Learn to bake sourdough bread - the best bread in the world!

Experiments with Autolyse (Autolysis) # 1

Experiments with Autolyse (Autolysis)

Click here for a PDF version of this article.

Click here for a printable PDF version with no photos.

An autolyse stage, as used in the making of bread, was first introduced by Professor Raymond Calvel. He noticed that when the dough was given a chance to hydrate when first mixed without the addition of salt (or other ingredients besides flour and water), it produced a superior outcome in the final bread.

The word Autolyse (sometimes pronounced… auto-leese)(dictionary.com  pronounces it as [awt-l-ahyz] ) means to consume self or digest self.  au·tol·y·sis  [aw-tol-uh-sis]  The breakdown of plant or animal tissue by the action of enzymes contained in the tissue affected; self-digestion.(dictionary.com)


It seems like an odd meaning, but it actually does make sense. When water and flour are first mixed together, an enzyme called Protease is activated (other enzymes are also activated as well). Protease is a protein. Protease goes after and breaks down other proteins in the dough, notably proteins called Glutenin and Gliadin which help to make up the gluten network in dough. When Protease does it’s work on these other proteins, it is actually breaking them apart or breaking them down, digesting them. Hence the label of self-digesting . We have a protein munching on another protein, kinda cannabilistic, don’t you think?


The Protease enzyme, breaks apart and realigns protein strands, helping to promote the formation of newly aligned gluten bonding. At the same time it also weakens the bonds so that the dough is not too tough/rubbery. Dough needs a certain amount of extensibility which means it can stretch out. You don’t want dough so tough that you cannot handle it or stretch it. It needs to stretch as well as have a certain amount of bounce back to it (elasticity).


The gluten protein Gliadin is responsible for the gluten strands being able to extend and the other main protein, Glutenin is responsible for allowing the dough to bounce back or be elastic. Too much of either one and you have a problem with dough that is too weak or dough that is too strong (information from Bread Science- Emily Buehler).

Autolysis is used on dough for different reasons and in different situations. It depends on the outcome you desire. A very high protein flour or a strong flour will be enhanced by an autolyse period as it will allow the dough to be more extensible (stretch more easily and get those large holes). A very weak flour, pastry flour, cake flour, some AP flours will not benefit from an extended autolysis. However a short autolyse period can be used to hydrate the gluten, help align gluten strands and cut down on the amount of mixing time.


Autolyse Experiment: (see formula below)

  • Mix up three pounds of dough at 66% hydration.
  • Divide the dough into three 16 oz/453g pieces.
  • Add salt (.2 oz/5g) immediately to dough # 1.
  • Let dough #2 Autolyse for 30 minutes, then add salt (.2oz/5g).
  • For dough # 3 Autolyse for 2 hours and then add the same amount of salt.
  • Allow all dough pieces to bulk ferment for six hours folding a total of three times (covered and at room temperature). 

Interesting things happened with this experiment.   Right after mixing I added salt to dough number one. The salt was easy to incorporate, there was no real gluten bonding yet and the dough was easily mixed and mushy feeling. After the 30 minutes autolyse period for dough number two, the salt was also easy to mix in, the dough felt sticky and was mushy like the first dough.


The surprise came with dough number three which was autolysed for two hours. The dough felt silky, smooth and was noticeably cohesive with a lot of gluten bonding, it was full of activity and bubbles. The salt was not as easy to incorporate into dough number three, but it was a small amount of dough, so it wasn’t overly difficult. All of the dough was folded three times during the six hour bulk ferment with no other kneading taking place except to incorporate the salt.

Loaf # 2 in the middle. Loaf # 3 on the right side

After the six hour bulk ferment, dough number one, which was salted right away had a nice feel and was moderately filled with bubbles. Dough number two which was autolysed for 30 minutes, was sticky and bubbly but was hard to handle and slumped during preshaping. Dough number three really surprised me, it felt better than dough number one, was filled with bubbles and sat up higher after being preshaped.

I was surprised enough that I will now autolyse for two hours before adding salt. My technical administrator on our forum gave me the idea for a two hour increment. I have autolysed overnight, multiple days and even for an hour, but not two hours. He was reading that two hours was an optimum time for autolysis. I have only done this experiment once, but it gave promising results for a two hour autolyse period. The dough used was from the same batch and the same room temperature was used. The only difference was the time when I added the salt to the dough.


