240 THE SALMON FLY. 



of that of the " Black Dog," seems to prove best in the vast majority of 

 cases. The " Jungle Hornet " is also a great favourite with me on small 

 rivers. It is, however, varied according to circumstances, but on the top 

 of a flood chiefly so in the materials used for the tail. For instance, on 

 the Usk I should put a tail of scarlet Ibis, on the Wye one of yellow 

 Macaw, and so on. Grubs with orange bodies, having cheeks of Jungle 

 at each hackle, are also required when the water remains discoloured from 

 mud or road washings. 



In peat water (porter coloured) a blue-over-black or a grey Heron 

 hackle also come into favour, as well as clarets and browns for the body ; 

 but the " Silver Doctor," " Silver Grey," and "Wilkinson " are sometimes 

 preferred. These latter patterns kill best on many rivers in bright water 

 and weather ; but when silver bodies are constantly used on dark, cloudy 

 days, as, for instance, on the Lochy they should never be left at home. 



When the bottom of the river is dirty, and a green slimy growth on 

 the stones sways to and fro, fish fast-running waters. Always reduce the 

 size of your fly when a Salmon rises after your pattern has passed him. 



In streams with plenty of uneven rocks temporarily or permanently 

 located on the bed of the river, I do not often find a rival for "Elsie." 

 Here again the dressing is varied ; light colours being reserved for open 

 situations, darker ones for those that are screened. This remark refers 

 both to the body of the fly and the wings. 



In the afternoon I usually dress flies with gold tinsel ribbing instead 

 of silver, which answers best earlier in the day. 



It may be taken as a general rule that one ought to mount com- 

 paratively large flies in cold weather, both for deep pools and level- 

 bottomed streams. But if a sudden change of temperature takes place, 

 the size must be immediately increased or decreased very considerably. 

 Suppose the wind veers suddenly round to the east, and the thermometer 

 falls much whilst the barometer rises, it is useless to persevere with any 

 fly for more than an hour or so after the change sets in. But during that 

 short period, although the fish will cease to show themselves, the Angler 

 who works hard and mounts any of the more suitable patterns as large as 

 those used in the early Spring may yet be rewarded for his pains. I have 

 worked on this principle myself for many years, and recently succeeded 



