Thread; A Tangled Mess

There is a lot of good information about fly tying out in the world these days.  Between books, magazines and the internet, there is no shortage of adequate instruction and advice on almost all aspects of fly tying.  Except for the apparent step child of fly tying tools; thread.

Not only is there just not much out there to clear up the understanding of this vital piece of every fly we tie, much of what is there is just plain wrong and misleading.  I have had the idea for this article in my head for quite some time and am often specifically asked to explain tying threads in my demos.  With that in mind, I will now give my best shot at explaining some of the finer details of this most misunderstood tool.

Note that I used the word “tool” rather than material when describing thread here.  In my mind, thread is a tool chosen and used to perform a specific task.  At the moment, I am certain I own no less than one thousand spools of thread in a huge range of sizes, colors and configurations and while every last one of them is good for something, none of them are good for everything.  This selection lets me choose the right tool for the job in every instance.  I’m in no way saying you need a thousand spools of thread on your bench, but hopefully what follows will help to clear up the details of why threads are different and what those differences mean to the tyer.

Before I get started, let me say just a few words about wax on tying thread.  Almost all conventional thread comes from the factory pre-waxed these days and if you wanted un-waxed thread you’d have to use silk, Kevlar, GSP or special order some of the older Danville threads like size A Monocord.  Waxed thread is not meant to be a substitute for dubbing wax, but rather, it is applied to the thread in the manufacturing process to keep the thread from fraying and unwinding wildly when it breaks.  Glad to get that off my chest.  Now on to the rest of the story…

Historically, fly tying thread was sized using the aught/ought system that came originally from sizing surgical silk using a range of diameters, and that is still used today for that purpose.  Using a zero as a baseline and adding zeros to denote smaller sizes, this system has been around since the late 1930’s and was for a long time the way all tying thread was measured.  In the case of surgical silk, the aught number referred to a diameter range and worked nicely as silk is a hard, round filament that can dependably be measured in finite terms. Many tying threads are still measured this way today, as an example, 000000 or 6/0 (six-aught) being smaller than 000 or 3/0 (three-aught) thread.  The first major fly tying thread on the scene (Danville) used this system for about fifty years without issue.  It was only when other thread companies came into play that the tying world became a lot messier.

In the early 1990’s UNI thread came into the market and while sticking with the aught system of sizing, used a somewhat different baseline for their thread, and that’s where everything went to hellUNI came out with what they called 8/0 thread and that really sort of screwed everything up from there.  Given the system of measurement already in place, the hapless tyer was led to believe that the new 8/0 thread was smaller than the old standard of Danville’s 6/0, and the newly introduced UNI 6/0 thread was much, much stronger than the corresponding Danville thread.  As it turns out, neither of those statements are entirely true and I’ll get into all that in a few minutes.

Enter Wapsi Fly Company and their introduction of UTC thread, sized using the thread industry’s standard of measurement, denier, rather than the historic, and now known to be somewhat useless aught system.  Denier is the weight, in grams, of 9000 meters of thread.   It’s a physical weight for a length of thread.  It says nothing of the thread’s configuration, material or strength, only the weight, but it does shed some light on the inadequacies of the aught system.  UTC debuted with thread in 70, 140, 210 and 280 deniers, and simplified thread sizing to a significant degree.  It’s pretty simple math to figure out that 140 denier thread is twice as heavy as 70 denier thread, 210 is three times as heavy and so on.  Now we’re getting somewhere, but it’s a long way from perfect.  Denier as a physical weight measurement has some liabilities in measuring fly tying thread as some materials are simply heavier than others.  Polyester is heavier than nylon and GSP is lighter than nylon.  Given diameters and lengths will have varying denier ratings simply based on the material of the thread…so we aren’t quite to that magic system of measurement that works for everyone quite yet.  Using the denier system retroactively tells us that UNI 8/0 thread is 72 denier, Danville 6/0 thread is exactly 70 denier and surprisingly, UNI 6/0 thread is 135 denier…no wonder it seemed so much stronger; it’s almost twice as heavy!  And what about that miracle thread, 8/0 UNI being actually a bit heavier than the old Danville 6/0?  It’s not the size or weight that varies so much between these two threads, but instead, the material and configuration of these threads is what makes them so different!  Now we see where the confusion comes from.  There is no constant baseline used by all manufacturers to clearly and concisely size and accurately label their threads!  As tyers, we all now know that 70 Denier UTC thread, Danville 6/0 and UNI 8/0 thread are all about the same weight, but we also know that these three threads are yet still very different in application and that is because of both the materials used to make them and the actual configuration of the thread.

