The Fresh Loaf

News & Information for Amateur Bakers and Artisan Bread Enthusiasts

Starters - What is your experience?

DanAyo's picture
DanAyo

Starters - What is your experience?

Much discussion has taken place pertaining to starters. And rightfully so. The sourdough culture is at the heart of all SD breads.

From this discussion, it is hoped that the experiences of others will shed light on this question.
Will radically different starters remain distinct, IF the various starters are fed and maintained identically in a single location?

Hypothetical Example - Three healthy and active starters are obtained from very reliable and trusted sources. One from Egypt, one from San Francisco, and the other from the original King Arthur starter (if there is one). All three starters are fed with the same flour and water in equal proportions. All starters are kept side by side and everything including the vessels and temperature are kept identical.

  1. Will each of the three starters remain unique
  2. Will each starter continue to manifest the characteristics (acidity, flavor, etc) of the original 
  3. Will the activity of each starter remain the same as the original

If you have other questions to add to the list, let me know.

For this discussion, let’s limit the input to those bakers that have experience in this area. There are many theories that most all of us have read. Many “authorities” hold diametrically opposing beliefs. For this discussion, let’s talk about the things we’ve experienced in our kitchens.

UPDATE  - up until now posters have replied in basic agreement. I would like to hear from bakers that have experienced a different outcome. Who has experienced various starters that have maintained their uniqueness, in-spite of being fed and maintained identically.

Danny

DanAyo's picture
DanAyo

The experience described below has happened to me many times and on numerous occasions through the years. Take what is written below and multiply that experience times 6 or more separate times over the past 15 years or so.

Earlier this month I lucked out and made a “super starter”. One baker named it the “Miracle Starter”. The starter was active and viable after 24 hours from being originally made. The starter ~quadrupled in 6.5 hours. And after 72 hours from first being mixed a bread was removed from the oven that it had produced. BUT after feeding on the counter for 7-14 days the beast became a pet. Now it MAY triple, but no higher. I decided to take my original starter from the fridge and compare the two. After feeding, maybe 3 days, I couldn’t tell the difference in any way from one starter to the other. As a matter of fact (and I’ve done this before) I forgot and cleaned the jars, removing all markings. Now I have absolutely no idea which starter is which. When each starter is pulled with a fork to show the gluten strands there is no difference. They smell the same and even taste identical. They are both rising the same amount and their feed cycles are the same. For all practical purposes they appear identical.

At this time it seems to me that the feed and environment (probably the water also) overwhelms the characteristics of individual starters causing them to take on like characteristics.

It is my belief that IF starters are feed and maintained the same (these assumptions are based upon starters maintained on the counter) that is a matter of weeks before they become indistinguishable. I have compared two completely diverse starters using the same criteria with the exact results. Never has it been my experience that the starters maintained their individual characteristics.  I am very familiar with Puratos and consider them an outstanding source of information. If the starters in their sourdough library remain unique (and I think they do) it seems to me that their individual environments (each starter s segregated from all others) and original flour is the key to their unique qualities. The sophistication of their testing equipment alone is enough reason to trust their findings.

I know that others will very much disagree with my findings, and I am eager to hear from them.
The truth makes us all better bakers...

Danny

DanAyo's picture
DanAyo

Here are both of the originally very different starters. If my life depended on it, I couldn’t tell you which one is which. Just over two weeks ago one of these starters were quadrupling in 6.5 hours. The other starter was never that active. Now compare the growth rates and top bubbles on each as pictured below.


The top 2 images were taken yesterday afternoon, after they had completed a 12 hour feed cycle ending in maturity.

The bottom images shows the same 2 starters after they fermented over night (12 hr) and reached maturity. So the top 2 images and the image below show the results of the same 2 starters after 2 refreshes totaling 24 hours. 

idaveindy's picture
idaveindy

Dan, what did you feed the two comparison starters? Strictly home milled flour, strictly store-bought white flour, or a combo?

I have plenty of maltodextrin powder (DE 10%) , and some dextrose powder (glucose monohydrate) with which I could experiment, (mostly) eliminating the addition of outside yeast/LAB that comes with feeding with flour.

I wonder if that's how the commercial places/labs do it.

I also have some DME (dried malt extract/sugar), potato starch, and tapioca starch.

I've been careful to feed my 3rd starter (Cultures For Health- San Fran style) with store-bought white flour only. 3 months on, it has still been distinct from my 2nd starter (1847 Oregon/Carl's), which I fed with home-milled 100% extraction Prairie Gold.

You've inspired me to experiment.  I'll split off a jar of #3, and feed the 2nd jar of #3 with home-milled Prairie Gold, and compare how it develops with the primary white-flour-fed jar of #3.

