OF THE ISLANDS NEAR SINGAPORE. 
599 
die on being exposed to the atmosphere, and on their death the 
breakers detach the masses from the living stock, to be washed upon 
the shallow reef, when, exposed to the action of the sun and rain, the 
polyps quickly decompose, and leave the calcarious matter to be roll¬ 
ed backwards and forwards by each successive wave, until reduced 
to a sandy detritus, it forms the beautiful beach of the coral girt is¬ 
lands, and by its mechanical effect when washed back by the ebbing 
tide destroys the living corals within its reach, and disposes them to 
undergo a similar transformation. 
Wherever we have coral reefs exposed at ebb tide we have a great 
destruction of coralline polyps, and a decomposition of animal mat¬ 
ter carried on in a gigantic scale, proportioned of course to the size 
of the reef. During this decomposition the same chemical and ani¬ 
mal changes occur, the same gases are generated and germs created 
as in a similar decomposition of animal substances on land. If ma¬ 
laria is produced from animal decomposition on land, and we have a 
similar decomposition at sea, I think I am entitled to make my first 
deduction, that wherever a coral reef is exposed at low water, ani¬ 
mal decomposition will go on to an extent proportioned to the size 
of the reef c<st: par: and that malaria will he the I'esult of this 
decomposition , which is one and the principal cause of the fevers 
endemic in such localities . In proof of this I refer my reader to 
what I have, on personal examination, found at Ayer Banddra, Bla¬ 
kang Mat!, Pulo Sakijang, Pulo Serai and Tanjong Pagar. Certain 
conditions increase or modify the applicability of this deduction. For 
instance, if a reef is extensive and a gentle trade wind in one mon¬ 
soon blows over it and wafts the effluvium, and of a necessity the ma¬ 
laria, on habitations, situated to leeward, and if that wind is prevent¬ 
ed continuing its progress by a hill, or a forest, by which perfect ven¬ 
tilation is prevented, more especially if we have the coral reef situated 
close to the base of the hill, which from its height lulls the wind, so 
that no benefit is received by the inhabitants from the contrary mon¬ 
soon that should carry away all effluvium and malaria, then we 
have the malaria from coral developing itself in its intensest form, 
giving rise to a virulent endemic fever, as we see at Ayer Bandera 
on the island of Blakang Mad, and, as we will see, in other coun¬ 
tries ; if the ventilation however is not altogether impeded, but merely 
imperfect, we have the fever much modified, so as to be changed to 
the intermittent from the remittent, as at k&mpong Blakang Mad. 
From these premises my second deduction is drawn, that from 
