Published Mar 31. 2014 - 9 years ago
Updated or edited Oct 28. 2021

Fly Tying Thread Blues

The old aught system for tying threads (6/0, 8/0, 10/0) isn't really good. So the newer Denier system is much better, right? No, not really!

All sorts - Diving into my tying thread stash reveals a large number of different threads - some from well known manufacturers and some absolutely unknown.
A selection - I have a large selection of spools ready to tie. These are just a random bunch, which has been in recent action
A selection
Martin Joergensen

Those of us who have tied flies for decades have been brought up with a very simple thread system. There was in essence four kinds of thread: really thin thread, 8/0, 6/0 and really thick thread.
You could buy thread called 3/0 (thick) or spiderweb (thin). Monofilament for tying was also available, but in general that was it. And if you were like me you tied with 8/0 and 6/0 95% of the time. 8/0 for small flies and 6/0 for the larger ones.
When I started tying I used Uni, Danville or Gudebrod. That was about it. Some places I could find Pearsall's silk and a few places had Bennechi thread, but certainly not in my average tackle shop.
This very simple landscape has changed significantly during the last decade or so.

If you were like me you tied with 8/0 and 6/0 95% of the time.

Many more threads

Today there are numerous fly-tying thread brands, and even though the number of manufacturers as such (who run actual thread mills) is definitely both smaller and probably different from the tying thread brands, there's still a much larger selection. Most of the tying thread we see is most likely just spooled and labelled by the "manufacturer".

These tying thread manufacturers use two systems to label the threads, aught and Denier.

A little history

The so called aught system (AKA the ought or naught system) with the numbers 6/0, 8/0, 10/0 and so on, was introduced back in the thirties where Danville started using it as a way of specifying threads. The more zeroes the thinner, so a 000 thread was thicker than a 000000 thread. For the sake of writing, reading and saying the specs, they were written 3/0, 6/0 and 8/0 and pronounced three-aught, six-aught and eight-aught.
The system was adapted by other manufacturers, prevailed and has lived on for decades. It's still in use, but lately (like the last decade), Wapsi's Denier specification has become more and more common, used alongside the aught system or on its own.

Wapsi introduced its UTC thread in the eighties, and unlike the other brands, Wapsi labeled its thread with Denier - 70 UTC and 140 UTC, 70 and 140 Denier respectively.
The thread wasn't commonly available (and can actually still be hard to find in some countries), but Wapsi was a true first mover with regards to actual physical thread specifications in fly tying. Few of the other brands noticed, none followed the lead.
That might not be a big problem for reasons, which you can read below. The current use of the Denier system is not the end of all confusion.

Measuring thread

For the sake of getting precise information about thread diameters I bought myself a micrometer with a precision down to 0.001 millimeters or 0.0005 inches, which should be sufficient to precisely measure tying thread, which is rarely thinner than about 0.025 mm or about 0.001".
Now, measuring with a decent precision is one thing. Another is getting a result that's actually true for the thread. There are many aspects to take into consideration:

Softness: thread is soft and the micrometer presses it together, flattening it a bit and seeing a thinner diameter than the actual diameter of the "relaxed" thread. I tried to avoid this by using the thimble on the micrometer in a consistent manner, letting it click once for each measurement, reading the number and then clicking a couple more times to see that the thread would actually compress. 8 out of 10 times it did, telling me that I hadn't compressed the thread too much to give useful results.

Twist: As noted twist influences the diameter of the tread. Most threads are thickest when relaxed and get thinner when you twist or untwist them. Twisting tightens the turns and narrows the diameter, and untwisting makes most threads flatter - thin and wide. Multistrand threads like GSP with many parallel filaments act the opposite way and become, if not thin then extremely flat when relaxed. Measuring such a thread may show it as very, very thin, but at the same time, it's very, very wide. I decided to let the thread I measured untwist to a "natural" degree by hanging a length of thread weighed down by the bobbin and a bobbin-holder and letting it untwist until it was still. This didn't necessarily give the "real" diameter, but gave sufficiently consistent - and in my eyes representative - results for all threads.

Sampling: As with all measuring, sampling method and statistical evaluation is critical, and I tried being as fair as possible here. Each type of thread was measured on several spools if possible (I only had one of some) and I measured different spots over different lengths spooling off thread to get to fresh material. I measured the thread many places and calculated the average. I ignored (but noted) really strange outliers like a series 0.050, 0,051, 0.055, 0.052, 0.037, 0,055. The 0.037 is either a measuring error or a "bump" or error in the thread, and should be ignored if it was a single occurrence. Many such errors would indicate that the thread was inconsistently made.

