CLYDE SEA AREA. 33 



for the most part over 50 fathoms in depth, the difference in the mean temperature of 

 the surface and bottom layers of 5 fathoms each exceeded half a degree on three 

 occasions only, and was less than quarter of a degree on eleven occasions. Four times 

 no difference in temperature with depth could be detected. 



Fig. 4, Plate XXIII. , reproduces several of the vertical temperature curves, all of which 

 show a practically homothermic condition. In No. 2 the dotted part represents the surface 

 temperature found off the Mull of Cantyre at 12 h .10 ; below 3 fathoms the curve coin- 

 cides exactly with that found south of Sanda at 10 h .50, showing that a perfectly homo- 

 thermic body of water rilled the whole extent of the Channel. The tide had turned 

 and was beginning to run westward when the " Medusa" was off the Mull of Cantyre, 

 and on the return journey the tidal stream was so strong that the vessel was scarcely 

 able, going 6 knots, to make head against it. Surface temperature observations were 

 taken every few minutes, and the results were curiously irregular. Patches of water at 

 45°*0 were once or twice passed through, but the prevailing temperatures ranged from 

 42° '0 (the minimum) to 42° '3. Great patches of oilily-smooth water were interspersed 

 amongst the generally rippled surface, and on the whole these seemed to be a fraction 

 of a degree cooler than the rippled surface. The smooth patches suggested the idea of 

 up welling water from beneath, possibly of greater density and different surface-tension 

 from the rest, forming calms in the same manner as a film of oil does. The late 

 Professor James Thomson suggested, in referring to this observation, that the accumula- 

 tion of floating objects along the lines of junction of masses of water moving vertically 

 in opposite directions, on account of tidal disturbance, would act as floating break- 

 waters or dampers for the small ripple undulations.* On the same afternoon the 

 temperature between Sanda and Pladda, on the Great Plateau, was found by two 

 different soundings to be 45° '3 on the surface, 43° at 5 fathoms, and 41°*4 at the 

 bottom in 25 fathoms, the low temperatures being obviously due to overflow from 

 the Arran Basin within, in which the whole mass below 15 fathoms was colder than 42°. 

 Whether the isolated warm surface patches in the Channel were due to direct heating 

 by the very strong sunshine or to isolated flakes of surface water from the Plateau, or the 

 narrow beach of Cantyre, whirling seaward, does not appear. It is quite certain, however, 

 that there was nothing of the nature of a continuous warm upper layer. Whether 

 heating or cooling, the water of the Channel changed its temperature homothermically, 

 and the straightness and parallelism of the vertical curves at all seasons, points 

 unmistakably to continual agitation and thorough mixture by the tidal currents. 



The one exception is observation No. 10, on 17th June 1887, when the superficial 

 5-fathom layer averaged 2°*3 warmer than the bottom layer of 5 fathoms, the whole 

 curve having a distinct positive slope. On that occasion the very hot weather appears 

 to have produced a strong effect on the upper layers. This observation was made south 

 of Sanda, the anticyclonic haze making it impossible to reach the more open water. But 

 even in this case the surface water was more than five degrees colder than the surface 



* Miller, " Tlie Clyde from Source to Sea," p. 293. 

 VOL. XXXVIII. PART I. (NO. 1). E 



