62 A YEAR OF SPORT AND NATURAL HISTORY. 



most numerous in places where the fish had no protection from 

 the sun, and in slow-flowing water and well-nigh stagnant pools ; 

 but where the stream had a rapid run, and had good shelter 

 from solar influence, such as the rocks of Dirlot supplied, the 

 normal state of health was maintained." 



Mr. Harper then gives a table of sea and river temperatures 

 taken by him during the months of March, April, May, June, 

 and July, 1886, which tend to show that the ordinary sea 

 temperature is the most healthy for salmon. 



From these figures it would appear that the sea during those 

 five months had a range of 12.8°; from 39.9° in March to 52.7° in 

 July ; while the river ranged from 33.3° in March to 66.6° in July. 

 July was exceptionally cold that year, a greater range would, 

 therefore, have been obtained in an ordinary one. For the 

 month of April the means of both sea and river were nearly 

 identical, being respectively 44° and 44.7°, and during that month 

 it not unfrequently happens that more fish. are killed with the rod 

 on the Thurso than during the whole of the rest of the season. 

 As the river is not then, as a rule, subject to any exceptional floods, 

 and as there are generally more fish in it in May, it would seem 

 that those conditions of temperature account for the angler's 

 success. That from 44° to 48° is the best temperature for the fish 

 is further proved by the fact that when the water stands between 

 those two degrees, old and stale fish take the fly more freely 

 than at any other time throughout the spring. In May, June, 

 and July, the ascending fish find themselves in water 12° or 14° 

 warmer than that they have felt in the sea, which quite 

 accounts for the inferior sport obtained during these months. 

 Again, the same temperatures of sea and river recur in September 