The configuration of a tying thread is, in my opinion, the most telling factor in deciding what it might be best suited for.  By configuration, I mean the structure of the thread itself.  Most threads are configured in either a round or flat shape when viewed in cross section.  There are degrees to both of these configurations of course, just to keep things interesting.

Most fly tying threads these days are made of nylon or polyester.  Polyester is slightly heavier than nylon and correspondingly will have a higher denier for a length of the same diameter as well as a bit more strength. There are specific use threads made of other materials like gel-spun polyethylene (GSP), Kevlar and yes, even silk and monofilament and I’ll get to those in a minute.

Danville and UTC threads are both made of nylon.  Nylon has a fair amount of stretch, (25-30%), naturally lies flat and typically features a silky smooth finish and a glossy sheen.  Being nylon, these threads allow dubbing to slide more than polyester threads and while still perfectly useable for dubbing applications, one should expect the slicker the thread the less the dubbing will adhere to it. Both of these threads are of a flat configuration, the UTC thread being particularly so and is a bit shinier than the Danville.  These are flat threads that wrap on the hook like a ribbon.  These threads are configured with multiple strands held together with a binder strand twisted around the center strands to hold the thread in shape.  Danville thread has a binder strand with more twist than UTC thread.  UTC lies very flat because it is held together with a binder strand that twists around the main core at a rate of only one revolution per inch.  This open twist allows the thread to spread out and lie flat on the hook shank where threads with more twist in the binder strand tend to bulk up vertically more.  The flatter, smoother Danville 6/0 and UTC 70 threads are particularly useful for thread bodied flies or flies requiring a smooth thread underbody, as they create silky, seamless tapers and bases.  I like both of these threads for flies like Black Beauties, Copper Johns and Humpies.  In the case of the Black Beauty and Copper John, this is because the smooth texture and low bulk allow me to build smooth bodies and underbodies without ridges.  Conversely, my friend Jay Zimmerman prefers UNI 8/0 for the Black Beauty as he believes the ridged thread body holds the wire rib more securely in place.  I can’t say he’s wrong and there’s room for opinion on all of this.  In the case of the Humpy, I find these threads to be a perfect for building both the upright divided wings with flattened thread to keep them from flaring, and then building the thread hump to create a vibrantly colored, smooth underbody.  I admit I do twist both of these threads into a more corded configuration (by spinning the bobbin) to attach the hair to a hook as I find round thread grabs, cinches and compresses hard hair better than flat thread.

In trying to better replicate actual tying procedures, here I used ten wraps of thread to approximate a thread head using a variety of threads. From left to right: UTC 140 Orange, Danville 3/0 Monocord Beige, UNI 6/0 Pink, UTC 70 Yellow, 8/0 UNI Olive, 14/0 Gordon Griffith’s Sheer Orange, 14/0 Veevus Gray, 16/0 Veevus Cream and finally UNI Trico (40 Denier, 17/0) in white. These are descending sizes amongst threads of varying configurations. The more flatly lying threads like UTC 140 and 70 tend to spread out a bit more while the corded threads like UNI form a somewhat more compact bundle.