After the six hour bulk ferment I preshaped, then final shaped and refrigerated(covered) all three loaves. In the morning I took out all three loaves at once, I was planning to bake them at the same time but I noticed right away that even though they were the same weight, they were different sizes and at different levels of being proofed. I ended baking loaf three first it was the largest(by volume not by weight) and most active, bubbly loaf throughout this experiment.

I baked loaf number two next, it was denser and ended up smaller than loaf number one. Last I baked loaf number one, the loaf which had the salt mixed in right at the beginning. It was also denser, took the longest to proof and did not get the oven spring and volume that loaf number three did. So I ended up baking the loaves backwards from loaf number three to loaf number one.


Baking was accomplished with a fully proofed loaf, which was slashed, sprayed with water and covered with a roasting pan lid for the first 18 minutes in a preheated oven at 450F degrees using a baking stone. After the 18 minutes, the roasting lid was removed and the bread was allowed to brown up and crisp for 12 minutes more, turning once. Preheat the lid before using for the next loaf.


Of the three loaves, loaf number three which was autolysed for two hours was more active and bubbly throughout the experiment at all stages, had a nice firm hand feel, it wasn’t as sticky as the loaf which had only been autolysed for 30 minutes and it lost the most weight during baking (a lightweight fluffy loaf).

Loaf three (two hour autolyse) weighed 16 oz when I baked it(they all started out at 16 oz) and was 13.4 ounces after baking. Loaf number two (30 minute autolyse) weighed 13.6 ounces after baking. Loaf number one (no autolyse) weighed 13.8 ounces after baking.

I am guessing loaf number three lost so much more weight because it had a much more open crumb and a greater volume so it was able to let off more water vapor during baking.  The volume of loaf number three is so much greater than the other two that it is easy to identify which loaf it is in the pictures.

Loaf # 3

Loaf # 3 Crumb

I would like you to know that this was not the expected outcome. The outcome I had expected was that the bread autolysed for 30 minutes would have been the superior loaf, with the two hour loaf being more sticky and slumpy (that a word?, if not, it is now) and a lower final volume.  The bread which had no autolyse did not surprise me. I have done extensive work with long autolysed doughs, but they were usually overnight at room temperature or days at cold temperatures. I have had good results with the long autolysed doughs. However, the shorter versions seem to be in their own category.

Loaf # 2

Loaf # 2

Loaf # 2 Crumb

Loaf # 2 Crumb


I will redo the experiment a few more times to see if the results are consistent. Please do the experiment yourself and report back. It would be nice to get some feedback from other bakers.


I believe the dough using a two hour autolyse, not only had the Protease working, but all of the enzymes and yeasts were able to get a good jumpstart before the salt was allowed to inhibit their action. I am guessing that is why the loaf had so much activity and was more bubbly and resulted in a larger volume. I also have the thought that the salt may not have been incorporated into loaf three as well as the first two loaves and that may have given it the advantage as far as volume goes.

Loaf # 3 and # 2

Loaf # 3 and # 2

This experiment will have different results on whole grain breads which have a weaker gluten bonding and would benefit from some inhibition of enzyme activity(adding some salt early on) or a short autolyse.


Here is the formula for making up the three pound batch of dough:

  • 11 oz/311g of starter @ 100% hydration, the starter had been fed 24 hours and then eight hours before using.
  • 14 oz/396g water
  • 24 oz/680g bread flour
  • Salt is used at .2/5g per loaf when the time comes to incorporate it.

This makes 3 lbs 1 oz / 1389g of dough at 66% hydration

Mix up all ingredients except salt. Weigh out three 16 oz/453g  pieces of dough and then follow the directions already given above for timing and adding salt.


Loaf # 3

Loaf # 3

I have so often heard people say they do not bother to autolyse because it is a waste of time or the salt is too hard to incorporate once the dough is mixed. I kept the dough a wettish 66% to help with salt incorporation. I have also used a slurry to add salt to a mixed dough later, like in the formula for Big Bear’s Bread (one of my very best breads). You can hold back part of the flour and add it after autolysis with the salt. So there are ways to incorporate the salt into mixed dough. Regardless, the results for a loaf autolysed for two hours was such a superior outcome, that I am convinced.


