Eye of the Beholder - The Search for the Perfect Fly

I haven't written anything recently because 7 rainless weeks have reduced the Ure to a stream and made fishing fruitless.  Fortunately (from my perspective - my wife disagrees) Yorkshire has been much cooler than southern England - today Wensleydale is 12 degrees cooler than Essex - so to date we've avoided serious depletion of the oxygen that would otherwise threaten the dozing salmon.  There are plenty of fish in the river following an excellent May spring run, but they are now semi-comatose residents unlikely to respond to a fly except during short periods in the early morning and late evening. Under these conditions I prefer to leave them in peace and await the arrival of the early autumn runners.  As the insect life on the Rye has gone on strike, my garden is in unusually good order: it's a very good rose year.

In parallel with this inactivity, the Islamic holy month of Ramadan (28th June - 28th July) has greatly reduced my Gulf clients' work-rate and thus given me ample thinking time.  Amongst other things I decided that I needed to do more on salmon vision and visual stimulation. My March 2013 post - 'Here's Looking at You' - explained how their eyes work and the effects of attenuation on what they might see.  From that I offered some general thoughts on the relevance of those factors to fly design and selection.  Put simply, in the dim and murky world of of an autumnal spate river in mid afternoon, the black component of the fly stands out better and is therefore more likely to catch the salmon's attention than the brighter colours.  In brighter conditions, dull grey or brown tones deny the salmon a clear enough image to see that the fly is something alien.  But neither of those points, valuable as they may be, addressed the inevitable challenge of "so what?"  The discussion was incomplete.

The "too good look"

Over the past month I have been enjoying Topher Browne's Atlantic Salmon Magic, a birthday present from my very generous wife.  That said, she is a Yorkshire lass with a keen eye for value, who would have baulked at the original cover price of £100, but found a 50% price cut irresistible.  There's vast experience and common sense sprinkled throughout its immense volume, in between stunning photographs and backwoods homilies.  Two nights ago one line caught my imagination, in words to the effect that "you need the salmon to notice the fly, but not to get too good a look at it".  In other words, in design, tying and presentation, we must remember that the fly's purpose is stimulation, not imitation.  Indeed, my own experiments involving the observation of salmon reaction (or more often not) to precise representations of food species, suggests that imitation may be an unproductive blind alley.  So the fly needs to trigger embedded responses, but need not be the thing itself.  

Of course we're only guessing and there are some big variables in play:

• We don't know the trigger mechanism and the largest variable is whether the salmon is in the mood to be stimulated and to what degree.
• Then there is presentation and the unanswerable debate between the 'right' fly presented badly and the 'wrong' one fished well.
• There's visibility, which varies from river to river, level to level, day to day and hour to hour, so the 'perfect' fly may only be perfect - or rather good enough - for relatively short periods.
• We don't know how the salmon perceives colours.
• And of course, superimposed throughout, is chance - Falkus' 'outrageous fluke'.  

Fine Distinction

With all these variables, can we know the merit of fine distinctions between patterns?  If nothing else, good salmon flies stimulate fishermen!  My grandfather and father in his turn held that most salmon flies were tied to catch anglers, not fish.  A well tied fly is aesthetically pleasing and often beautiful in the eye of the beholder when viewed in crystal white light.  But the acid test is how it appears when viewed at the taking range of 2m/6' while immersed 30cm/12" deep, tracking through the moving water of a spate river running at +30cm/12" at 3pm on a dull September day.

Selection of shrimp patterns
Images by Fulling Mill on Sportfish website

Here we have a selection of 6 very popular shrimp double patterns.  The differences between them are readily discernible in white light (albeit there's a small white balance problem with No 6).  Of course, there's nothing in the natural environment of these orange/red shades, so there's no imitation here.

Photoshop simulation, not size corrected for distance

Using Photoshop, this picture simulates the brightness and definition of the same flies, in the test conditions described above. You may agree that the differences between them are becoming less readily detectable.  However, as the contrast falls the definition of the black elements is relatively stronger than in white light.  

Photoshop simulation of sinking head presentation
1" Cascade Conehead, depth 1m, range 2m, poor light
Not size corrected for range

Moreover, as depth or range increases, or ambient light declines with the season and conditions, the red component of the light reflected by the fly diminishes rapidly (about 24% per metre).  In those circumstances the orange dressing takes on an an increasingly grey appearance, whilst the definition of the black remains significant. Of course a salmon capable of catching grey sprats in a grey ocean at high latitudes has no problem detecting such objects, provided that they are moving, and for optimum performance, above the horizontal sight-line (i.e. in Window 2 or 3).