I like the larger sizes, UTC 140, UNI 6/0 and good old Danville 3/0 Monocord (rated at 140, 135 and 116 denier, respectively) for streamers and saltwater flies.  I find the slightly narrower nature of the Danville 3/0 to be particularly useful for many of my streamer patterns like the Baby Gonga and Dirty Hippy, but again, the colors here are somewhat limited and muddy.  UNI 6/0 is very much like the Danville 3/0 and I use it in many of the same instances. It is one of my favorite threads for flies like Crazy Charlies, as the round configuration bites into the D-Rib overbody material as well as the hard calf tail wing for a more secure tie down.  Medium sized saltwater flies like my Ragin’ Craven and Flip Flop get UTC 140 as there is plenty of room for the bulk on the larger patterns and this flatter thread creates a smooth, clean profile around the weighted eyes.

I find I only use the largest sized UTC threads, 210 and 280 denier, on flies that require quick and heavy thread coverage like the Pigsticker or for the hot head on Cliff Watts’ Kilowatt.  In these cases, thread that builds and covers quickly and smoothly is paramount and both of these fit the bill perfectly.  I find these typically too big for even gargantuan sized streamer patterns and frankly, overkill for even some of the biggest flies.

Flies like the Pigsticker are built entirely of thread over a lead wire underbody. Large size threads like UTC 140 and 210 denier make this a quick easy job and their flat nature makes for a beautifully smooth body.

UNI’s polyester threads have less stretch (15%), are more roughly textured and lay a bit more round or corded on the hook.  The rougher texture of polyester threads accepts dubbing particularly well.  Colors are typically a bit more muted and subtle in polyester threads than they are in nylon with more of a matte finish.  UNI thread is built with a round configuration, although it can be flattened a bit by diligently unwinding it or wrapping the thread left handed.  Yes, you read that right.  Most threads are twisted such that as a right handed tyer wraps them around the hook they contribute to the factory twist in the thread.  For us left handers, we enjoy an unintended advantage in that as we wrap these threads we unwind them…causing them to lie much flatter than our right handed friends.  UNI thread is a bonded thread that is reinforced with multiple binder strands that twist around the core strands at a much higher rate.  This means that in cross section, UNI thread would look more like a rope while Danville and UTC would appear more flat and ribbon-like.  8/0 UNI thread is a great example of a thread that is very similar in denier to both the UTC 70 and the Danville 6/0, but made up in a very different configuration and material.  8/0 UNI thread, as well as all the other sizes from UNI, are much more round and tend to build bulk vertically on the hook shank rather than horizontally as the other nylon threads do.  Think of a stack or line of o’s in this case and a stack or line of –‘s in the case of the nylon threads.

The round cross section coupled with less stretch and a more textured surface of the UNI poly threads allows for a bit more ‘bite’ and ‘grab’ to the thread in tying applications.  Round thread bites into and flares hard hairs like deer, elk and moose better than a flattened thread does.  The rougher texture of poly threads also holds dubbing better than the slicker nylon.  Because of this and my inherent left handedness, I really like 8/0 UNI thread and use it on many of my trout patterns.  Its ability to be flattened or twisted into a cord and the harder nature of polyester versus nylon make it a great choice for smaller flared hair wing flies like a Comparadun or X-Caddis or anything made predominantly out of dubbing.  My biggest gripe with UNI thread is the colors are not as vibrant as the nylon threads but that’s simply a matter of my opinion…the fish really don’t seem to care. I should mention here that the popular Gordon Griffith’s Sheer 14/0 (72 denier) thread is made of polyester as well but is configured more like the Danville 6/0, striking a nice compromise between a hard, round poly thread and a flatter nylon.

When tying in hair wings, I prefer to twist my thread up slightly to cord it. The round thread creates a harder edge that bites into the hair, is less likely to slip and totally compresses the fibers under the wraps. Flattened thread spreads thread pressure out a bit too much for my liking in these situations.