Edited to add:  I have had much feedback on this experiment and bakers wanting to know about autolysis without added levain or yeast. My next experiment will be with just the flour and water vs. flour/water and levain. Stay tuned….



Similar posts
  • Experiments With Freezing DoughI’ve been experimenting with freezing dough. My first experiment was with a two loaf batch of seeded dough.  After bulk ferment, I put the dough into the fridge to cool it down before shaping it. Chilling the dough minimizes excessive rising before it freezes. After refrigerating the dough for two hours, I shaped the loaves [...]
  • Slowing Down FermentationMost likely ambient temperatures are warmer now that it’s summer here in the northern hemisphere. Your starter and dough can over-ferment easily in hotter weather. Here are some ideas for slowing down your starter and dough: Things that help slow down a starter [...]
  • Summertime Wheat BreadI had some Yecora Rojo berries wheat left from the bag of wheat berries that Guy Frenkel sent me so I thought I'd use them up. I wanted to do something a bit different too, so I added a new ingredient that I hadn't tried before.It's not quite summer yet but it will be really [...]
  • Fresh Ground Flour and Home MillingI have two home mills, the Blendtec Kitchen Mill and the Mockmill, well maybe three if you count the cheapo coffee grinder I used in the past before I got the Mockmill. The Blendtec Kitchen Mill has been a mainstay in my kitchen for years. I’m actually on my second one as the first one ended [...]
  • Sweet and Sour SourdoughIt’s that time of year when lots of folks around the world are baking. We are heading into spring in this hemisphere and on the other side of the world they are heading into fall. Often in the summer many people give up baking for a while, so during summer half the world isn’t as [...]


  1. KellyKelly
    January 30, 2017    

    Thank you for sharing this great experiment! I just finished a 2 hour autolyse thanks to reading this, and I’m very excited about the results! I have a question for you: Is there a reason you’re only doing a 6 hour bulk ferment? I’ve been doing a 10-12 hour bulk ferment on the counter top before putting my dough in the refrigerator overnight. I’m wanting the overall time from beginning of autolyse to baking to be at least 24 hours. My bread is turning out well, but I’m always looking to improve it. Is there a benefit to a shorter bulk ferment? Thanks!

  2. John PleasantsJohn Pleasants
    December 7, 2016    

    Hi, John from Australia, Head Baker at Firebrand in Melbourne, baker of twenty years. Thanks for your very interesting information. Lovely to read, very informative and quite enjoyed the science involved. Keep up the good work!

  3. SolveigSolveig
    November 23, 2016    

    Have you ever done autolyse for an extended period, day 3 days? Resources I find say up to 5 hours.
    I’ve had same experience with 2 hour autolyse dough – incredibly smooth and silky. I also use a small tangzhong – 25gr flour to 1/2 cup water for a small dough, adding this to the water, then flour, then rest for autolyse.

    • November 24, 2016    

      Yes, I’ve done autolyse for up to six hours at room temperature and cold autolyse for six day or more but you need to do a microlevain inoculation for very long fermented dough.

  4. MartinMartin
    October 30, 2016    

    This is so interesting. I just started reading the artices on your web site and it’s such a treasure of information. You did all the experiments I always wanted to, but was too lazy to actually do it. I will pay much more attention to the autolyse the next times I’m baking.

    I have a book about baking that also includes some chapters about dough properties and the effects of different ingredients. It said that salt improves gluten strength because it helps unfold the glutenin molecules so they more readily form crosslinks. They also included photographs of three loaves with the same formula, but different salt content: 0%, 2% and 3%. The difference wasn’t huge, but the 2% had the most volume, followed by the 3% loaf.

    Knowing this, I would have expected loaf #1 of your experiment to have the biggest volume because it has the strongest gluten network. But as it turns out gluten development isn’t the only important thing, extensibility and increased availability of the dough’s biomolecules is equally important.

    • MartinMartin
      October 30, 2016    

      I forgot to say, thank you for the valuable knowledge you’re sharing with us.

  5. evrenevren
    September 16, 2016    

    question: does incorporating salt later damage the gluten while mixed in, either by hand or a mixer? or you you just sprinkle the salt and it penetrates the dough during folds.

