134 MISS L. 8. GIBBS: A CONTRIBUTION TO 
Levu in 1901 to investigate this question, divides the geological formations 
of the island into two main groups : the first including continental rocks of 
high but undetermined geological age ; the second, Tertiary to recent forma- 
tions of voleanie and sedimentary origin. Dr. Woolnough considers that 
Viti Levu consists of a core of very ancient, perhaps archaic rock, surrounded 
and partially covered by marine deposits of Tertiary and recent age, and 
Cainozoie lavas; but so far no traces of Palæozoic or Mesozoic formations 
have been observed. Не accounts for the absence of these formations by a 
prolonged subsidence since very early geological time. He admits that there 
is no direct evidence up to the present to prove a continental origin for Viti 
Levu, and he states that the rocks collected by him show no marked similarity 
with either New Caledonia or New Zealand, the nearest undoubted con- 
tinental areas. Also rocks characteristic of continental areas have nowhere 
been met with so far from land-masses. The great depth and extent of the 
ocean between Fiji and the nearest areas of continental land is also an 
argument against such origin. The latter difficulty is not considered 
insuperable, and the case of Madagascar is cited amongst others—the 
Mozambique Channel, due to extensive faulting, being quite as deep, though 
not so broad, as the sea which separates New Caledonia and Fiji. Other 
arguments are brought forward to prove the possibility of such continental 
origin. 
Professor Woolnough notes the interesting fact that evidence afforded by 
land-shells is more towards affinity with the New Hebrides to the west, than 
with Samoa and Tonga on the east. 
The botanical evidence, as far as it goes, shows no such preponderating 
influence. The flora is markedly Indo-Malayan, which the low altitude of 
the mountain ranges and the small area of open country practically restrict 
to forest types. In mosses and liverworts the affinity would apparently lie 
more with Samoa, but these plants have been systematically collected in 
Fiji and Samoa, which, as far as I know, has not been the case in other 
Pacific areas. 
That it is inadvisable in the present imperfect state of our knowledge of 
the flora of these groups, to draw any far-reaching conclusions as to the 
distribution or endemism of particular genera or species, is proved by 
the results of the present collection. It was made in one locality, within a 
radius of 20 miles, and an altitudinal range of only 1800 feet, starting 
from 2700 ; yet it iucludes 40 new species and 7 new records, comprising 
7 genera not previously recorded for the islands. 
Of the new records the most interesting is Melicytus ramiflorus, widely spread 
in New Zealand, and also known from the Kermadec and Norfolk Islands and 
Tonga.  Cyrtandra glabrata has previously only been recorded from Tahiti. 
Podocarpus elatus, growing as a slender tree in the forest, and one of the 
chief denizens of the mountain ridges about Nadarivatu, is known from 
