The temperature observations recorded in this communication were all taken from the yacht “Medusa,” except those in Loch Morar, which were made from a small rowing-boat, but the sameinstruments and the same methods were used as in the other lochs.

All observations beneath the surface were made by means of Messrs Negretti & Zambra's reversing thermometer in the Scottish frame. The readings are published as they were observed, except that the instrumental correction is applied. The readings may, as a rule, be taken as exact to one-tenth of a degree when the sea was smooth, and when the temperature of water and air had a range less than six degrees. Experiments have shown that if a thermometer be reversed in water at 40°·0, and then brought to the temperature of 46°·0, it would change its indication slightly, and would read 40°·1. At first sight it would appear sufficient to subtract 0°·1 from the reading for each 6° of excess of air temperature over that registered by the instrument, and to add similarly in case the air temperature should be lower. This has not been done, because it was believed that, in summer at least, the cooling caused by evaporation from the wet instrument would reduce its temperature very considerably, and probably enough to make no correction necessary.