    • September 16, 2016    

      No, it doesn’t damage the gluten, in fact it helps preserve it. The enzymes break down the gluten and help the dough to be more extensible, adding salt help halt that process because salt impedes the action of the enzymes. Salt also strengths and tightens gluten. So yes, you sprinkle in on after autolyse and work the salt into the dough, you can feel the change right away as the dough stiffens up. Salt also adds flavor and hold moisture in the bread, besides discouraging mold.

  6. WilliamMardWilliamMard
    May 6, 2016    

    I am so grateful for your post.Much thanks again. Want more.

  7. Vanya DinevaVanya Dineva
    February 19, 2016    

    Good afternoon from Bulgaria.
    Thank you so much for sharing your experience with us.
    My question is about the mixing/kneating of the dough.You wrote: “Mix up all ingredients except salt. Weigh out three 16 oz/453g pieces of dough and then follow the directions already given above for timing and adding salt.” There is no kneating, is that rihgt? How long do I need to mix (by hand using danish dough whisk or with Kitchenaid)?
    Thank you.

  8. fff takafff taka
    October 22, 2015    

    thanks for this post.

  9. July 20, 2015    

    Thank you!

  10. April 22, 2015    

    Hi, I would like to subscribe for this website
    to take newest updates, thus where can i do it please help out.

    • northwestsourdoughnorthwestsourdough
      April 22, 2015    

      There is a large subscribe button on the top right hand side of the page. 🙂

  11. KatyukKatyuk
    April 2, 2015    

    as a beginner I have followed your instructions and ended up with a beautifully risen and active dough in a banneton. This then collapsed when I transferred it to a baking tray. Do you use a banneton or just bake it from the final proving container?

    • northwestsourdoughnorthwestsourdough
      April 22, 2015    

      Hi Katyuk, I pretty much always use a banneton or couche. You just have to catch the dough before it is over proofed. Try proofing it less before baking and be gentle when turning your dough out of the banneton.

  12. BrunoBruno
    March 31, 2015    

    Great website. Annoying shit unit system.
    Does not make sense to have ounces, pounds, and grams.
    Use freaking metric system and NO OTHER.

    • northwestsourdoughnorthwestsourdough
      March 31, 2015    

      I guess I really have to agree with you Bruno. I have finally gone over to the metric only myself, but it has taken me a while. It is something people get entrenched with. It would make it so much easier if everyone made the leap.

      Anymore, if people won’t do metric, they can’t follow my formulas. It’s as easy as changing the mode on your scale. Most scales have standard and metric.

  13. bbkeobbkeo
    March 3, 2015    

    Super interesting blog and posts. Thank you for taking the time to answer the responses so well. I learned a lot from reading them.

  14. KevinKevin
    May 19, 2013    

    Protease is an enzyme, not a protein…

    • northwestsourdoughnorthwestsourdough
      May 19, 2013    

      Yep, that’s what it says.

    • RasmusRasmus
      October 13, 2016    

      Those two are not exclusive. Protease is indeed a protein with a catalytic function. 99% of all enzymes are proteins.

  15. XieweideXieweide
    January 3, 2013    

    I am a new comer into the bread making, and this experiment really caught my attention.

    I will follow thru…


  16. PeterSPeterS
    April 17, 2012    

    NW SD: this is interesting stuff; good job.

    Just a technical note regarding the weights of the breads and loss upon baking. The differences you observe are probably statistically insignificant. If you were to repeat this 10 times and measure the weights, all other things being equal (which they probably aren’t), e.g. formulation, oven temp, baking time, cooling time and humidity, you would likely find the weights to average out to be the same.

    Appearance differences are real, of course, but it’s a stretch to imply it had anything to do with the weight

    • northwestsourdoughnorthwestsourdough
      April 17, 2012    

      I don’t imply that anything had to do with weight. I merely observed that the more expanded dough lost more weight and was larger and filled with larger holes. A smaller denser loaf will always be heavier due to it’s not being able to evaporate as much water during baking. Several bakers wrote me and had the same outcome with this experiment, so it is significant.

  17. steve kennedysteve kennedy
    March 13, 2012    

    Did you notice any difference in the taste of the three loafs?

    • northwestsourdoughnorthwestsourdough
      April 17, 2012    

      Hi Steve, tastewise they were similar, mouthfeel was different, the crumb and crust were superior on the largest loaf, giving one a “wow, this is good bread” reaction.