Veevus is a new brand of thread that has recently come on the scene and become quite popular.  It is the only modern era thread that is un-waxed. All of the Veevus middle sized threads (8/0, 10/0, 12/0 and 14/0) are built in a conventional fashion with a bit of twist but with the ability to be flattened beautifully by unwinding, while both their largest and smallest sizes (6/0 and 16/0) are built of just two intertwined strands in an counterclockwise twist.  These threads too flatten and cord up nicely and in the case of the 16/0, create negligible bulk when tying.  Both of these two margin sizes lend themselves wonderfully to split-thread dubbing, in case you’re into that sort of thing.  Oddly, a little bit of online research has shown me that their 6/0, 8/0 and 10/0 threads are all rated at 110 denier, and the 12/0 and 14/0 are rated at 70 denier with the 16/0 rated at a fine 50 denier.  I cannot for the life of me figure out how they have managed to make distinctly different sized threads out of a material of the same denier, but they have.  The 12/0 and 14/0 threads are clearly different sizes and honestly are, really, really nice threads.  They lie flat or cord up at the tyer’s whim, they have surprisingly good strength and the colors offered are more than adequate for all the tying I do.  I just wish I could give you a straight answer as to how they make the stuff but, after an extensively long email string with the head of the company, they either don’t know, or don’t want us to know! Their advice was not to concentrate so much on the technical aspects of the thread but to instead buy a few different sizes and see what works best for your own tying. This isn’t bad advice, although I must say it wasn’t very helpful in writing an article about those very technical aspects.

Forty turns of Veevus thread Left to right: 6/0 Black, 8/0 White, 10/0 Cream, 12/0 Chartreuse, 14/0 Orange, and 16/0 Brown Note the corded nature of both the 6/0 and the 16/o threads in relation to the more flat center sizes. The largest and smallest Veevus threads are built in a twisted two strand configuration while the other sizes are formed more like conventional thread. The two strand threads lend themselves quite well to split-thread dubbing techniques as the strands are easily separated.

Close up view of flat UTC 70 Denier on the left and UNI 8/0 on the right illustrating the difference between the configurations of the threads. Both of these were wrapped as they come from the spool with no twisting or untwisting done beyond what normally occurs when wrapping them around the hook. Both configurations have a purpose and they are both good for something, but neither is good for everything!

Pictured here is UTC 210 Denier started at the rear of the hook shank and flattened to spread out to its maximum width and smoothness, then twisted and corded to reduce the diameter on the front of the shank. We can clearly see the two dimensional nature of these threads and how they might vary according to the amount of twist introduced or removed from their factory state. The flattened thread is about four times as wide as the tightly twisted thread.


GSP (gel-spun polyethylene) threads are very slick, very strong for their size, have very little stretch (3%) and lay very flat on the hook.  They are typically used in applications like spinning deer hair or synthetic materials that call for an abnormally strong yet small diameter thread.  GSP thread holds up very well to toothy fish and creates very durable flies.  Their slickness can be somewhat of a liability in most other applications however, particularly dubbing retention, and dye used to color them is not as stable on GSP thread as it is on other materials.  I only use this thread in white for reasons I’ll give below.

Kevlar is sort of the precursor to GSP threads.  It naturally lies quite flat, is incredibly strong and comes in but one size…big (200 denier).  It is most often used in larger spun deer hair patterns like bass bugs.  Again, dye coloration is not very stable in this material and can run when head cement is applied, so again, I only use this thread in its natural coloration of pale yellow.  Nothing like tying a perfect white bass bug and having the thread leach dye into it when you add a final drop of head cement.

These two threads are, for all intents and purposes, interchangeable.  GSP thread is much slicker however and both can be hard to cut with your tying scissors.  I’ve heard horror stories about the abrasiveness of both these threads eating bobbin tubes, but can’t yet, after almost forty years of tying, confirm this for myself.  GSP thread does come in a wider range of sizes, from 30 denier up to 200, where Kevlar seems to only have ever come in 200 denier.  I might lean slightly more toward the Kevlar thread for larger hair bugs, as it is not as slick as the GSP and I believe that makes for a more securely tied bug without the use of glue and epoxy at every step.  This is just my opinion here.  There are plenty of good tyers these days who swear by the GSP stuff and I’m getting to be an old guy so take that for what it’s worth.  I do like GSP in smaller sizes for compact spun deer hair work in some cases.  I like to use it in slightly oversized sizes as I find the really small stuff can cut right through a clump of softer deer hair, particularly if it’s not flattened.

