100 DR HUGH ROBERT MILL ON THE 



maximum in 1886, and 185 days in 1887. The minimum similarly occurred about six 

 months later than the minimum on the surface. The curves show considerable 

 irregularity, particularly the bottom curve, for successive years ; but the general facts of 

 retardation of phase and reduction of range with depth are clearly brought out. The re- 

 tardation of the annual maximum with depth is expressed graphically in fig. 42, Plate XXX, 

 where depths are ordinates, and the time, in days after the occurrence of the maximum 

 annual temperature at 5 fathoms, are abscissas. In 1886 the bottom was 120 days behind 

 the surface layer, in 1887 only 95 days. In 1886 the phase was simultaneous at 5 and 15 

 fathoms, in 1887 tliirt}^ days were required for the maximum to sink to the latter depth. 

 But in 1887 ten days more saw the maximum at 25 fathoms, while in 1886 twenty days 

 were necessary. From 35 to 45 fathoms the descent of the maximum was at the same 

 rate in both years, and 14 days more brought it to the bottom in 1887, while 48 days 

 were necessary in 1886. The curves at all depths are, as in all cases, widely spread in 

 heating up, and drawn much closer together while cooling down. The approximation is 

 more apparent for curves of middle and bottom temperature, the surface remaining 

 markedly higher during heating and lower during cooling, probably on account of the 

 much lower salinity of the surface layer. The variations of the temperature of the 

 water of the loch as a whole are much more regular than those of the separate layers, 

 and admit of comparison with those of the Channel and Arran Basin, with which they 

 are compared in Table LXIV. in the General Summary. 



Starting from a minimum of 42°, say on April 15th, 1886, the mass of water came 

 to its maximum of 49° - 9 on September 30th, a gain of 7°"9 in 168 days, or an average 

 of 0°"047 per day. This time of heating is practically the same as that for the Arran 

 Basin, but the rise of temperature was 1°*7 less, or 5°*1 less than was gained by 

 the Channel water, starting from the same minimum in 20 days less time. The daily 

 gain was one-fifth less than in the Arran Basin, and practically only one-half of that in 

 the Channel. The minimum (probably 43°"8) was reached on March 5th, 1887, 

 showing a total loss of 6°'l in 156 days, or of 0°*039 per day on the average. Here 

 also the period of cooling was practically the same as that in the Arran Basin, although 

 the minimum was slightly higher, and the daily fall of temperature one-quarter less. 

 The next maximum occurred on September 27th, 1887, when the temperature was 51 0, 5, 

 showing a gain of 7°7 in 206 days, at the rate of 0°-037 per day. Here the rate of 

 heating bore the same relation as before to that in the Channel and Arran Basin, but 

 the period of heating was proportionally longer. A minimum of 43° "2 on March 30th 

 showed the loss of 8°"3 in 184 days, or o, 045 per day. This rate of cooling was almost 

 as rapid as in the Arran Basin, and its duration 15 days less. The last maximum 

 observed (48° # 7 on September 15th, 1888) showed a gain of 5°*5 in 170 days, or at the 

 rate of o, 033 per day. The mean duration of heating in Loch Fyne for the three years 

 observed was 180 days, and for cooling in the two years 170 days. For the two years in 

 which the observations for the three years are comparable the ratio of the period of 

 heating to that of cooling was 100 : 115 in the Channel, 100 : 100 in the Arran Basin 