My measuring setup - I hung the threads on a bobbin holder from a round tube and let them spin until they were static and then measured each length about 5 places and noted the numbers. Then I spooled off more thread to get to a new, \"virgin\" stretch and repeated.
The micrometer - My micrometer can measure down to 1/100 of a millimeter or about 5/1000 of an inch. This thread is measured to be 0.060 millimeters
The measure station
Martin Joergensen

The Denier system is a physical standard, and is an expression for the weight in grams of 9,000 meters of thread - equaling 29,500 feet, 9,840 yards, or 5.6 miles.
The reason for this seemingly odd measure is that 9,000 meters of silkworm silk allegedly weighs one gram and has set the base unit - 1 Denier. Spin a thread of 100 strands of silk, and you get a 100 Denier silk thread.
The measure is widely used in the textile industry, which use a number of other specifications too, and the world of Denier, tex, Decitex, filaments, fibers, threads and yarns is a really complex one.

8/0, is 8/0, is 8/0... or is it?

Unfortunately none of the common measurements - aught or Denier - are really useful for specifying fly-tying threads.
The worst one is the aught system, which is basically useless in my opinion.
Just looking at different manufacturer's threads specified using this system reveals its inadequacy. Here's an example using the very common 8/0 polyester threads (see also the table below):

1) UNI's 8/0 is 0.051 millimeters thick and breaks at about 425 grams.
2) Bennechi's 8/0 thread is 0.056 millimeters thick and has a breaking strength of 822 grams.
3) Griffith's 8/0 thread is the same diameter as Bennechi's 0.056 millimeters, but about half the strength, 425 grams.
4) Gudebrod (discontinued) had a diameter of 0.046 millimeters and also broke at app. 425 grams.
5) Veevus 8/0 thread is 0.059 mm and has a breaking strength of "over 1 kg", more than 1000 grams.

If we include 8/0 threads made from nylon and GSP, it just blurs the image even more:

6) Montana Fly Co's 8/0 nylon thread is 0.036 millimeters and breaks at 397 grams, so it's much thinner, but also weaker than the average 8/0 thread.
7) Roman Moser Power Silk is GSP and is only 0.033 millimeters, but holds more than a kilo like Veevus' polyester thread, so it's also much thinner, but significantly stronger than the average.

And the Denier measure for these 8/0 threads will not make you wiser. They should be near identical if they were to be generally helpful, but vary from 55 Denier to 150, so a factor three between the "thinnest" and the "thickest", which are nowhere near being three times thinner or thicker than each other.

They should be near identical,
but vary from 55 Denier to 150

Very different

The 8/0 threads are in other words very different and in no way comparable, having great variation in diameter and even larger variation in breaking strength. Of course you can add to that the other properties of the thread: coarseness, ability to flatten, waxed/unwaxed and more.
To the average fly-tyer one 8/0 thread will be assumed identical to the other, but as you can see that isn't the case: diameter varies greatly and strength and physical properties depend on the material and the way the thread is made. Notice that the first threads are all polyester and almost homogenous regarding thickness, but when we include nylon, not to mention GSP, the variation is colossal.

This little table shows that essentially neither the aught system nor the Denier specification is any good for indicating the properties of a fly tying thread. You learn nothing about its thickness or breaking strength from either of these numbers - and of course absolutely nothing about the nature of the thread.

8/0 tying thread table

Click the headers to sort. 16 tying threads listed.

Brand Product Material Denier Dia
Benecchi Fine 8/0 Polyester 150 0.056 822
Cascade Crest Euro Thread Polyester      
Cascade Crest Spider Web 8/0 Nylon      
Gordon Griffiths Wisp Polyester 108 0.056 425
Gudebrod 8/0 Polyester 67 0.046 450
Montana Fly Co 8/0 Nylon 72 0.036 397
Orvis 8/0 Polyester   0.055  
Petitjean Split Second Thread     0.025  
Roman Moser Power Silk 10/0 (fine) Dyneema Thread GSP 55 0.033 1077
Semperfli Tying Thread 8/0 Polyester      
Semperfli Classic Waxed Thread Polyester     754
Sparton Micro Polyester 72   566
UNI UNI-Cord GSP 75 0.036 1928
UNI Waxed Midge Polyester 72   450
UNI UNI-Thread 8/0 Polyester 72 0.051 450
Veevus 8/0 Polyester 110 0.059 1000

Notice that not all info is available on all threads. Some manufacturers supply very little informaiton on their products.

