The Fresh Loaf

A Community of Amateur Bakers and Artisan Bread Enthusiasts.

How to use white wheat flour starter to make rye bread?

Doc.Dough's picture

How to use white wheat flour starter to make rye bread?

Does anybody have a definitive reason why I should not be able to use a starter that is being maintained on white wheat flour to make rye bread? If I want to do that is there any guidance on a good way to go about it?  Do I have to allow the starter to adapt to rye flour or can I just pitch a batch of white starter into a bowl of rye flour and expect it to go to work?

BaniJP's picture

Nah, white flour starter works just fine. Most rye breads are not completely 100% rye (more like 60-70%), due to the properties of rye flour in baking.

widdersbel's picture

I've used my white wheat flour starter to make a 100% wholemeal rye loaf, with great success - at least, I was very happy with the results.

If anything, the dough seems to rise faster with rye... I wonder if this is anything to do with rye supposedly being more nutritious than wheat?

DanAyo's picture

The key word here is “definitive”. Although I can’t be definitive, experience leads me to believe that unless there is a very high percentage of pre-fermented flour in the levain, the difference is relatively inconsequential. Surely others will disagree, but a small percentage of PFF coupled with a long ferment will build all the acids desired for most bakers.

Keep in mind, I typically work with very low percentages of preferment.


I fully expect you to call me on this. <Laughing Out Loud> I imagine you’ll be conducting TTA test to reach a “definitive” conclusion. The forum wouldn’t be the same without you.

starvingviolist's picture

It should work ok, but it won't be optimal. The bacteria and yeast in a start adapt to their conditions over time. If they are fed a diet of rye flour, they will adapt to become a better rye starter, bacteria can break down starches found in rye flour into sugars, and that will thrive under conditions similar to a rye dough, will flourish. If you start a rye culture with a bit of your wheat culture it will take about three days to evolve into a rye starter.

Our Crumb's picture
Our Crumb

Perhaps you could try feeding your white wheat starter with a flour mix equivalent to that of the rye you seek to bake. You’d have an answer in 5-10 hours about the ability of your starter to raise the dough you want to mix. 


HansB's picture

I used to maintain separate wheat & rye starters. I got tired of that and now only maintain a wheat starter. Now when making rye bread I just build the levain with 50/50 wheat/rye flours. Makes really nice rye breads.

DanAyo's picture

I’m with Hans. My starter readily accepts any food it receives. It will go form 100% AP to 100% whole rye in a single feeding and not miss a beat.

Only keep one starter.

suave's picture

It may work if your starter is capable of generating sufficient acidity or percentage of rye is relatively small.  For predominantly rye breads I would suggest taking an extra step and building rye starter from your wheat mother.

pmccool's picture

I usually, not always, feed my starter a combination of whole rye, whole wheat, and bread or AP flours.  It's never the same from one feeding to the next, since I'm just eyeballing the amounts.  However,  it's ready and able to seed any levain made with any of those flours since the organisms are already accustomed to that type of food.

If I make an all-rye bread, I'll usually give the starter one or two feedings of nothing but rye flour.  That's as much to cut down on the non-rye flours in the bread, which the starter would contribute, as it is to ensure the starter's readiness for an all-rye diet.  But, even if I use it straight out of the refrigerator to seed a levain, it will still work fine and the bread won't be affected.


Doc.Dough's picture

Thanks to all for the opinions expressed.  I will tailor the experiments based on the expectation that it probably doesn't make a large difference (What, me worry?). I will refresh my white wheat starter with 100% white, 25% rye, and 100% rye, then track how fast they lose weight to CO2 generation at 38°C in a temp controlled water bath.

I will post the data with what will serve as analysis, hopefully in a few days.

DanAyo's picture

Looking forward to your methodical post! We would expect the 100% whole rye to lose more weight in the same amount of time. But we often don‘t get the results we expect. <So True, at least in my case>

But Doc, I am of the opinion that the ‘magic’ really takes place in the dough. It seems to me that dough that is subjected to extended fermentation are somewhat similar to a starter or levain, just on a larger scale. 

Example - 1000g total flour and 10% PFF and let’s say 70% total water. If it was called a levain it would be (starter @ 100% hydration = 100g water & 100g flour), plus 600g water and 900g flour OR 1:6:9. And then the dough might have 2% salt.

What do you think?


Doc.Dough's picture

No surprises = no information.  So if you are myth busting, you expect no surprises.

In the limit, holding over bread dough to act as a chef for the next batch is just making starter with no additional work.

Doc.Dough's picture

Based on 12 hrs of data, those who predicted that there would be no impact seem to have been correct.

All samples have lost >2.5% of the weight of the added flour in 12:00 from mixing (this is the weight of the CO2 produced and lost to the atmosphere).  Measurement errors are on the order of ±.03% but the samples were not burped before every measurement so there are residual systemic errors.  That was fixed as you can see in the second run of this experiment shown a few posts below.

The 100% white flour sample had something strange happen that caused it to apparently gain weight early in the test so I will probably re-run it tomorrow but start each container with the end product of today's test.

Today they were all started with ~2.45g of white starter that was a couple of days old, and 16g of flour added to each and differing amounts of water added to get to approximately the same viscosity (which meant 13g, 15g, and 20g of water for the 0%, 25%, and 100% dark rye respectively).


DanAyo's picture
  1. What was the ambient temp during fermentation?
  2. I estimate approximately a 1 to 16 ratio of pre-fermented flour to virgin flour. Did the starters recede after 12 hours?

The varying weights of water for each flour mix makes a lot of sense.