  18. SRSR
    January 22, 2012    

    I’ve made this bread a few times with the 2 hour autolyze. As long as I use pendelton mills power flour it turns out ok (occasional large holes though not much oven spring). However the moment I use pendelton bread flour the dough craters and tears when it is final proofing in my small loaf banneton (about 16oz). I thought I wasn’t getting enough gluten development at first but after five tries and 7lbs of flour wasted…I kinda doubt it. I’m at a loss. Could it be my starter? I feed it every 24 hours and fed it 8 hrs before I began.

  19. BenBen
    December 31, 2011    

    Autolyse is meant to be only water and flour. This experiment was not autolyse, your third bread was larger because your yeast had more time without the hygroscopic reductive effect of salt making it more effective.

    • northwestsourdoughnorthwestsourdough
      December 31, 2011    

      Exactly Ben, that was the point. Thank you!

  20. hibobhibob
    December 18, 2011    

    “All of the dough was folded three times during the six hour bulk ferment with no other kneading taking place except to incorporate the salt.”

    I wonder – how much influence did kneading at 2 hrs (as opposed to not kneading or kneading at 30 min) have compared to adding the salt at 2 hrs? Perhaps revisit but knead #1 at the same time/in the same way as #3?
    Years of working in a chemistry lab has taught me the problem with only changing one variable at a time: lots more experiments to run!

  21. Joe MacBuJoe MacBu
    December 7, 2011    

    Very interesting!

    Did you notice a difference in flavor? I think Calvel mentioned that he preferred to have some salt included in the autolyse because it serves as an antioxidant, thus preserving more of the flavor.

  22. November 17, 2011    

    Hi Teresa,
    Very interesting experiment.. What would you say was the room temperature for your 2-hour autolyse?

    • northwestsourdoughnorthwestsourdough
      November 18, 2011    

      My room temp is right around 68F degrees during the colder months.

      • northwestsourdoughnorthwestsourdough
        November 18, 2011    

        Actually I laughed when I re-read what I had written, room temperature is also right around 68F in the warmer months in WA 😉

  23. October 27, 2011    

    Thanks for doing this experiment and sharing your results; great information!

    I have 2 questions:

    1) For the slurry, can one use baker’s percentages and use 7/8 of the flour and water ingredients for the main dough and 1/8 (or 1/4?) flour and water and all the rest for the slurry?

    2a) The initial autolyse dough is just the flour, water, and yeast, right?

    2b) For a recipe that has other ingredients; for example, I’m making a pizza dough that calls for some oil, should that be added at the end, with the slurry, or at the beginning?

    Thanks again.

  24. Ken JordanKen Jordan
    October 27, 2011    

    I love this experiment. Living in the tropics with fairly high heat and humidity affects my timing a great deal. I am always concerned that extending my autolyse will cause my dough to over-ferment. In fact, to keep my dough inactive for the 24 hour bulk refrigerator ferment I use frozen flours and ice water and even do my autolyse in the fridge. Should I just live wild and let it autolyse at room temperature (currently 85F but often in the low 90’s F on Saturday morning when I bake). Should I stop being so concerned with dough temperature? Thanks so much!

    • northwestsourdoughnorthwestsourdough
      October 27, 2011    

      Hi Ken, I lived in Hawaii for a while and was challenged as well. I think using the cold ingredients and autolysing for two hours would work. The cold ingredients would chill the dough enough to make two hours feasible. I did have to cut my usual fermentation times in half while in tropical conditions. You may have to keep a bread journal and write down what is working for you. I would definitely experiment if I were you. Maybe start a blog on tropical bread baking……

  25. tonyktonyk
    October 27, 2011    




    • northwestsourdoughnorthwestsourdough
      October 27, 2011    

      Hi Tony, yes, you have it right, go for it! Remember to do the folds though.

  26. DeanDean
    October 27, 2011    

    My concern with not using a mixer is that it’s happened to me that the salt is not distributed evenly throughout the dough. Nothing worse than biting into fresh bread and getting a mouthful of salt! May I ask how you achieve even distribution of salt throughout the dough without using a mixer? Thanks!