See the full table with more than 110 threads here, and if you want to be really nerdy, you can see some scatter charts that show correlation (or rather: lack of same) between thread diameters, deniers and breaking strengths even between threads from the same manufacturer or threads made from the same material.

Still confusing

I have been unable to find any references to the physical aspects of the aught "measurement", which in other words fail pretty miserably as a specification, not to mention as a standard. It is - as we have seen above - nothing that even remotely can be called a standard or even a useful specification.

"So", you say, "Simply use Denier and the matter is solved!"
Unfortunately not. Some manufacturers like Veevus won't specify Denier for their threads. (the Denier number for Veevus above comes from another source).
The 8/0 threads listed also vary from 55 to 150 Denier. Now, there's nothing wrong with that and it might be very accurate. Remember that Denier is linear weight, in other words weight per length of the thread, so the Deniers just tell us that these threads have very different weights per length.
But compare the Bennechi, Griffith and UNI, all basically the same diameter. They are 150, 108 and 72 Denier respectively! How can I as a fly-tyer use that for anything? It's simply confusing. In another way than the aught system, sure, but still confusing.
And like the aught system, it's insufficient to properly specify the physical properties of the thread.

And like the aught system, it's insufficient to properly specify the physical properties of the thread.

Twisting and turning

Other aspects of the thread can also be important.
Material of course, where polyester is the most common, nylon also quite common and GSP (Gel Spun Polyethylene) has become common the latest years. You will also find real silk as well as Kevlar. Each has its benefits and drawbacks, but while polyester and nylon are rarely specified, GSP almost always is and silk as well as Kevlar always is.

The different materials have fundamentally different properties, like stretch and smoothness. Nylon is pretty elastic and has up to 25% stretch, polyester is less stretchy with about 15%, and GSP is very numb with just 3% stretch. Polyester is quite rough while nylon and silk is smoother. GSP is very smooth while Kevlar is smooth, but has a special bite because of the surface of the fiber. This has great influence on the tying, and is one of the reasons than many tyers wax their threads or buy waxed threads, because it increases bite and eases the tying.
Add to that the production method or thread type of which single filament (mono), simple twist, bonded, spun and rope types are just some. Threads can be made in many ways, filaments spun and fused in many ways, and each way produces a thread significantly different from the other.
This is very rarely specified by the manufacturers even though it can be quite important for the tyer. If you like to split dub, some threads are useless because they can't be separated to open to the dubbing. If you like to flatten thread to avoid bulk, some threads flatten really well while others are more rope like and unwilling to flatten.

Large variation

As noted elsewhere I bought myself a micrometer and measured a bunch of threads. I knew that there would be some variation, but for some threads it was pretty large - up to 20 and 30% over even short stretches of thread like a few inches. It's usually not a big issue when you tie, but for the average and weaker threads it can mean that you have parts of the thread that are significantly weaker than the rest.
Thread diameter also varies depending on the twist. As most tyers know, you can untwist and flatten many threads, and while they do become wider on one side they become significantly thinner when the filaments become more parallel.

What will work

The only physical measurements, which are interesting for tying threads, is diameter and breaking strength. Fishing line manufacturers have found that out decades ago, and all their spools are clearly labeled with thickness in millimeters and/or inches and test (breaking strength) in kilos and/or lbs. And that goes for nylon and fluorocarbon as well as different spun and fused lines, which are fundamentally different from nylon in material and production method. Still these lines are easily comparable thanks to two simple numbers.
How hard can it be?
It's a very simple thing for the tying-thread manufacturers to measure thickness and test, and would make selecting a tying thread so much easier.
If they added material and thread type or production method to the labels, it would be a fly-tyer's Nirvana!
Simple, uniform, standardized and comparable.
Manufacturers and spoolers of fly-tying threads, put this on the label:

  1. Thickness in millimeters and inches
  2. Breaking strength in grams and ounces
  3. Material (Polyester, Nylon, GSP etc.)
  4. Thread type: monofilament, twisted, spun, braided or whatnot

But that's probably way too much to ask...

How hard can it be?

