Be sure to visit the Hand Tool Headlines section - scores of my favorite woodworking blogs in one place.  Also, take note of Norse Woodsmith's latest feature, an Online Store, which contains only products I personally recommend.  It is secure and safe, and is powered by Amazon.


Sloped Gullets: The Finer Points of Sharpening

Click to enlarge

 I find it hard, even after many years, to carry on a conversation about "sloping gullets" with at least a little chuckle.   Not at how it refers to a certain method of sharpening a hand saw mind you - its more that every time I hear the phrase "sloping gullet", I can't help but think of some sort of deformed fish...

That, and when researching the origins of this icthyological pursuit, I ran across a reference to a photo of a WWII era front-line French infantry "installation" (actually a shack the infantrymen had set up as a bar) called "L'Auberge des Gosiers en Pente" - or "The Inn of the Sloping Gullets" - that is to say, always thirsty... (from "The French in love and war: popular culture in the era of the World Wars" By Charles Rearick)

When sharpening a saw, there are several angles you are concerned with. The terms associated with these angles which are most important to this conversation include rake, fleam, and of course - slope (as shown in the graphic above -you can click on any of the images to see a larger, clearer version).  I'll try not to go into too heavy technical detail on saw sharpening as that's another subject, and it's been well covered by others... as well as myself.

 What I would like to do is discuss just what sloping gullets can - and cannot - do for you.  I've seen several discussions on the web that indicate there is some confusion as just how it is they work....

Click to enlarge

First, I'd like to dispel a myth, one that inspired this writing - and that is that you can have sloping gullets on a rip-filed saw.  That is simply not the case.  Here is a saw (pictured at right) that was presented to me, the sender claiming it was filed rip with sloping gullets:

Looking closely at the teeth you'll see they really are not filed rip at all - the effect of sloping gullets has essentially made them all filed to what I would call an aggressive crosscut pattern - and they are no longer "rip" teeth.

 Ok, perhaps a quick review is in order of just what the term "rip" filed means.

Click to enlargeLooking at a comparison of rip and cross-cut filed teeth on the left, you can see that the rip teeth are flat across the top (to chisel through the wood with the grain), vs. the crosscut teeth that all come to a point (to slice through the wood at a perpendicular angle to the grain).  If you look at the photo of the saw above, it's easy to see that the teeth resemble the crosscut profile at left far more than the rip profile.

So, how is that then?  It's because the act of filing using a slope on the file files a point on the teeth.  Think about it - the upward angle of the file will naturally sharpen the tooth to a point...  To say it another way , the very act of sloping the file while sharpening is to file the saw to a crosscut profile.

Click to enlarge

To be fair, there are some instances where you do put a slight slope on rip teeth - but it's 5 degrees or less, and then you get what my dad would have called a "modified" rip profile, which is one that you use where you have some wild grain...  but I'm getting ahead of myself, I'll touch more on that in a later post.

Also - for purposes of clarity - fleam (for the purposed of this post) is the angle produced on the tooth of the saw and can be produced by angling the file either to the left or right, or by sloping the file up or down (sloped gullets).  Just so I keep them straight, when I refer to fleam (or flat fleam) in this conversation I will only be referring to the angling of the file (while sharpening) from left to right;  I will use the term slope when referring to the angle of the file in the up/down direction.  In reality, it is all fleam in the end...  or should I say sawdust in the wind?

 The exact origins of the sloped gullet are lost to history, though perhaps we can deduce a few things about it's beginnings.  The term "sloped gullets" itself I think is a more recent moniker, having been coined - or at least more commonly used - in the last several years than previously.

It's my belief that the sloped gullet was the first iteration of the cross-cut saw, as many (very - like 1700's or so and older) if not most old saws were commonly filed rip...  some experimenting sawyer back in the day discovered that the addition of an angle - or slope - to the filing made saws perform better when sawing across the grain.  Further experimentation - in my opinion - then led to the development of fleam and eventually the crosscut saw.

