianum), Cenchrus calyculatus, Ipomaea pes-caprae (which 
forms the well-known plant-formation on so many tropical shores), 
