of the Fishery Board for Scotland. 



413 



mean being 7°*8, that of the afternoon being 8° '2. It was hottest in 

 August (9 a.m. 12°-3, 3 p.m. 12°4), and coldest in March (9 a.m. 3° -8, 3 

 p.m. 3°'6) — a range of 8 0, 5-8 0, 8. The highest monthly mean was a 

 degree lower than in 1891, but the minima cannot be compared, since 

 observations at Dunbar were not made until the latter half of that year. 



The extreme range was 12° '4, the maximum of 15 0, 4 occurring at 

 3 p.m. on June 8, and the minimum of 3 o, 0 at 10 a.m. on March 11. 



The curve obtained by plotting Dunbar observations has the same 

 characteristics as those of the other stations, and closely follows that of 

 the air temperature. The arrest of the fall in November and again at 

 the end of 1892 is seen here very clearly. 



5. OBSERVATIONS ON THE WEST COAST, 1892. 

 Ardrishaig. 



Observations were made regularly at Ardrishaig except during the 

 months of July, August, and September, when they were very irregular. 

 No readings were taken from September 8 until October 2. The ther- 

 mometer, which is graduated to fifths of a degree Centigrade, was read to 

 fifths only. 



The mean annual temperature was 8° -8 in the morning and 9 o, 0 in the 

 afternoon — about half a degree lower than in 1891. 



The maximum mean monthly temperature was that of August — 12°*3 

 at 9 a.m. and 12° 5 at 3 p.m. The minimum monthly mean was 5 o, 0 at 

 9 a.m. and 5°'5 at 3 p.m. in March — a mean range of 7°*3 in the morning 

 and 7 o- 0 in the afternoon. The range is a degree less than in 1891. 

 The surface water does not vary so much in the afternoon as in the 

 morning, owing mainly to its greater warmth at 3 p.m. than at 9 a.m. in 

 the early months of the year. 



The extremes show a much greater range (13°'7), for the maximum 

 temperature was 14°'6 at 3 p.m. on August 23, and the minimum 0 o, 9 on 

 more than half a dozen occasions — in January (18, 21. 22), February (22), 

 and March (10, 11, and 14). 



The curve of surface temperature is more regular than that of 1891. 

 The minimum at the end of 1891 gave place to a maximum at the end 

 of January 1892, and then the curve turned down to another minimum, 

 which was reached in March. The upward motion of the curve consists 

 of several leaps and rests, the two practically equal maxima on the ten- 

 clay means being reached in July and again at the end of August. These 

 maxima of ten-day means are only 12 0, 9 on July 11-20, 13°'l August 

 21-31, compared with 15°-6 for June 21-30, and 13°"2 August 21-31 in 

 1891. A steady fall, broken here and there by slight rises, characterises 

 the rest of the curve, the most marked elevation occuring in November. A 

 minimum occurs at the beginning of December, and then the curve 

 rises 2 degrees, to fall again through the same distance during the last 

 period of the month. 



The line of mean annual air temperature crosses the curve of the ten-day 

 surface means about the end of April and the middle of October. The 

 mean annual surface temperature gives a line cutting the curve of ten-day 

 means at the beginning of May, at the end of the third week of October, 

 when there is a minimum, and again in the middle of November. 



Brodick. 



The record of observations made at Brodick is complete. The mean 

 surface temperature for the year was 8°*7 in the morning and 9° "l in the 