Click to enlarge

 In "The Art of Saw-Filing: Scientifically Treated and Explained on Philosophical Principles" By Henry Wells Holly, first published in 1864 or so, sharpening saws is gone into great detail, though exact rake angles are not discussed.  The illustration at left (illustrations from this book were later used by the Disston company in their reference manuals) clearly shows an approach that uses a sloped gullet in combination with an angle to the file to produce fleam.  So the practice was well established by that time.

  Yet not twenty years later (1880) Robert Grimshaw's seminal treatise on saws, Saws:  The History, Development, Action, Classification, and Comparisons of Saws of All Kinds (aka Grimshaw on Saws) doesn't show or refer to a sloped gullet at all when referencing hand saws of the kind we are discussing; some slope is obvious in the diagrams of pruning saws and logging saws, so he was obviously aware of it .    But none when referring to a hand saw destined for dried hardwood.

 Now granted, Grimshaw's book tends more towards the industrial than Holly's, and therein may lay our explanation of what happened in those twenty years.  Industrialization..

Saw-making had grown along with the industrial revolution, and benefitted greatly at the hands of the very revolution that would one day spell it's near doom.  One of the advents of this was the development of saw filers - and they don't lend themselves well to sloping gullets  My hypothesis is it's possible that the flat-angle fleam we all know and love grew out of this mechanization. 

On the retail end, while most saws came flat fleamed, there were a few higher end saws back in the day that were sharpened with slope.  A case in point woul be the Disston Acme line of saws - that came filed from the factory with factory with sloped gullets.  In the case of the acme saws, they also came without set...  The Disston Acme saws were a special case, in my opinion.  First, he steel was extra hard, so did not lend itself to being set. But because it was extra hard, the steel could hold a sharper edge... Next the plate was heavily tapered to make up the difference for having no set.   The extra depth of the sloped gullet could only help.  With a regular saw, the set of the saw would make up the difference easily.  So to me, the Acme was a saw specially made that benefitted from the sloped filing, and outside of the scope of this writing - which I'm aiming more towards the regular user with a regular saw.  The great majority of saws sold after 1880 or so were filed with a flat fleam from the factory.

Click to enlarge

So - what are the effective differences between a sloped gullet and a flat-filed fleam?  On the top of the illustration at left is a fleam-filed saw.  The bottom illustrates a sloped-gullet saw filing.  The most noticeable difference is the depth of the vally on the side being filed - it's obviously deeper on the sloped saw.

As far as the flat-fleam filing goes, once you reach a certain angle it starts to become exponentially more difficult to sharpen simply because of the angle of the file in relation to the saw plate.   Adding slope allows you compensate for that difficulty and achieve a greater fleam angle when used in concert with a standard fleam angle attack. 

Because of it's angle, an additional benefit to the sloping gullet is that it also makes that slightly larger valley between teeth to carry and the theory is this will evacuate more sawdust out of the cut. 

The end result is that it allows for a sharper tooth that cuts more cleanly and one that runs cleaner because of how it can handle sawdust more efficiently. At least in theory, anyway...

In my own experience - the people I've known that used logging saws almost always used sloped gullets.  Likely - I think - it's due to their experiences sharpening their larger saws.  Larger toothed saws seem to benefit more from the sharper angle than smaller toothed saws in my experience.  Most old-school carpenters and cabinet makers that I knew used the flat fleam method of sharpening.

 Does that mean that one method is inherently better than the other or vise-versa?  So what's not to love?  There's a couple things to consider, as it's not all that black and whiteClick to enlarge.

The first thing one needs to understand is about just how fine of a point you can put on a hand saw.  This relates to all styles of sharpening...The steel in a hand saw is much softer than the steel of say, a chisel.  It simply will not hold as fine of a point as a chisel will.  The very tip of the points when filed to greater angles will fail quicker, hence the saw will dull more quickly, and require  that the saw be sharpened much, much more often.  Or, optionally not to use sharp angles, just use slope for the sake of using slope - perhaps to take advantage of the deeper gullet. More on that in a bit.

