The N.Z. Journal of Science and Technology. 
190 
[Aug . 
which the melted ice was run of! separately by a tight-fitting rubber tube 
into a measuring-jar. End effects were thus avoided. 
Thermometers, T, were placed in tight-fitting brass side tubes, with 
their bulbs on a level with the surface of the outer cylinder. The average 
readings of these thermometers gives T 2 , which was only one or two 
degrees less than air-temperature. The temperature of T x of the inner- 
cylinder was assumed to be 0° C. Thus T 2 — T t = T. 
Tn general, measurements were not commenced until about two hours 
after the ice was first placed in the cylinder, conditions being then practi¬ 
cally steady. The ice melted at a rate of only about 2 grammes per minute, 
or J lb. per hour, so that the cylinder was still practically full of ice at the 
end of a test. At least two samples of each material were tested, and 
the average results are given in the following table. Samples of the same 
material did not differ by more than 2-3 per cent, in their conductivity. 
77 % 79*0 
A =-—-, where m is the mass of ice melted per minute in grammes. 
r 2 = 12 in. ; r± — 6 in. ; log e r 2 /f i = ’693 ; l = 38 cm. 
K = 
Q log r 2 /fi _ m x 79-6 X -693 m w nAOQK 
2ttIT ~ T X 2* X 38 X 60 ~~ T X 
Material. 
T = tempera¬ 
ture of outer 
cylinder. 
Mass of ice 
melted per 
minute = m. 
m/T. 
Conductivity. 
Calories per cm 3 . 
B.Th.U. per 
hour per square 
foot per inch 
thickness per 
degree F. 
Sawdust 
26-5 
1-77 
-0668 
•000255 
0-74 
Coir 
26-6 
1-395 
•0521 
•00020 
0-58 
Charcoal 
26-2 
2-05 
•0782 
•00030 
0-87 
Chaff .. 
26-3 
1-49 
•0567 
•00022 
0-63 
Air 
26-5 . 
4-31 
•1625 
• • 
All the materials used above were sun-dried. It must be remembered, 
however, that the humidity in Samoa is high—about 85 per cent, on the 
average—while the average temperature is also high—about 26° C. Thus 
the materials, particularly the sawdust, will contain a larger percentage 
of moisture than is usual in more temperate climates. This probably 
explains the fact that the value found for sawdust is larger than that 
usually given in tables. It tends to emphasize the primary importance 
of air-tight walls for the space surrounding the ice-chamber and containing 
the insulating-material. 
The sawdust was from American redwood. The coir was prepared by 
hand, and lightly rammed. The charcoal was the ordinary product pre¬ 
pared by the Natives. Practically the same value was obtained with 
pieces up to 1J in. as with a more finely powdered material with pieces 
up to J in. The chaff was imported. 
The results show the undoubted superiority of coir, which is also 
suitable from the point of view of durability, and less from the fact that 
it is less hygroscopic than the other materials. 
The author is indebted to the kindness of Mr. Menzies, of the Crown 
Estates Department, for kindly supplying the ice ; to Mr. Lofley for the 
construction of the apparatus ; and to Mr. A. Tyndall, Chief Engineer, 
Public Works Department, who kindly supplied the materials. 
