CLYDE SEA AKEA. 133 



the surface layer of 5 fathoms is of considerably greater amplitude, and a little earlier in 

 phase than that for Loch Fyne, corresponding more closely with that for the G-areloch 

 (fig. 47, Plate XXXI. ). In 1886 the curve is so flat, and the apparent maximum so late 

 and low, that it seems possible that the true maximum fell between the dates of observa- 

 tion. On the other hand, the rate of heating was less throughout than in other years, and 

 it may be that the exceptionally low minimum temperature of the spring gave rise to the 

 retardation. It is difficult, however, to account for the rate of surface heating being less in 

 Loch Fyne, whereas in subsequent years it was greater. Probably the fact was due to 

 local causes, such as the direction and force of the wind, or to the amount and temperature 

 of the rainfall entering the loch, and it is, unfortunately, impossible to obtain data for 

 these conditions. 



The bottom layer of Loch Goil gave a curve corresponding closely in form and 

 irregularities with that of the layer between 30 and 40 fathoms in Loch Fyne ; but in 

 1886 its range was less, on account of a lower maximum, and the period of the maximum 

 was both in 1886 and 1887 fully a month later than that of water at the same depth in 

 Loch Fyne, occurring as it did about the end of December. The retardation was not 

 so great as for the bottom water of Loch Fyne. but approached it sufficiently to show 

 that the steepness of the sides of a basin has a great deal to do with the temperature 

 regime of its water, and that depth alone does not condition the manner and rate of 

 change. The maximum of the bottom layer was retarded about 180 days after the air 

 maximum in 1886, and 150 days in 1887, comparing with 126 and 120 days for the 

 retardation of the maximum at the same depth (30-40 fathoms) in Loch Fyne. In the 

 two years the rate of sinking of the maximum was very nearly the same ; for the 

 maximum temperature at 20 fathoms it was 60, at 25 fathoms 90, and at 40 

 fathoms 105 days later than at 5 fathoms, while in Loch Fyne the retardation was 20, 

 30, and 60 days respectively ; the average retardation at the bottom being only 105, 

 although the depth is twice as great as in Loch Goil. Thus the isolation of the Loch 

 Goil basin is twice as efficient as that of Loch Fyne. 



Considering the variations in temperature of the mass of water in Loch Goil, it is 

 found that from the minimum of 42° to the maximum of 49°"6 in 1886, the time was 

 198 days, and the rate of heating only o, 038 per day. This was a longer period of 

 heating by 30 days than in Loch Fyne, and by 45 days than in the Gareloch, while the 

 rate of heating was less than half that of the Gareloch, and less by one-quarter than that 

 of Loch Fyne. The following cooling period was 144 days, when a minimum tempera- 

 ture of 43 0, 8 was reached, the daily rate of cooling being 0°*040. This was practically 

 the same rate of cooling as for Loch Fyne, but the time was 12 days less, and 22 days 

 less than the Gareloch. In 1887 the maximum was 53°'0 after 160 days' heating at 

 the rate of 0°*058 per day, nearly as much longer and slower than the Gareloch as it 

 was shorter and quicker than Loch Fyne. The succeeding minimum of 43° '4 was 

 reached by an average cooling of 0°*046 in 210 days, the rate being the same as in Loch 

 Fyne, but the time 26 days longer, although 41 days shorter than in the Gareloch. 