Such sharp angles are probably OK, if you sharpen your own saws and are good enough at it, but not if you are going to send  the  thing out to be sharpened that often, you'd go broke quickly.  Which leads me to my second consideration - I think it depends a lot on the skill of the saw filer.  Just above is a photo of some sloped gullets as filed by Mark Harrell of Bad-Axe Toolworks.  Masterful work!  There's no doubt he can make them work ...  Now I'm no where near as accomplished at filing as Mark is, but still - I'm not a complete slouch.   Most of the time anyway... Click to enlarge

 I decided to do a little experimenting of my own.  I took a 20",  eleven point Richardson panel saw and a 28" seven point Disston D-8 (pictured below) and filed them each with sloping gullets. Performance-wise, the end result was the saws worked as good or perhaps even better than any previous results I've had.   It did seem the seven point saw had the most distinct improvement.  Of course, that may have been "an undigested bit of beef, a blot of mustard, a crumb of cheese, a fragment of an underdone potato..." (from Dickens - A Christmas Carol).  I can't say with authority it was the case.

 The first issue I had was with setting up and maintaining the correct angle, all while keeping the teeth consistently sized.  It's not that the angle was so difficult to figure out, but it did require a bit of practice - I had do an extra pass over the saw (one each one, none-the-less) to re-set the angle so it was appropriate after deciding I the results I was getting were not satisfactory.  It worked - and I suspect with practice it would come easier, as is usually the case.  It's not like I've had the greatest amount of practice lately either - so I'll admit my skills have gotten like my saws - a bit rusty. 

The consistently sized teeth were more troublesome.   The deeper gullet on one side sets up a bit of an optical illusion, making every other tooth look larger than the one next to it.   I've gotten fairly good at shaping teeth by sight over the years - but when shaping the sloped gullets I found myself bending over to look at the saw straight on to make sure that I was keeping the teeth consistently sized, and even to go through re-shaping the teeth on the larger saw one time to fix where I screwed it up.  Again, I suspect practice would remedy the problems I had - but it did take me quite a bit of work to get it right the first time.  Not to mention using up an entire file just re-shaping that 7 point animal - there was a lot of metal to remove to convert it to a sloping gullet!  The next time it gets sharpened that won't be the case, of course...

My verdict?  I think that a sloping gullet can produce a sharper crosscut saw.  The caveat is that it requires a skilled filer to do it.  For 99% of the users out there, who are averse to sharpening their own saws to begin with or simply don't have the time to practice the craft daily as was done back in the day, the extra angles introduced and the difficulty of maintaining the consistent tooth sizes simply isn't worth the effort.  I was able to sharpen my saws with a sloped gullet, and perhaps did notice a slight improvement in their performance - but not enough of a difference that I could say with authority that it was entirely due to the sloped gullet.  It might have simply been the fact it was freshly sharpened.  It's hard to quantify these things in the real world when writing - but for my next sharpening I will be returning to the simpler flat fleam, if that says anything.  It's just easier. 

bibliography and special thanks:


The Art of Saw-Filing: Scientifically Treated and Explained on Philosophical Principles By Henry Wells Holly

First published in 1864, still in publication
ISBN-10: 1408667150
ISBN-13: 978-1408667156

Available at Google books here:

You can download a PDF copy from Norse Woodsmith here: 


Grimshaw on Saws

ISBN-10: 096180887X
ISBN-13: 978-0961808877

Available on Google books here:


One of the best charts I've seen to explain the angles of saw sharpening is on Joel's blog at

Tools For Working Wood


Mike Wenzloff also goes through a great discussion on fleam here:

Fleam Theory by Mike Wenzloff


Brent Beach has done an exhaustive study on sloping gullets here that you should take the time to review:

Brent Beach's page on Sloping Gullets


Erik von Sneidern's Disstonian Institute - a great resourse for all things Disston, including the Acme line of saws;


And finally, special thanks to Marv Werner, Mike Wenzloff, Mark Harrell, Andrew Lunn, and all the other saw-makers out there with whom I've traded emails.  Thanks, guys.


This article is subject to revision...  as I'm still always learning things.



In Leonard Lee's book, he says to drop the file handle so that its angle to the horizontal plane is roughly equal to the fleam angle.

