58 DR HUGH ROBERT MILL ON THE 



The importance of the Skate Island observations is exceptional, for several 

 reasons. Being the deepest water in the Clyde Sea Area, it is the part of the Arran 

 Basin most completely cut off from the sea, and consequently the characteristic thermal 

 changes of the Basin should here be most readily detected. The station was 

 also important because, from some happily situated landmarks, the exact spot where 

 observations were made could be readily picked up at any time, and any tendency to 

 drift while observing could be at once detected and checked. A set of observations had 

 also been taken by Mr J. Y. Buchanan in the summers of 1878, 1879, and 1885, thus 

 allowing of some interesting comparisons. 



The most striking feature of the temperature curves is that, with scarcely any 

 exceptions, the lower 50 or 60 fathoms is straight and perpendicular, showing a 

 predominating homothermic state in the deeper water. 



There were positive slopes on twenty-three occasions, zero slope once, and negative 

 slopes only ten times. Compared with the Garroch Head observations, this shows that 

 negative slopes are more difficult to establish in the deeper water. Otherwise the two 

 stations are very similar. The maximum positive slope observed, 9° '4, is very nearly the 

 same as for Garroch Head. The maximum negative slope was 2° '9, and only on this one 

 occasion was the negative slope greater than 1°*8, while the positive slope was only twice 

 less than that figure. Positive slopes over 4° "9 occurred fourteen times, — twice in June, 

 thrice in July, five times in August, and four times in September. Negative slopes 

 were only found in the months from November to March ; that is, in the months of 

 rapid cooling and of minimum temperature. Positive slopes less than 4°*9 were shown 

 three times in April, once in May, three times in June, and once in October ; that is, at 

 the commencement of the periods of heating and cooling respectively. 



The rate of gain or loss of heat in the vertical section in a short interval of time 

 could be studied on four occasions, when observations were made on successive days, or 

 with an interval of a few days. Fig. 18, Plate XXV., shows curves 17 and 18 of Table XIX. 

 representing the state of matters on the 15th and 16th June 1887. The observation of 

 the 15th (curve A) was made at 12 o'clock, when the air-temperature was 57°*0, a 

 light breeze was blowing from the north-west, and the tide was 4f hours' ebb. On the 

 following day the observation was made at 15 h 30, when there was a very light southerly 

 breeze with air-temperature 63° '0, and the tide about half an hour's flood, practically at 

 low water. The change of temperature does not seem to be due to the movement of 

 warmer water, for the up welling off Otter effectually isolates the surface layers of the 

 Arran Basin from those of Loch Fyne, and, with the ebb tide, the effect of the warmer 

 water in the seaward division could hardly be felt, although the fact of a distinct increase 

 in surface-salinity on the second day hints that this might have been the case. The 

 gain of heat, averaging nearly a degree for the first 20 fathoms (below which there was 

 practically no change), may thus be taken to represent the heating effect of 21 hours' solar 

 radiation, minus the cooling due to G|- hours of darkness ; and supposing the average rate 

 of gaining heat in sunshine to be the same as that of losing it at night, the result would 