Mono Threads are just what they sound like; fine, round, single strand nylon monofilament…like fine tippet material.  They are used in very specific applications when you’d want the thread to disappear under a coat of epoxy or head cement.  Their single strand configuration leads to excessive twist in the tying process and thus they must be managed to control and unwind this twist while in use.  Mono threads are not particularly strong for their bulk but perfect for flies like EP Minnows and other synthetic hair patterns as the clear coloration allows the body materials to show through the wraps for continuity.

Silk was the first thread used to tie flies and is still wonderful stuff.  Two twisted strands are used to create this thread and they can easily be twisted to flatten or cord them up as the tyer sees fit.  It’s not particularly strong and nor very common these days but still useful in historic applications and frankly, it adds a bit of class to many patterns.  Silk thread is also unique in its propensity to change colors when wet.  I still like this thread for soft hackled patterns as well as some ribbing applications, but as a general tying thread it has sort of gone by the wayside.

After all this, I am afraid the best advice I have to offer as far as truly understanding threads is to do what you have likely been doing for years…buy several different kinds and sizes and see what suits you and your tying best.  While a more consistent and understandable sizing system would be a truly lovely advancement, fly tyers are really a far too opinionated lot to ever take someone else’s word for anything without trying it for themselves.  We’ve probably gotten what we deserve!

If you’re now finding yourself a thread nerd and want to really dig in for a better understanding of thread sizing, materials and breaking strength, check out Martin Joergensen’s fantastic thread table found here on the Global Flyfisher website.  I used this chart for reference to the denier sizes and diameters listed in this article and found it to be a plethora of useful information in a sea of drivel.

As a fly tyer, I believe a simple coding system giving the dimensions of the thread (height and width), breaking strength and material used would be so much more useful and easy to use and decipher than any of the measurement systems currently in place. A labeling code that listed the dimensions of the thread in partial millimeter scale (microns?) such as .001X.004 for the vertical height (.001) and horizontal spread (.004) of the thread would give us a much better idea of the threads dimensions in use. As most threads truly do have two dimensions depending on if they are twisted up or flattened out, a scale of both of these would be incredibly useful in thread selection. The measurements would have to be taken with the thread at a completely untwisted state for the horizontal measurement and in a factory twisted state for the vertical measurement. Making these measurements in common, standardized terms would prevent any marketing or hype from slipping into the labeling system and misleading anyone who doesn’t know their way around this mess. Round threads would be easily determined by the width and height numbers being the same (.002X.002). The actual breaking strength of the thread is an easily determined number, most often measured in grams with a standardized measurement system already in place and verifiable. The material used to make the thread is a known variable and could be designated with a single letter, i.e., P=Polyester, N= Nylon, G=Gel Spun Polyethylene, K= Kevlar, M=Mono. Given this system we could do away with the ought system as well as denier and have real, viable, comparable measurements as well as configurations in a simple code. A UNI 8/0 thread might read something like this 02×04-450-P. This code tells us that the thread is .002 mm tall, .004 mm wide, breaks at 450 grams of pressure and is made of Polyester. (As a side note, I completely made these measurements up as an example because I am not nearly smart enough to figure out how to measure these dimensions myself…but someone out there is!) Overall, this theoretical system devised by Jay Zimmerman and I on a quiet day in the shop will need a few bugs worked out and would take a bit of getting used to, but long term it makes a lot more sense than anything else we have in place at the moment. I’m just spit ballin’ here and I’d be amazed if any of us lived long enough to see any standardization in the fly tying thread industry, but hey…a guy can hope, can’t he?


Very valuable info.!

Ed Walker

I totally agree. Craven, YOU have ALWAYS amazed me …kudos to YOU for stepping forward! The credit belongs to YOU! Like you said, it’ll only happen after we (who are here now) are long gone, but it will have to happen eventually. Hopefully, someone able to change things will read this.


Leave a comment