186 Dr. Marcet on the specific gravity and temperature 
in which the experiments were made, could not be relied upon 
as to perfect accuracy.* 
It is obvious that these defects in the methods employed, 
though affecting the precision of the results, and rather tending 
to render them less striking, could not in the least degree in- 
validate the general conclusion, that in Davis’s Straits, and 
in Baffin’s Bay, the sea, at great depths, is considerably colder 
than at the surface ; while to the east of Greenland, and in 
rather higher latitudes, the temperature of the ocean follows 
precisely the opposite law. 
These various facts having an obvious and immediate con- 
nexion with the density of water under different temperatures, 
my attention was naturally directed to that circumstance in 
respect to sea water, which had not yet, I believe, been the 
subject of direct investigation. It had been long suspected, 
but was first established by Deluc, and afterwards correctly 
ascertained by Sir Charles Blagden, that water, in cooling 
towards the freezing point, ceases to contract when its tem- 
perature reaches about the 40th degree ; but that, on the con- 
* Captain Ross, who generally used a register-thermometer, might easily have de- 
tected, by a comparative observation, any material error made in ascertaining the tem- 
perature of the mud which he brought up by his apparatus ; and as he appears to 
have occasionally availed himself of that mode of checking his observations, we may 
presume that his results were free from any considerable error. Lieut. Franklin, 
on the other hand, when he could not reach the bottom, and was therefore unable 
to make use of my machine, employed that used by Dr. Irvin g, consisting of a leaden 
cylindrical vessel with two valves ; a convenient apparatus, but which, as I before 
observed, is liable to some inaccuracy. He sometimes also used a corked bottle, 
which he sunk to a great distance from the surface, and by means of which he 
obtained, doubtless, water from considerable depths ; but it was obviously impossible 
to estimate with exactness the precise depth from which this water was procured, or 
the change of temperature which it had undergone in traversing the upper strata. 
