PROCEEDINGS FOR 1895 XXI 
release, which is mechanical, is apt to take place prematurely. Also, if 
there is much current, the steamer should be free and not anchored, as 
there is then so much stray line that great depths cannot be correctly 
measured. It is thus necessary to use time in good weather, specially 
for this purpose. In the present instance two thermometers were used, 
which were kept in perfect working order. The thermometers were 
checked against each other by duplicate readings at the same depth, and 
were also compared directly with a standard thermometer. Any read- 
ings which there was reason to suspect of inaccuracy are omitted from 
the results given. 
“Summary for Cabot Strait. 
We thus find in Cabot Strait a current running out of the gulf on 
the western side, and into the gulf on the eastern side, while in the middle 
the current is weak and uncertain in direction. The temperature of the 
water is practically the same in both cases, except within eight miles of 
Cape North, where it is appreciably warmer. The depth of water in 
motion appears to be greater on the western side, while on the eastern 
side the width of the flow is greater and the under-current weaker in 
proportion. From a comparison of the under-current with the tempera- 
ture at various depths, it is to be inferred that the movement of the water 
does not extend to a greater depth than perhaps 60 or 80 fathoms 
at the most, and below that depth the water appears to be per- 
fectly quiescent. The volume of water leaving the gulf on one side is 
thus balanced by the volume entering on the other; and as the tempera- 
tures are nearly the same, the loss or gain of heat to the gulf is much less 
marked than it would be if the balance of volume lay between a surface 
and a bottom current, The actual balance, however, is on the side of loss 
of heat, as it is the outflowing water near the western side which has the 
highest temperature, 
‘As to causes, it is not possible to speak very definitely from observa- 
tions so limited in time and extent. It is more than likely that the layer 
of very cold water between 20 and 50 fathoms is the result of the chilling 
of the water during the winter, and that’ the warmer water at the sur- 
face is duc to rise in temperature with the progress of the season. But 
the reason that the coldest water does not sink to the bottom is by no 
means clear. If fresh water were in question, the temperature of 39° or 
40° would then correspond with maximum density, but the density of 
sea water increases uniformly as the temperature falls. The increase, 
however, 1s exceedingly slight for the range from 41° to 32°, which we 
have here to deal with ; and this allows a possible explanation to be sug- 
gested. It would require the admixture of less than 14 per cent of fresh 
water with sea water at 32° to give it the same density as unmixed sea 
water at 41°. The cold water might thus be prevented from sinking if 
