62 The N.Z. Journal of Science and Technology. [April 
A village famous for its lava-tunnels in an old lava is Tapueleele, situated 
about three miles inland in the Fa’asalele’aga district, on the east side of 
Savaii, at a height of 160 m. (525 ft.) above sea-level. One of these tunnels 
has evidently fallen in for some distance, and forms the bed of a stream, 
which, followed up, ends abruptly in the tunnel. The mouth is occupied 
by a deep pool, through which one must swim before entering the tunnel, 
and I did not explore it. Another is approached by a short deep 
downbreak through over 50 ft. of rock, and is of large dimensions, with a 
height of over 20 ft. in places. The floor is strewn with fallen blocks, and 
the walls and sides are quite irregular in shape. This tunnel is used as a 
nestmg-place by an indigenous swift (peapea), which rests on the walls high 
up, and gives its name to the cave, and by a bat, which lives there in 
hundreds. The floor is covered by a deposit of a sort of guano, composed 
largely of wing-cases and other chitinous remains of insects and of feathers. 
The same swift (peapea) also nests in the short end of the tunnel at Fale- 
mauga. Friedlander mentions that at Tapueleele there are two tunnels, 
one above the other, apparently in the same lava-field. 
There is stated to be a lava-tunnel in Tofua descending to the sea in 
the direction of Apolima, and another on the rock Nulopa, near Manono, 
directed towards Tofua, and it is popularly believed that these are one and 
the same, and that there is an underground passage between the two places. 
As lava-tunnels can exist only within lava-fields, this would necessitate a 
lava-field connecting the two places, and as lava, like water, always descends 
in its progress, a continuous tunnel of the kind is manifestly impossible. 
Presumably the rock underlying the Apolima Strait is the palagonite tuff 
seen in Tafua and Apolima. 
At Safune, on the north side of Savaii, there is a lava-tunnel which 
has a downbreak a short distance from the sea and lies only a few feet 
below the surface. Here on the seaward side there is a pool of fresh water 
which is affected in height by the tides. The tunnel also opens at the 
seaside a few feet below the surface and gives rise to a bathing-pooi there 
which is fresh at low tide. 
Similar bathing-pools amongst the rocks, sometimes not open to the 
sea and sometimes open on the seaward side, are common all around the 
coasts of Savaii and Upolu. In some cases these may mark the outlet of 
a lava-tunnel, as at Safune, but a more general explanation seems necessary 
to explain their abundance. Inland the level of the ground-water may 
be fairly deep, owing to the jointed and cracked nature of little-altered 
lavas, but, of course, the level must rise nearer the surface as sea-level is 
approached, since it must always be higher than sea-level except at the very 
coast-line. In country with little run-off, but a high rainfall, the discharge 
of water by underground percolation to the sea must be very great ; con¬ 
sequently it is not surprising to find numerous and strong springs appearing 
on the coast at about sea-level. 
There is no doubt, however, that lava-tunnels must very frequently 
be occupied by underground streams and carry off a great deal of water 
which otherwise would move seaward in some porous layer underlain by 
impervious rock or as ground-water. The presence of such underground 
streams would tend to depress locally the level of the ground-water, and 
their existence renders problems of underground water-supply more difficult 
than would otherwise be the case. 
A feature of the streams of Upolu, which has been very generally 
remarked upon by writers on Samoa, is that there is generally much more 