Orange, grey or downright dull?

Given that all the patterns in the sample are successful, and all display about the same sort of colour in the specified underwater conditions, then we a heading towards a conclusion that a fly that appears grey with some contrasting black seems to work well in autumn conditions.  This is consistent with the experience-based ghillie aphorism for autumn fishing of turning your fly with the leaves, towards patterns with a predominance of red or orange.  Taking that argument a stage further, let's look at two closely related and very popular patterns.

The Cascade is almost the universal fly for dirty water and anything other than bright conditions.  If I was told that I could only fish with one pattern for a whole season (or life), this would be it (albeit I'd cut back some of the dressing in clear water). But is its success owed to its orange component, the black, or the balance between them?

Its first cousin, the Ally Shrimp, has much less black in the dressing and a red body; and less mass and length in the tail (even after allowing for the smaller size #10 vs #8) .  I find it works better than the Cascade in brighter conditions.  This one certainly did, as you can observe from the damage to the hackle, inflicted by 2 good fish on a bright, warm September day.

Ally Shrimp & Cascade
Photoshop simulation of clear water & bright sky
Depth 1'  Range 3'

Now put them into relatively clear spate river water, at mid-morning light on a bright day, at a depth of about 1'/30cm, and examine the salmon's eye view from a range of 3'/1m as it approaches to take.



This is a very different prospect to the dull September afternoon with the river at +12".  The level of absorption of the red light is greatly reduced, so the orange elements stand out.  The Cascade has very strong definition and contrast.  Does this allow the salmon to get "too good a look"?  In comparison the Ally Shrimp appears more bland and anonymous.  Could it be that its dullness is the key to its superior performance when fished shallow in brighter conditions?

Low clear water flies

Thunder & Lightning, Hairy Mary, Silver Stoat, Stoat's Tail
Clear day, depth 1'

Putting a selection of four popular low water flies together in the same simulated conditions, we observe an almost complete absence of red/orange tones.  However, at this sizing there is strong black definition of the outline.  They are popular because they are successful.  Is there are connection between their success and the absence of red/orange?

If we re-scale my favourite Blue Charm to represent the smaller #12-14 fly size I tend to use in such conditions, and correct for the distance from the approaching salmon, the black definition is slightly suppressed, when viewed horizontally against the background of Window 1.

Viewed obliquely in Window 2 it will look more like this (relative adjustment for background brightness), and in Window 3 it becomes almost entirely black.  It's an approximation of some kind of nymph, but certainly not a close imitation.  As a result the salmon will not be able to discern much and so is less likely to be deterred until the last moment.  In both horizontal and oblique views the black is critical to the detection of a very small moving object at distance (which is why all the RAF's trainer aircraft are now painted black rather than red and white).

Conclusions

So where did all that tell us about fly design?  Everything that follows is wholly conditional, but it appears that:

• In dull conditions and at depth in coloured water, the fine distinctions between broadly similar orange/red shrimp patterns are irrelevant.
• The real value of orange may not be in its high visibility in air, but rather that in water it looks an interesting shade of grey in dull conditions and/or when the fly is fished deeper.
• The orange based grey differs in some way from pure grey, especially at closer ranges.
• Black is the key to detection - alerting the salmon to the fact that there's something there, however small, over extended distances in all water conditions and depths.
• As the water clears and lightens, or when fishing shallow, bright colours and excessive definition may allow the salmon to get "too good a look" at the fly and may therefore be disadvantageous.
• For clear bright water, dull anonymous dressing seems to work well, possibly by preventing the "good look".
• None of the most popular low water flies have any orange/red component.
• There is no obvious way of combining all of those qualities in a single fly.
• The perfect fly is probably two black-bodied flies: orange for dull conditions and deeper fishing; and grey/brown for bright and shallow.
• The open question is the optimum proportion of black in each design, which I can only refine through practical trials on unpredictable fish.

So what?

There's not much point reaching some conclusions and not putting them to the test.  So the next step is to design, refine and test some prototypes.  As I don't tie my own flies the challenge was to find someone open-minded and willing to work with me on the project.  The perfect candidate was the very talented Darragh Digney in Belfast, who despite having little salmon fishing experience, shows an extraordinary and original eye for colour and design in tying flies.  Fortunately he agreed to help me, and after consultation the first prototype of the MCX Dark Shrimp is now complete.

In subsequent posts I shall describe our joint development journey, but I shall certainly need some water for testing and trials!  For the sake of my farming friends (and to avoid being lynched) I'll postpone ordering Dave Bamford's Wensleydale rain dance for a couple of weeks.

For those of you lucky folk with water, tight lines.