A huge selection - This is the impressing thread selection in Whitetail Fly Tieing - the shop formerly owned and run by Chris Helm
A huge selection
Joe Cornwall

A conclusion

Well, if you are expecting me to tell what's the best thread on the market, you will be disappointed. That's like asking me what's the best rod... are you fishing #28 nymphs on a small stream, salmon in a rough river or tarpon in the tropics? Even knowing this, I'd have a hard time recommending a certain rod for you.
Same thing with thread.
The ones labelled "Very weak" in my table might make you shun away from them, thinking that they will break all the time. Sure they are more fragile, but if you are tying those small flies they might work for you. But if you look for thread for small flies, the table also tells you that other threads might be worth looking into, like the thinnest GSP threads, which are all thin and strong, or the thin polyester threads from several manufacturers, which are both thin and pretty strong without being as smooth as the GSP.
In the other end of the scale a thread that's more than 3-5 hundredth of an inch thick like the UNI Bigfly or the Dyneema might scare you off for being thick and clumsy. But if you want practically unbreakable thread for tying big flies, 4-6 lbs breaking strengths might be just your thing in spite of the bulk.

Finding your thread

This article and the table enables you to find a thread that suits your needs based on diameter and breaking strength where it's available, and also enables you to compare threads which are usually difficult to compare.

An example: I really like the Gudebrod 8/0 polyester thread. Unfortunately Gudebrod is in limbo, and their threads very difficult to find. But looking at the table tells me that Griffiths Sheer, Veevus 12/0 and 14/0 as well as Bennechi 12/0 and Lagartun 150 Denier, all polyester threads, are the same diameter, much the same type of threads and both Veevus 12/0 and Lagartun 150 Denier are actually stronger than the Gudebrod, so they are very good alternatives. My empiric findings have already confirmed this, because for a long time my favorite thread has been Veevus 12/0, found simply by tying with it and liking it.

Many other fly-tyers have written about this subject, like Chris Helm, who wrote a series on thread published here on GFF, Mike Hogue "Choosing the Right Thread", Scott Sanchez "Choosing the Right Thread" or John Wood "Fly Tying Thread Explained".

Links to thread manufacturers

Most of the thread manufacturers have underwhelming web sites - to say it nicely. Very few have well ordered, systematic records of all products online. And as always: links often change or disappear, and may not work when you read this.
Danville Chenille Company Inc.
Montana Fly Company has very little info on its tying threads
Roman Moser
UNI Products
Veniard/Gordon Griffiths

Hareline Has a large selection of threads from many manufacturers

Lagartun and Giorgio Benecchi don't seem to have own web sites.
Gudebrod has closed its thread production.


A lot of the basic information for this article is based on work done by US fly-tyer Chris Helm, the former owner of Whitetail Fly Tieing where his thorough coverage of tying thread is still found and kept updated by the new owners. Chris has written about thread online and in magazines, and his articles "The Introduction of Denier" and "Using Denier to Standardize Fly Tying Thread" are considered a basic work with regards to using Denier in specifying fly-tying threads.

Steven H. McGarthwaite has also written a very thorough article on fly-tying thread on FlyAnglers Online.

The article "Hanging by a thread" by Dr Paul Davis is also a very thorough coverage of threads, with exact measurements of a large number of threads. Some of the data in the present are taken from Paul's article. You will find Paul's shop on

Paul Marriner's "How to Choose & Use Fly-tying Thread" is a small booklet with information on selecting and using tying thread. It's made for the manufacturer UNI Products and mainly covers their threads.

In the book "Tying Small Flies" by by Ed Engle there's also a short section on tying thread and the aught system, and the author comes the the same conclusion that I do: the system is pretty much useless for comparing threads between manufacturers.

The article "Thread Essentials" by Carol Laflin Ahles on sewing threads is actually very informative too. It concludes the same as this article, saying that there's no one standard for threads, but at the same time covers a lot of thread types and materials in detail, much of which is also applicable on fly-tying thread, which by sewing standards is generally very thin by the way.

Articles in this series

Work in progress

This article on tying threads and all the data that is part of it is work in progress and will be updated as soon as I get more or better information on the various threads.
If you have things to add, comments or corrections, feel free to contact me on You can also leave a comment to the article, and I will see it and respond if required.
I welcome contact from tying thread manufacturers with information, specifications, remarks or corrections - even criticism. The more the better!


Martin Joergensen's picture

Paul, I tend to a...


