The following is a report by Jim Winn on some experimental archaeology he preformed and presented to the limited audience of Knappers-R-Us in January 2003. I was extremely impressed with his research and report, and I ask him if I might add it to my WEB page. Jim is an accomplished knapper who lives in Los Angeles and works for the city as a Operations Supervisor at the main waste water treatment plant. You can contact him at Knappers-R-Us is a restricted mail list populated by knappers from around the world and a few crazy non-knappers like me.

Flakes from Flat Surfaces

by Jim Winn
January 27, 2003

I did some more test flaking on obsidian slabs to observe the shape of the flakes removed. Flakes were removed with pressure using an Ishi stick with a copper tip and another with an antler tip on the 1st test. And a 2nd test was performed with percussion using both copper and antler. All of the slabs were photographed afterward showing the slabs as well as the tools used.

The following picture shows the results of the first test done with pressure flaking.

Two slabs that were pressure flaked using an antler tipped Ishi stick are shown on the left. Two more slabs that were pressure flaked using a copper tipped Ishi stick are shown on the right. The slotted rubber/leather pad shown at the bottom was used to support the slabs in the left hand and allowed the flake to travel without interference. The flakes removed with both antler and copper were all elliptical in shape and only slightly longer than their width. I was unable to significantly increase the length to width ratio using copper or antler regardless of the direction or the amount of applied force.

This next picture shows the results of the 2nd test using percussion flaking.

The two slabs on the left were percussion flaked using antler. Those on the right were percussion flaked using solid copper. In both cases all of the flakes removed were elliptical in shape and similar to those removed with pressure flaking. Some of the flakes were slightly wider than their length which was mainly a result of the edge angle (platform) being nearly 90 degrees. These flakes feathered out along the sides and terminated a little short at the distal end. The lower left flake has a platform closer to about 60 degrees and the flake was nearly round in shape.

OK, so what does all this prove? Well, it seems that the type of tool used (copper or antler) as well as the method of removal (percussion or antler) have little impact on the shape of the flake removed! There may be some minor differences that I could not detect, but they are insignificant. Of course, this only applies to a flat surface which provides ideal conditions and repeatability. Is this knowledge of value when removing flakes from an irregular surface, such as a biface? I think it is. That is just my opinion, but the whole purpose of performing a test like this is to gain insight into what can be expected under typical conditions that are not perfect (the surfaces we encounter in bi-facial reduction). I'm going to make the assumption that the shape of a flake removed from any surface has little to do with the type of tool used (copper or antler) or the method of removal (percussion or pressure). Instead, I believe that the primary determining factors in the shape of any flake removed from any surface are primarily a result of the following.

  1. The shape of the surface where the flake is to be removed. The flake will follow ridges if they exist and it will fan out on flat surfaces.
  2. The point at which the pressure is applied (when pressure flaking) or the point of impact (when percussion flaking). This will determine the initial thickness of the flake as it begins its travel.
  3. The amount of applied force. This factor affects the mass and shape of the flake removed. More force is required to initiate fracture as the platform depth is increased (or the depth below the surface at which the pressure flaker makes contact).
  4. The direction of applied force, including both the depth (or downward) direction as well as the direction across the face of the bi-face whether it be at 90 degrees or diagonally.

All of these factors directly affect the final shape of the flake detached. The type of material used to initiate the fracture (copper or antler) is mostly a matter of personal preference. The copper will require a stronger more heavily ground platform than the softer antler. And, if a massive flake is to be removed, more force can be applied using percussion instead of pressure due to the strength limitations of the person applying the pressure. If a mechanical levering device is used to apply pressure (such as in fluting), the flake removed should resemble that done by percussion. Again, the flake removed is primarily a result of factors 1 thru 4 above.

The bottom line is the stone simply follows the laws of physics and reacts according to the forces being applied to it. You cannot force a flake on a flat surface to be much longer than its width. You can, however, cause the flake to terminate short by applying pressure (such as with you fingers) to the surface of the flake as it is traveling. This usually results in a step fracture. I have heard of some knappers who are able to extend flake length by applying pressure along either side of the flake as it travels but I am not familiar with this technique and did not try it. My guess is that the flake might terminate short along the sides but continue straight ahead. It would be interesting to see what others are able to do using this or other techniques, and I'd really like to hear from them.

One final picture:

The above photo shows the results of flakes removed sequentially along the edge of a piece of opaque flat white glass with a flat surface. The first flake removed was elliptical. Each flake removed after the first followed the ridge created by the preceeding flake scar. In this case, the overlapping scars are 2" long by 3/8" wide and traveled to the opposite edge. They are 5 times longer than their width. This is the type of flaking observed on Egyptian Gerzean knives and demonstrates how length to width ratio in flake removal can be controlled and extended by preparing an ideal ridge for each flake to follow.

OK, that's it! I know a lot of folks may not be in agreement with my assumptions or conclusions, but I invite others to perform similar tests and report their results so we can all learn from one another and improve our skills.

Return to Paleoindian & Other Archaeological Stuff