188 The N.Z. Journal of Science and Technology. [Aug. 
induce in the metallic Morse wires electrical potentials which, although 
not sufficient to interrupt Morse signalling, rendered the adjustment of 
these circuits difficult and sensitive. 
At Wellington the last effects of the aurora occurred at 1.30 a.m. on the 
17th, and very little effect was noticed at any other station except Dunedin 
after this hour. In Otago mild interference was experienced on the morning 
of the 17th, and this finally disappeared during the afternoon of that day. 
As in the case of the wireless signals, there were brief periods during 
which little interference was experienced, and the whole phenomena were 
erratic in the extreme both as regards intensity, duration, and direction of 
disturbances. 
T elephone-lines . 
A rather unusual interference was experienced at certain telephone 
exchanges, such as Nelson, Palmerston North, Featherston, and Pahiatua, 
where there are a number of long-distance earth-working telephone-lines. 
These lines were subject to disturbances similar to those described in 
connection with the Morse telegraph-lines, but the greatest degree of 
interference appears to have occurred at Pahiatua late on the afternoon 
of the 15th. At this station the call indicators on earthed lines were 
operated by the induced voltages for as long a period as two minutes ; and 
here, as at some of the other stations mentioned, lightning-protectors on 
all lines were operated in much the same manner as if the lines had been 
struck by mild lightning-discharges. 
Although Morse circuits operated upon the metallic or loop principle 
were not entirely immune from interference, the speech on metallic telephone 
circuits was unaffected. This, however, was not due to the absence therein 
of foreign voltages, but to the fact that these telephone circuits are 
balanced and are operated upon such a principle that the voltages induced 
in the two wires of the circuit are rendered mutually compensating and 
produce no effect upon the telephones connected therewith. 
In conclusion, it may be stated that the interferences with telegraph, 
telephone, and wireless services experienced during the recent auroral 
display were of greater magnitude and more far-reaching in their effects 
than any of which the New Zealand Post and Telegraph Department has 
record. It is the first time, also, that the effect of the aurora upon wire¬ 
less communications in this, or probably any other, country has been so 
noticeable. 
In addition to the outline given above a considerable amount of specific 
data has been accumulated, which it is hoped will assist in some degree 
towards the elucidation of the nature of the interference to which telegraph 
and telephone circuits are susceptible. 
SOME TROPICAL HEAT-INSULATORS. 
Bv E. Marsden, D.Sc. 
The present note deals with tests of the thermal conductivities of coconut- 
charcoal, coconut-fibre (coir), sawdust, and chaff. The tests were made 
at Apia, Samoa, under the auspices of the Public Works Department, and 
were rendered advisable by the advent of a local cold-storage chamber 
and an extended use of ice-boxes. The materials tested were such as 