It took me a while to get the hang of this for the reason that you mentioned--it was very difficult to see if you were being consistent about taking off metal because the reflections from the newly-cut metal between the teeth played tricks on your eyes. Eventually I got the hang of it by doing three things:

First, I started trusting the tops of the teeth more. Once the flat from jointing was gone, it was time to stop.

Second, for jobs that require serious reshaping, I take more than one pass across the saw; I don't try to shape an entire tooth at once. This has helped me a lot in cutting new teeth, too.

Third, if I really want to see how I'm doing, I try to shade the saw so that there are no reflections and look at it against a light background.



Hey, Brian!

"In Leonard Lee's book, he says to drop the file handle so that its angle to the horizontal plane is roughly equal to the fleam angle."

That would pretty much correspond with Holly's book, if you look at the diagram above.

I agree totally that - with practice - its certainly feasible to sharpen a saw with sloped gullets. Like anything, if you do it enough it will come. 

Thanks for commenting!



Something that also occurred to me on this subject is that I don't really know the exact angle at which I slope the file. It's not a lot (I don't use a tremendous amount of fleam, either). In general, the triangle profile of the file roughly fills out a freshly-sharpened tooth, whereas with no slope, there's a little gap between the file and the bottom of the gullet until you file the whole thing out. There's something about the shape of the tooth that results, but I can't really put my finger on what to say about it.

As far as practicing, I just went with whatever Lee wrote, and started out sloping the gullets. I guess I assumed that there was no other way, or at least, I didn't want to read any more web sites other than Taran's. I recall it took about three or four tries before I got teeth that looked halfway reasonable. Once I got the hang of it, it wasn't much more difficult than filing rip teeth, but I was also very silly and started out filing crosscut teeth instead of learning rip teeth first.



Hi Brian!

One of the things I like about Leonard Lee is he is one of those types of authors that can speak plainly, matter-of-factly, with a good dose of common sense (Charles Hayward is another) - his book on sharpening is still one of the best out there, if still not the best.  I guess I'm trying to agree with you and also endorse his book, I don't think you can go wrong by following his teachings. 

For those out there that do prefer to slope their filing in such a way, I think there's a great deal of historical and practical evidence that shows it's not a folly pursuit - and I commend those who can and do, as it shows a level of competence in sharpening that most do not pursue.  The simple fact a craftsman sharpens his own saws speaks volumes to me about what kind of product that person is trying to put out, and to the amount of dedication they have to learn their craft.  Kudos to all that do.

So many out there today send their saws out to be sharpened, and thats what seems a waste to me...  There's nothing that difficult about it and if you use saws you need to do it often enough that sending it out become expensive and counterproductive.


P.S.  Brian's got a post on sloped gullets on his website, Galoototron.  His excellent posts can also be found on the Norse Woodsmith blog feed...


Fantastic article Leif! I agree with everything you've written here! I sharpen all my own saws and have for some time, but not daily, or even weekly. Add to that the fact that most of my saws are filed with rip teeth (long rip saws, tenon saws, dovetail saws...all rip teeth) and that the saw cut is almost never the finished show cut in any of my work, I just can't see the point or benefit of the extra care required to sharpen with sloping gullets.

I own two crosscut saws; one 26" #7 and a 14" sash saw I built. I can't see a need for any more saws with crosscut teeth, and hence I have no need for sloping gullets. My crosscuts are basically never seen in a finished piece; tenon shoulders are knifed, ends of tenons are burried in a mortise (for blind M&T) or planed flush (for through tenons), and end grain that does show in a finished piece is either planed or molded in some way. So for the most part, I'm not concerned with how clean the cut is, I just want the saw to cut fast.

I think this is likely why there is little historical evidence that crosscut teeth existed as well. They simply weren't really necessary. Almost all crosscuts are hidden from view, cleaned up in some downstream process, or covered up with molding. I've used my rip filed tenon saw to crosscut countless times (after knifing the cut line) and never felt that it did an inferior job to the crosscut filed sash saw, because I wasn't interested in the surface the saw left behind. If the surface left by the saw was important, it needed to be cleaned up regardless of whether the saw used was a rip or crosscut saw.

For someone who does their own, semi-frequent sharpening, it's just easier and more consistent to file flat fleam.

