CLYDE SEA AREA. 



135 



cool, while those in which depth and isolation retard temperature changes gain and lose 

 heat, on the whole, at the same rate ; or, as in the case of Loch Fyne, cooling is more 

 rapid than heating. The fact that the two years, 1886 and 1887, were very unlike 

 in their thermal relations deprives these averages of any general application in the case 

 of Loch Goil. 



The rate of change of temperature in fractions of a degree per day is shown in 

 Table LIIL, and, together with the same data for Loch Fyne, graphically in fig. 42, Plate 

 XXX. The mean daily change of temperature for the whole mass is o, 043, or a change of 

 one degree in 23 days, as compared with one degree in 1 3| days for the Gareloch, and in 

 25 days for Loch Fyne. For the surface layer of 10 fathoms it is 0° - 061, or one degree in 

 16^ days, compared with 19 for Loch Fyne. From 10 to 20 fathoms the rate of change 

 was the same in Loch Goil and Loch Fyne ; from 20 to 30 fathoms the Loch Fyne change 

 was slightly greater, and from 30 to 40 fathoms both were alike, 30 clays being required 

 for a change of one degree. 



As in all other cases, the descending curve of air-temperature cut the curve of surface- 

 temperature at the maximum, and also cut the curve of mass-temperature at its 

 maximum a month or so later. In each case the ascending air-curve cut the others 

 about 10 or 15 days after their minimum. 



The air was warmer than the surface layer of 5 fathoms for 172 days in 1886, 104 

 in 1887, and 132 in 1888, averaging 136 days for the three periods ; while it was colder 

 than the surface layer for 224 days in 1886-87, and 248 in 1887-88, an average of 236 

 days. The average of the two seasons is 138 days of air warmer, and 236 days of air 

 colder than surface water, or 4^ months to 7 J. This was exactly intermediate between 

 the data for the Gareloch and Loch Fyne, in the former the period of warmer air being- 

 shorter, and in the latter longer than in any other division. 



Table LIV. — Mean Anmial Temperature of Air and Water for Loch Goil. 



Year. 



I. Air. Mean 

 for Area. 



II. Air. Mean 

 for Helensburgh 

 and Callton Mor. 



III. Water. 

 0-5 Fathoms. 



IV. Water. 

 Mass. 



Difference 

 Air II. and 

 Water III. 



Difference 

 Air II. and 

 Water IV. 



1886 

 1887 

 1888 

 Mean 



46-2 

 47-0 

 46-7 

 46-63 



46-0 

 46-4 

 46-9 

 46-43 



47-1 

 48-9 

 47-9 

 47-97 



45-7 

 47-9 

 47-1 

 46-90 



— IT 



-2-5 



— I'O 



-i "54 



+ o- 3 

 -i'5 



— 0'2 



-o"47 



By interpolating probable values for the first three months of 1886 and the last 

 three months of 1888 the curve (fig. 47, Plate XXXI.) gives the means of estimating the 

 average annual temperature of the three years. This is given in Table LIV. 



The average temperature of the surface layer for the whole time of observation was 



