258 SALMON AND TROUT. 



2. The twist or oval should be applied sparingly ; a few turns — say three 

 complete \.\3X'c\% altogether — I find, are better than many. 



3. The hackle should always be turned on nearly upright — or rather, 

 curling very slightly over towards the hook-bend, in the shape and position 

 shown in the two illustrative diagrams. If hackles lie fiat they only hide the 

 body of the fly, and entirely lose the life-like movement which it is their 

 special function to impart to the fly by opening and shutting with its move- 

 ments in the water, somewhat like the motion of a jelly-fish. 



4. For the same reason flies should never be compressed in a book, but 

 kept loose in a box from the first moment of tying. This is also most im- 

 portant, at any rate as regards my patterns of hackle flies. 



5. The length of the hackle should always be very nearly as long as the 

 ength of the fly from where the hackle starts to the extremity of the hook- 

 bend [vide cuts). 



6. In all these and other flies dressed on my eyed hooks, it is really 

 essential that the neck of the hook, between the eye and the head of the fly, 

 should be left clear for the gut, as shown in the diagrams. The length of the 

 hook-shank is, moreover, especially designed to allow of this, and if the fly 

 were dressed right up to the loop, not only would not the attachment to 

 the gut be so perfect or complete, but the body of the fly would in most cases 

 be disproportionately long. 



I have furnished patterns of these flies to Messrs. Farlow, 

 191 Strand, who have been in the habit of making my flies 

 according to instructions for many years. To this course 

 I have been driven by the monstrosities — they can be called 

 by no other name— constantly met with at tackle shops, or 

 shown to me by fly-fishers as ' my flies.' (I refer to my three 

 original patterns of ' typical flies '). Briefly, my collection of 

 them forms a sort of ' chamber of horrors ! ' 



The same cause has led me to substitute, in the present 

 patterns, natural for dyed hackles. I am satisfied that the latter 

 have in some cases, and in certain shades, advantages ; but these 

 are more than compensated for by the uncertainty as to the 

 nature of the dye used by different fly-dressers, and the colours 

 they produce. 



A word as to Stream Trout flies (I do not refer now in any 

 way to dry-fly fishing). These I fancy look somehow best, and 

 realise better the idea of * a poem on a hook,' when dressed 

 on the sneck-bend (see cut), and I would suggest the following 