Doc, why is there always a lag time for freshly mixed cultures?
I know, that was 3 questions :-)

Doc.Dough's picture

Fermentation temperature was 38°C as noted above.

What appears to be lag time is just the delay during which the initial production of CO2 is being absorbed by the liquid phase of the starter until it becomes saturated and only then begins to inflate the alveoli.  At that point excess CO2 becomes available to diffuse through the surface and into the container (and then out unless you have a gas-tight seal). Another reason for some delay in CO2 production is a period of time while the pH is declining to the optimum point for fermentation and as acid is produced and drives the pH down, the fermentation rate increases. There are two components of pH change.  Both the absorption of CO2 to make carbonic acid and the production of lactic acid and to a lesser extent acetic acid are contributors to this process.

This is not the same as the lag phase of yeast or LAB growth rate when an initial slug of sugar becomes available and the cells have to build the structures necessary to ferment the newly available food before they can divide/multiply.

Doc.Dough's picture

I re-ran the experiment today with a little better control and the results are more tightly clustered. The three samples and their refresh ratios were:

100% white (3:13:16)

25% rye/75% white (3:15:16)

100% rye (3:20:16)

The 25% rye/75% white continues to be the best performer.  This set was run at the same temperature (38°C) but with slightly larger initial inoculations (2.96±.02g) and the seed came from matching leftovers of yesterday that had been refrigerated overnight. The 100% rye sample seemed slightly dry and that may account for the slightly lower activity relative to the 25% rye but that would require a few separate runs to confirm.

So the conclusion seems pretty well founded: that I can keep a white starter and initiate batches of rye bread without pre-conditioning the starter with even one feeding of rye flour.

DanAyo's picture

Doc, I am surprised that you fermented your starter for 12 hr at 100F using a ration of 1 to 5.3 (starter to flour). Did the starters recede completely by the end of the 12 hr?

It looks like it took 9 hr to lose 2% of the weight of the added flour, which I believe is your goal for a fully mature starter.

Am I missing something? I would have thought that an active starter would mature in 6 hr or so (give or take an hr) at such a high temp of 100F (38C).


You wrote, “then track how fast they lose weight to CO2 generation at 38°C in a temp controlled water bath.”
-   Your water bath is interesting. What a great way to normalize the temp throughout the dough! How do you set this up. Sous Vide? That concept has my mind running in high gear. My favorite bread ferments for 16- 17.5 hr at 78-79F. A Brod & Taylor is used, but the temperature is very difficult to maintain. The bottom and top of the dough temps vary considerably. A water bath might be a great help. I say might because there is also the issue of the heat generated from the fermentation itself. Please enlighten me and share any helpful thoughts.


inquiring minds want to know...

Doc.Dough's picture

I stopped at 12 hrs even though all of the samples were still going strong. It was time to clean up. You might argue that there are indications that the growth rate is beginning to roll off but typically they would run another six or eight hours before even beginning to suffer from lack of sugar.

This experiment was designed to compare the growth rate of three different flours.  It could have been run at any arbitrary hydration, and in fact each sample had a different amount of water in the mix. The combination of flour and water determines (for the most part) the viscosity of the starter, and the viscosity largely determines when and whether a dome forms and when/if it collapses. The 100% rye sample never formed a dome as it was stiff enough to just leak CO2 through fissures in the (gluten free) structure. The other two would form a dome between measurements after the first few hours, but they would collapse when the sample was burped to release trapped CO2 just before measurement (that is one reason why the data from the second run is smoother than the first). 

Remember that the samples could have been mixed at a much higher hydration where the CO2 freely migrates to the surface and forms (at best) a foam before the bubbles burst and the CO2 is released.  In that case there would be no need to burp the samples since the CO2 has an escape path that is not dependent of physical manipulation. Or the samples could have been subjected to a vacuum to flush out trapped CO2 (which works differently for samples of different structural strengths and viscosities). So the physical manifestation of collapse is a strong function of both hydration and the effective viscosity which reflects the gluten formation (or lack there of) within the structure of the starter. No gluten (as in the 100% rye sample) means less trapped CO2. Thus forming a dome and falling back is not a good indication of either CO2 production or starter maturity. 

In fact, the development of the weight loss methodology to measure growth rate was driven by a set of experiments in which the levain hydration was above 200% (to dilute the acids produced and facilitate more total acid production during levain build and thus produce a more sour resulting bread). In that case, monitoring starter maturity by watching it rise and fall was totally ineffective and measuring TTA was not measuring the population density of the yeast and LAB.  The resulting starter maturity criteria of losing 2% of the weight of the added flour proved to be a robust indicator which also works effectively at larger scales as well.

As for how long a particular starter takes to mature, there are lots of variables in play, including the specific culture, the inoculation fraction, and growth temperature.  You are correct that they took nine to ten hours to reach the 2% weight loss point.  But for this experiment the focus was on comparative performance not maximum growth rate. For this particular culture 38°C may be above the optimum growth temperature.

Yes, I use what passes for a sous vide circulator to maintain water bath temperature - it is a Lauda E100 which was designed as a laboratory temperature bath controller long before sous vide became a popular form of home cooking (it controls to ±0.05°C).  I float the samples in little boats that consist of 5.5 oz polypropylene food service cups ballasted with 2" dia/.125" thick/44g steel washers.  The starter is grown in a second (covered) cup that slips into the boat and thus keeps it's feet dry. The water bath is a 1/6-sheet x 6" deep stainless food service pan filled with ~4" of water and covered with a foil wrapped piece of double-weight corrugated cardboard cut to fit.