NOTES 
On the Occurrence of Bauxite on Truk 
In a paper appearing elsewhere in this issue 
(Pacific Set. 2(3): 216) mention is made of 
the identification of bauxite on Moen Island 
in the Truk group. Although the Japanese 
had probably discovered this material on these 
islands in the course of their comprehensive 
search for mineral deposits throughout the 
former mandated area, the writer knows of no 
published reports on the subject. The purpose 
of this note is to record this occurrence and 
to give a brief description of the deposit. 
The Truk group lies in lat. 7° 25' N., long. 
151° 30' E., and consists of about 15 large 
and small volcanic islands rising out of a 
lagoon about 30 miles in diameter and sur¬ 
rounded by a reef ring. The total area of the 
islands within the lagoon is about 37 square 
miles. 
The only bauxite found to date in the Truk 
group is located on Moen, the second largest 
island in the group (area about 7.3 square 
miles). The deposit appears to be confined 
to a small area of about 200 acres on the 
summit of Mt. Witipon (Takeun on some 
maps), an 890-foot peak which dominates the 
eastern end of Moen (see p. 218, Fig. 4, this 
issue). The material, represented in the col¬ 
lections (see p. 220, this issue) by specimen 
TK-23, consists of reddish-tan, irregularly 
shaped, vesicular nodules, which are from 2 
to 3 inches in diameter, and which superficially 
resemble small, dried sponges. Freshly broken 
surfaces are mottled in shades of red and 
yellow, and have a fine, granular texture. These 
nodules occur in a pale, yellowish-buff, fine¬ 
grained clay, but are most abundant at the 
surface as a result of concentration during the 
erosion of the clay. The clay itself is the 
weathered residue of a fine-grained trachytic 
flow (see p. 219, this issue) and is quite dif¬ 
ferent from the gray and deep-red clays derived 
from the weathering of the basalts which cover 
most of the island. 
Early in 1947 a sample of one of these nodules 
was tested on a portable thermal analysis unit 
(Hendricks, Goldich, and Nelson, Econ. Geol. 
41: 64-75, 1946) and was estimated to con¬ 
tain about 60 per cent gibbsite and 20 per cent 
kaolinite. From this it was estimated that the 
sample contained about 42.5 per cent alumina. 
This was determined from the gibbsite fraction 
alone, and no allowance was made for the 
alumina in the supposed kaolinite. 
Recently a chemical analysis of the same 
nodule was made in the laboratory of the U. S. 
Geological Survey (W. W. Brannock, analyst), 
and the results were as follows: 
Insol. 9.37 
A1 2 0 3 53.08 
Fe 2 0 3 7.26 
Ti0 2 0.66 
Loss on ignition 29.68 
An X-ray determination of the same material 
made in the laboratory of the Geological Sur¬ 
vey, by George Switzer, showed that the pre¬ 
dominant mineral present was gibbsite, mixed 
with some halloysite. Boehmite is not present, 
and kaolinite occurs in very small amounts. A 
re-examination of the curve produced by the 
portable thermal analysis machine shows that 
it carries a low peak at 140° and a strong peak 
at 610°. These peaks agree closely with those 
in the curve for halloysite given by Hendricks, 
Goldich, and Nelson ( op. cit., Fig. 5) and indi¬ 
cate that the second mineral should have been 
interpreted as halloysite. Twenty per cent of 
halloysite would add about 7.9 per cent of 
A1 2 O s and 9.3 per cent of Si0 2 to the sample, 
and, considering the very rough nature of the 
estimate, the new totals agree very well with 
the analysis. 
Using the standard assumption that each 
unit of per cent of Si0 2 makes 1.1 per cent of 
alumina unavailable in the Bayer process, the 
available A1 2 0 3 of the sample is reduced to 
42.78 per cent and is therefore not of com¬ 
mercial grade. The low iron content of this 
bauxite is noteworthy, especially when com¬ 
pared with analyses of bauxites from other 
islands in the Pacific, notably Bintan in the 
Dutch East Indies and Palau in the western 
[223] 