    • northwestsourdoughnorthwestsourdough
      October 27, 2011    

      Well the loaves were only one pound loaves, so it was not hard. A large amount of dough would be a different story. However, if you do a search on my blog for Big Bear’s Bread, I have a slurry technique that works great for incorporating salt into larger amounts of dough.

  27. northwestsourdoughnorthwestsourdough
    October 27, 2011    

    I do not use a mixer anymore, haven’t for over two years now. You need to add the levain early on for dough that is low hydration or where the percentage of levain is very high.I have used overnight autolyse and multiple day cold autolyse and have had great results.

    I need to experiment with the flour and water only autolyse. I did find though, that the jump start in activity was really good for the loaf and you can’t get that with a flour/water only autolyse. But I won’t prejudge it, it might be better, have to try it….

    • mariemarie
      March 9, 2016    

      Theresa – I’ve been baking for about 2 years now with pretty good success following Chad Robertson’s method, but I’ve been wanting to experiment with other methods. This week, I used KA bread flour with 20% White whole wheat and did an overnight autolyze with just the flour and water. Then I added 20% levain and waited 30 minutes before adding the salt with a touch more water (total hydration ~80%). This dough was so slack! I did bulk ferment in oven with hot water pan for about 4 hours and then preshaped a number of times cause the dough seemed so weak. Then I did an overnight final proof in the refrigerator. Any clues about what I did wrong?

  28. DeanDean
    October 26, 2011    

    I know that “autolyse” means different things to different people; an old French sourdough baker insisted that autolyse is water and flour, nothing else. I’m curious how different it would be if you waited and added the levain and salt at the stated intervals.

    I’m curious to try this myself, and have been curious about doing an overnight autolyse since listening to an interview with the Senegalese baker who won the coveted “best baguette in France” award this year. He described his routine, which includes what sounds like an overnight autolyse with just flour and water.

    The thing is that for me sourdough breadmaking is part of my daily routine; I have it down to about ten minutes a day and that includes mixing the salt in with the flour – then adding the starter/water mixture. If I did the autolyse I’d have to drag out the mixer to incorporate the salt, which would add considerable time by the time I cleaned it etc. Maybe one of these days!

    • July 14, 2016    

      I am in the process of milling my own Non GMO wheat and hitting walls. Yet, you were right, Dean. Many have said mixing flour and salt is imperative. Then I learned that salt is a inhibitor to the enzyme process. What to do? All this is confusion and resulting in wheat bread like a rock. I need a formula I can count on.

  29. October 26, 2011    

    Very interesting experiment!
    I do not have a mixer at the moment..I had to leave it in storage when I moved to France…so I autolysis for one hour at 76deg making the assumption that it would help the development of the glutten…and I found it really does! Your experiment helped me realize that I was not imagining it.It does produce a superior crumb! The only diffidence is …I do not add the starter till i knead it along with the slurry of gray sea salt. My thinking was…the levain would lose some of its strength over that extra time.Have you tried adding the starter after the 2 hour period?
    Thanks for a great post…I will have to really look at your links to the articles.

No Pings Yet

  1. Stuff you come across | Charissa's Kitchen on March 28, 2017 at 2:06 am
  2. Autolyse Experiment | Melissa Johnson, Editor and Writer on January 22, 2017 at 10:45 am
  3. Teff Mash Bread on November 19, 2015 at 1:33 pm
  4. Autolyse – Rest and Go to Work - Discovering Sourdough on March 31, 2015 at 9:45 pm
  5. 100% Whole-wheat Bread – Weekend Fun with Tangzhong & Semi-Autolyse Method | Simply a Food Blog on March 18, 2015 at 5:22 pm
  6. Whole wheat walnut raisin boule using the King Arthur method | The Sourdough Journals on January 24, 2015 at 6:17 pm
  7. Result of Pizza recipe changes? - Page 2 - Forno Bravo Forum: The Wood-Fired Oven Community on January 13, 2012 at 11:29 pm
  8. Experiments with Autolyse # 2 - Discovering Sourdough on November 3, 2011 at 4:35 pm

Leave a Reply

Your email address will not be published. Required fields are marked *

Free E-Book

Get, "Make Your Own Sourdough Starter" FREE!

Discovering Sourdough, The Books:
Discovering Sourdough - The bread books!
Only 10-15 dollars  



Favorite Bread Baking Equipment