312 
SAGO, 
ent aspect, owing to the soft, loose soil. The largest of the two 
groups is evidently of considerable age, and is in reality one con* 
nected tree. From a central point six thick roofs spring which 
spread along the ground in different directions to the distance 
of 6 to 8 feet, when they bend upwards and rise into trees. 
From these trunk-like roots numerous rootlets descend into 
the soil, while large branch rocts strike out laterally from which, 
other trees rise. At present there are 4 large stems, and 14 young 
trees of different sizes rising between them and sending up lofty 
leaves, but without any stem. Around the stole of each of 
these a great number of shoots are constantly springing, each 
capable of becoming a tree if it had room. The hollow bases 
of some trunks which have been cut down are seen. The grow¬ 
ing stems are about 16 inches in diameter and covered with 
moss and small ferus. About the middle of the space covered 
by this multiplex tree no shoots rise, the oldest parts of the 
roots appearing to lose their productive power. 
The other tree is younger, the large roots having only ex¬ 
tended 3 or 4 feet from the centre and still preserving their full 
vigour throughout, for while each has a tree rising from its 
extremity, a thick growth of shoots extends along both its sides. 
Forrest states the average produce of a Molucca tree to be 
336 pounds, but Rumphius makes it from 600 to 800 pounds, 
and according to the writer in the Singapore Chronicle, who 
seems to have paid great attention to the subject, good Suma¬ 
tra trees yield from 760 to 950 pounds, and the very worst 475. 
Perhaps therefore 700 pounds may be assumed as an average for 
the Sumatra trees, which at 10 feet apart (the distance stated by 
Forrest and followed by Crawfurd) would give about 300,000 
pounds for the harvest from one acre, and, allowing that the 
harvests are 15 years apart, and not 7 as Forrest assumes, this 
will give an annual average produce of about 20,000 pounds. 
We believe however that 5 or 6 feet is about the average dis¬ 
tance of the large stems in the Sumatra forests. When a plan¬ 
tation has once arrived at maturity there will be a constant har¬ 
vest, because the natural mode of growth secures a continual 
succession of new plants from the time those first planted have 
begun to extend their roots, and this succession can be regula¬ 
ted by the knife in any way the planter desires. The sago 
tree when cut down and the top severed from it, is a cylinder 
about 20 inches in diameter and 15 to 20 feet in height. If 
we assume 20 in. by 15 ft. to be an average size, the contents 
will be nearlly 26 bushels, and allowing one half for woody fibre, 
there will remain 13 bushels of starch, which agrees very close¬ 
ly with ©ur previous calculation, 700 pounds being equivalent to 
bushels. It may give some idea of the enormous rate of this 