Bob Rozaieski


Hi Bob!

You know that by 1864, Holly was using sloped gullets, so the practice was definitely well established by then.  The early 19th and late 18th centuries saw a lot of refinement in hand tools such as saws, so it's not beyond imagination to consider that this refinement came about en-masse during this period.  Another refinement was the addition of set - or should I say refinement?  Set had been around at least since Roman times, if not Viking (google Mästemyr tools) and Egyptian, but was refined to be per individual tooth sometime before the advent of the crosscut saw.

You know I I had more fun researching the historical aspects of this more than anything.  It's fascinating to see the developments and refinements of these tools.




Hi Leif!

Thank you for doing this article on sloping gullets.  I love reading stuff I agree with.

Just a few thoughts on rip teeth.  Rip teeth are always referred to as small chisels.  However, when one pictures what the rip tooth is actually doing down in the kerf, it is not chiseling, it is scraping. Even with zero rake, it is still scraping.  To make it act as a chisel, negative rake would have to be employed. Talk about grabby.  It would be like trying to push a Japanese pull saw.

Relaxing the rake angle on a rip tooth to say 10 to 15 degrees will allow it to scrape crossgrain without tearing the fibers as much as a rip tooth that is normally between 4 and 8 degrees for example. Also it is very possible to crosscut with a rip saw if the teeth are quite small, like maybe 13, 14PPI or smaller, more like we often see on backsaws.  Maybe it's just me, but I like a smooth cutting saw, both by feel and the end result. 

I once sold a big O 5PPI rip saw to a guy on eBay.  I had freshly filed it and tested it and it cut like a rip saw should.  When he got the saw, he emailed me and said he couldn't get the saw to cut. He was quite irate. He said it just wouldn't start a cut.  I emailed him back and ask him what kind of cut he was trying to make. I suspected what he was doing, but thought I should ask.  He answered back and explained that he was trying to cut a board cross grain. I then explained the difference between a rip tooth and a crosscut tooth.  I had filed the saw with an 8 degree rake.  I suppose if it had been filed at say a 15 degree rake, he would have gotten the thing to cut cross grain, but boy would it have been a rough cut.  And when ripping with the grain, it would have cut like it was super dull. 

The point is, there is good reason for filing rip teeth for scraping or ripping and good reason for filing with what you call a "flat fleam" tooth for crosscutting.  I'm with you regarding sloped filing. Slope filing a rip tooth simply compromises its ability to scrape.  That same tooth is also compormised if expected to cut cross grain because usually when adding fleam to a rip tooth, less fleam is commonly used compared to the amount of fleam on a crosscut tooth. 

It is just good common sense, as you point out, filing crosscut teeth with sloped gullets makes the tooth more pointed and it will dull faster, especially in hardwoods.  Probably do ok in soft wood such as pine for example, but what about those big knots we know pine to have.

Your articles are always well written and thought out and informative. Much appreciated.




Hi Marv!

Thanks for commenting.  You bring up some good points, and I think your experiences highlight some of the ignorance - for lack of a better term - of sharpening saws.  It's my hope that these types of articles and conversations can help demystify the process, at least for some. 

Just for the record, I don't believe that a sloped filing is a bad thing - I actually believe tthat (when done properly) it is a totally valid way to sharpen a saw.   My biggest beef has been with the misunderstanding of how saw teeth work and I think your experiences bears that out.  My experience with sloped gullets is purely personal, but what I've found is that the additional angles are difficult for me to work with - perhaps that's just because I've been using a "flat fleam" for so long, working the angles with a slope is quite alien to me, and introduces too many angles, which would be OK, I guess, if it produced a better cut - I didn't really find that to be the case, at least when we are discussing my skills at sharpening saws.  Obviously, others may have a different view and skill set.

Thanks again Marv!  Keep up the good work!



Great post Leif, thank you very much. I think I little disagree on a couple of points, but I have to read it again to be sure. Nevertheless, I share your conclusions and I think it's a very good article, thanks again. Auguste P.S. Never tried a mixed configuration as for example 5 teeth rip and 2 crosscut, 5 rip 2 cross and so on?