I tend to agree and then disagree with you at the same time. Sure you can't measure the diameter of threads, which are not round, but in the cases where I have actually measured the threads, I have done so repeatedly on several threads using the methodology described, and even though you might think that such measures would yield very different results when repeated over a length of thread, the numbers actually varied very little for the threads I worked with. Sure the measuring itself can maybe affect the result, but my impression is that the result is actually a valid measurement for a thickness - attempting to avoid using the term diameter here. It indicates a circle round thread, which only very few threads are. A more fair measure could be dimensions - measured on the smallest and largest thickness when imagining the thread cross section as a rectangle. Unfortunately it's beyond my skills and my tools to do such a measurement.

Regarding the use of Denier, you are right inasmuch as it's a true, physical measurement, but there are simply too many aspects and factors to take onto consideration for it to be useful for the average fly-tyer in assessing the properties of a certain thread - not to mention comparing thread from different manufacturers made of different materials using different methods. It's still insufficient to be able to select a thread for the far majority of fly tyers. One 70 Denier thread can be as different from another 70 Denier thread as one 8/0 thread can be from an 8/0 thread from another manufacturer. As long as neither materials nor fusing or spinning methods are specified, the Denier number is as confusing to most as the aught system.

I have your tying thread booklet and can warmly recommend it. GFF readers can order it through your web site. The aim of my article was not to recommend certain threads for certain uses, but to raise a discussion about the very confusing standards used to label threads mainly sparked by my own frustration and confusion.


Martin, I disagree w...

Martin, I disagree with some of your methods and conclusions. First, you cannot measure the diameter of thread (except for the very specific case of a round monofilament thread). That's because thread isn't round, and so has no "diameter." Moreover, as you partially recognised, in many instances the act of measuring affects the measurement. In statistical terms your measurements are reliable but not valid. In other words, you are consistently and systematically measuring the wrong, or a meaningless, value.
Denier, or decitex (a similar measurement in general commercial use), is the only way to reasonable assess thread "sizes" AND only when comparing threads of like material and construction. On the other hand, breaking strength can be reliably measured and so is a legitimate basis for thread comparisons. Of course one understands that even saying something is constructed of polyester filaments doesn't guarantee identical properties as the filaments are subject to scores of manufacturing tolerances. If you had total information about a thread's properties (such as specific gravity among others), one could calculate a cross-sectional area and hence an "effective diameter," for whatever that would be worth. For example, given the identical "effective diameter," a 1-ply thread would tie differently from a 2-ply one, a twisted one differently from a flat one, a waxed one differently from unwaxed; etc..
The "ought" system has been completely adulterated and so, at least on a global scale, is now completely meaningless. At one time responsible respoolers kept the numbers at least relative. But, because there was no standard, not even a definition, some respooling brands simply invented numbers to make their thread appear more attractive. In some cases they even gave the same thread different numbers to make it appear that they sold a range of sizes.
Is there any hope for standardization of fly-tying thread designations? No. It's a small market and the wild-west of "ought" designations suits some players. If you want to see a struggle to get some consistency in a "similar" angling product, one need only look at the case of monofilament line in the European market (relative to fly-tying thread a mega-market). Even in monos, there is variation in "roundness" (some are oval) and so pure diameter measurements aren't always possible.
What is true is that one should select thread based on the use (my booklet, How to Choose and Use Fly-tying Thread, can help here) and whether or not you trust the brand to provide accurate information. So while I disagree with what you say "will work" - besides, attempting to put that much information on a spool label is completely impractical - I applaud your attempt to try and clarify some of the issues.

Paul Marriner

Outdoor Writing & Photography. Owner: Gale's End Press. Member: Outdoor Writers of Canada. Author of: (new) Mahone Bay Mornings, Modern Atlantic Salmon Flies, 2nd Edition; A Compendium of Canadian Fly Patterns (co-author); Stillwater Fly Fishing: Tools & Tactics; How to Choose & Use Fly-tying Thread; Atlantic Salmon: A Fly Fishing Reference (OP); Modern Atlantic Salmon Flies (OP); Miramichi River Journal (OP); Ausable River Journal, and Atlantic Salmon (OP).

good article...

good article


Log in or register to pre-fill name on comments, add videos, user pictures and more.
Read more about why you should register.

Since you got this far …

The GFF money box

… I have a small favor to ask.

Long story short

Support the Global FlyFisher through several different channels, including PayPal.

Long story longer

The Global FlyFisher has been online since the mid-90's and has been free to access for everybody since day one – and will stay free for as long as I run it.
But that doesn't mean that it's free to run.
It costs money to drive a large site like this.

See more details about what you can do to help in this blog post.