CLYDE SEA AB,EA. 81 



this result may be due to the fact that No. 16 was fixed by 12 points, No. 17 by only 9. 

 No. 16 showed a rapid rise from 45°*9 on the surface to 50°"0 at 10 fathoms, a fall increasing- 

 more rapidly as it deepened to 47° at 53 fathoms, and then a gradual fall to 45°*5 at the 

 bottom. No. 17 was practically constant at 49°"8 from surface to 25 fathoms, a possible 

 maximum of 49°"9 being shown at 20 fathoms. Below this depth the curves practically 

 coincided. The upper 25 fathoms of No. 16 gave an irregular parabola with mean tempera- 

 ture 49°"5 ; this is straightened in No. 17 into a vertical line with mean temperature 

 49° - 8 ; the disturbance practically does not extend below 25 fathoms, and shows a definite 

 warmino - as well as mixing in the upper layers. The 5th was calm, while on the 7th 

 a strong north-easterly wind blew down the loch. The curves of the same date for Cuill 

 and Dunderawe may be compared with those for Strachur, and the relation between these 

 is well brought out in Sections XIV. and XV., Plate XIV. 



Observations on 16th and 18th October 1888, curves 26 and 27 (fig. 31, Plate 

 XXVIII. ). — These show a rise of 0° # 2 in two days. The curves were very fully traced 

 out, and are remarkably irregular. The upper part of both for 10 fathoms shows mean 

 temperature of about 49° '8, nearly uniform, then a uniform convex parabola descending 

 to 44°'l at 50 fathoms in No. 26, and a concave parabola, joined at 25 fathoms by 5 

 fathoms of uniform temperature to a regular convex parabola descending to 44°*2 at 50 

 fathoms in No. 27. This shows two areas of considerable warming about 20 fathoms and 

 35 fathoms. On the 16th the observation was made at 15 h 45 in a calm : the temperature 

 of the air was 51°*3, and during the earlier part of the day there had been a very light air 

 from the south-west. October 17th was calm, overcast, and a little cooler, while on the 

 18th the observation was made at 12 h 30, with the air temperature at 51°"2, the weather 

 misty, and a freshening breeze blowing from the south-east. There is nothing in the record 

 of weather to account for the remarkable heating at 30 fathoms ; and were it not that the 

 temperature recorded at 35 fathoms points in the same direction, one would be tempted 

 to look on that at 30 as a misreading of one degree. 



Observations on 16th and 17th September 1886 at Inveraray, Nos. 5 and 6 in Table 

 XXIX. (fig. 32, Plate XXVIII. ). — The data for curve No. 5 are too incomplete to allow of 

 comparison with No. 6, but those of No. 7 for 27th September (see fig. 33) are exception ally 

 complete. Below 50 fathoms the three curves are very much alike, the temperature being- 

 close to 44°. In the ten days elapsing between Nos. 6 and 7 the surface layer showed slight 

 cooling, and the intermediate layers slight warming, but the total change of temperature 

 was only a rise of 0°*1. These two are the most perfect examples found of S-shaped 

 curves, i.e., compound heterothermicity without intermediate maxima or minima. They 

 are compounded of two parabolas, the upper concave, the lower convex to the origin. 

 The limiting form of such a curve would represent the superposition of a mass of uniformly 

 hot water upon a mass of uniformly cold water. The maximum change of temperature 

 with depth took place about 25 or 30 fathoms, where a fall of 3°'4 took place in 

 5 fathoms. 



Observations on 25th and 27th August 1888, Nos. 23 and 24 (fig. 34, Plate XXIX.) 



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



