412 
NATURE 
[JUNE 20, 1912 
three currents. When the ice melts it cools the salt 
water, which sinks down by convection. A stream 
of warmer salt water moves in towards the ice, 
giving rise to a horizontal current. The melted ice 
consists of fresh water, which does not mix with the 
salt water on account of the difference of density. 
This fresh water rises around the ice and spreads out 
over the surface. Very soon the ice is seen to be 
surrounded by a layer of fresher water, which tends 
to remain on the surface. As the ice moves the fresh 
water moves with it. Pettersson believes that this 
circulation has an important influence on the currents 
in the sea. 
Icebergs which have been left high and dry on the 
shore by the tide show the action of the melting. 
Bergs which become top-heavy and turn over also 
bear evidence to the underwater current producing 
the melting. The form of the ice shows a deep 
furrow running all around where the melting process 
has proceeded, and this is often the cause of the roll- 
ing over of a berg to find equilibrium in some other 
position. 
Signs of the Proximity of Ice. 
Before ice can be actually seen there is a peculiar 
whiteness observed around the 
berg on a dark night. This is 
about one-eighth of an inch. The interval of time 
between the dipping of the water and the report of 
the reading may be anything from five to ten 
minutes. In the meantime, the ship has sailed some 
miles beyond the point of observation. 
The Recording Micro-thermometer and its Use in 
Locating Ice. 
For many years I have been studying the effect of 
ice on the temperature of the St. Lawrence River. 
I found that the ordinary thermometer was useless in 
finding the small changes in the water, and it was 
only through the use of exceedingly delicate electrical 
instruments that the temperature changes were 
observed. It was my desire to make use of one of 
these sensitive thermometers to test the influence of 
an iceberg on the water temperature. I accordingly 
devised a _ practical form of  electrical-resistance 
thermometer which was capable of recording 
thousandths of a degree of temperature. This instru- 
ment, which I have called the micro-thermometer, 
works on the well-known principle of the electrical- 
resistance thermometer. The thermometer coil is 
called by mariners the ice ‘ blink.” i 
|50° Centigrade 
48° 
It is caused by the reflection of 
the scattered rays of light from 
the sky from the white surface of 
the berg. Thus it is a contrast 
between the black absorbing 
water, which reflects none of the 
light, and the ice, which scatters 
nearly all. It is stated that on a 
clear day over the ice on the 
horizon the sky will be much 
paler or lighter in colour, and may 
be distinguished from that over- 
head. 
During foggy weather ice can 
sometimes be made out on account 
of its darker appearance. In this 
case it is a contrast effect again, 
but this time it is the shadow of 
the berg against the white 
shadowless fog particles. 
Icebergs are sometimes detected 
by the echo from the steam- 
whistle or fog-horn. They are also 
frequently heard for many miles by the noise they 
make in breaking up and falling to pieces. The 
cracking of the ice or the falling of the pieces into 
the sea causes a noise like thunder. 
The absence of swell or waves is sometimes a sign 
of ice or land, and the presence of flocks of birds far 
from land is an indication of ice. The temperature 
of the air usually falls as ice is approached, and 
“mariners describe a peculiar damp cold, as dis- 
tinguished from the cold caused by a change of 
wind. 
Failure of Previous Efforts to make Use of 
Temperature Changes in the Sea. 
Navigators place no reliance on _ temperature 
measurements. As a matter of shipboard routine, 
the temperature of the water is taken, but very little, 
if any, attention is paid to it. The method is to 
dip a canvas bucket over the side and bring up a 
sample of sea water. The quartermaster then inserts 
a good household thermometer in the water, waits 
for a few minutes, and then reports the reading to 
the bridge. The thermometer is usually graduated 
in two-degree intervals, representing a length of stem 
NO. 2225, voL. 89] 
Fic. 1.—Temperature gradient near an iceberg. 
composed or a large-size iron wire, silk covered, 
wound between concentric cylinders of copper. The 
connecting wires pass through a cable to the observ- 
ing room, where a recorder gives the temperature- 
curve and variations on a chart. The relay galvano- 
meter is of special design to be independent of vibra- 
tion, and is exceedingly strong and quite portable. I 
placed this instrument on one of the Canadian 
Government ice-breaking steamers in charge of my 
assistant, Mr. L. V. King, who rendered most valu- 
able and efficient service in helping me in its opera- 
tion. Through the kindness of the Canadian Govern- 
ment a passage was secured for Mr. King in one of 
the hydrographic survey-boats sailing to NHudson’s 
Bay in 1910. One of the thermometers was placed 
over the side of the ship immersed to a depth of 
about 5 ft., and a record of temperature was made 
through the Straits of Belle Isle, along the Labrador 
coast to Hudson’s Bay. Several icebergs were passed 
in the northern journey at a distance of about half a 
mile, and these were recorded on the chart by a rapid 
fall of temperature of from one to two degrees as the 
bergs were approached. It was found as the ship 
drew near the berg that a rise of temperature took 