--

Update:  I have fed my starters at a ratio from  1 : .5 : .5, up to 1 : 1 : 1.  "Maintenance" feedings are closer to 1:.5:.5.

DanAyo's picture
DanAyo

Most comparisons conducted through the years have used King Arthur All Purpose flour.

It seems the starters used in the comparative test take on identical characteristics in a matter of weeks. I can’t tell the difference...

mwilson's picture
mwilson

You probably can keep a starter "the same" under laboratory conditions. But even the flour varies from season to season and batch to batch...

Here is an anecdotal observation of mine...

I ordered the same flour (type 00) in bags of 25KG from the same producer and fed my starter as normal with this new sack. But immediately I noticed something wasn't right. The consistency was softer and within an hour it didn't look right, it turned dark grey and became quite soft and didn't rise as normal.

To this day it was the strangest sourdough event I have witnessed.

I honestly think that was the result of a shift in dominant microbes, of course I can't substantiate that!

DanAyo's picture
DanAyo

Is it feasible to think that the microbes contained in the starter’s feed flour is able to overtake the microbes within an established starter in a relatively short amount of time? This goes against common belief. Sometimes (maybe often) the “rules” need to be challenged and broken when applicable. Doc.Dough is my man...

By the way, my feed are generally (but not always) large. Often 1:3:5 or 1:5:5. The large feeds are used to extend the feed cycle out to 12 hours, since the starters are often maintained on the counter.

mwilson's picture
mwilson

Think about how active your rye starter was...

I have read a paper that described such a thing. It was even more notable since the dominant microbes shifted after just one feed where the same starter was fed in the same way with the same batch of flour but in a different location.

I think we like to give our starters an identity sometimes even a name.. But in reality they can be quite dynamic. But as always that depends on a number of other factors...

It is categorically true that low ratio feeds promote stability. e.g. 1:1 . That is why the mother or stock starter is typically fed this way.

Doc.Dough's picture
Doc.Dough

This is a bit technical and has more math than some will be comfortable with, but I think it captures the complexity of the challenge you offer:

Thoughts on Starter Stability

And if you look at the complexity of some of the starters that Rob Dunn and his team have characterized then you can understand that the test matrix is truly overwhelming

JeremyCherfas's picture
JeremyCherfas

That's what I would have said too.

MTloaf's picture
MTloaf

I have the same starter that I started 10 years ago and have not used another. It has never been neglected but has been through several phases of hyperactive and sluggish performance, from stiff to liquid and fed a lot of different flours. Reliably erratic. The best tasting loaf was one of the first ones I made with it and I have never been able to recapture that. If it wasn't such a bad time to be wasting flour right now I would start another one just to see how much flavor has been lost to refrigeration.

My bet is that starters are like the snowflakes in my yard this morning individually different but all the same thing.

Our Crumb's picture
Our Crumb

Danny,

I know you want contributions from bakers with personal experience in experimenting with starter stability.  I am the counter example.  I have been feeding the same starter with the same feed at the same frequency and feeding ratio for as long as I've been baking sourdough.  I guess I just like the rut I'm stuck in.

That said, humor me.  I taught college biology for decades, including population and evolutionary biology, so I think about these things a lot.  Doesn't mean I claim to know or have all the answers -- only some long considered thoughts.  To wit:  Every minute aspect of a baker's sourdough starter maintenance routine constitutes a component of the niches each baker creates for each of her/his starter's microbes:  The feedstock (grain species and particular genotypes and their consequent biochemical makeups, time since harvest, fresh milled or not, flour particle sizes/distribution, milling method, storage conditions including temperature, humidity, pests, microbial hitchhikers), the feeding ratio, frequency, the water, the overall culture volume, the vessel size, shape and material, temperature, humidity, mixing instrument and method (see below) and many other variables our bugs keenly sense to which we are blissfully oblivious (Latitude? Elevation above sea level? Light?).

That constellation of niche components constitutes and imposes a selection every time we feed our starters.  That is, it differentially favors the reproduction of the various microbial populations in the starter community with respect to the biological potential of that species (that being the highest reproduction rate the species/genotype is capable of under ideal conditions for it).  On top of that, there are higher order interactions among microbial populations and between them and abiotic aspects of their environment, all of which change over the course of the growth cycle (acid production being a widely recognized one).

Any alteration in that constellation of niche-defining factors will quantitatively shift the distribution of suitability of the growth environment for one species versus another.  Ever so slight variations can have substantial effects because growth is logistic (exponential), quickly amplifying changes in the relative abundance of the starter's populations by changing the probability that a member of a particular species will divide during a given period of time (or stated another way, change the frequency of cell division in that population).  For a simple example, my starter feed contains a tiny amount of rye.  It wouldn't take many minus-the-rye refreshments to generate a measurably different starter from it because some bug species' cell division probabilities are increased in the presence of rye and others are decreased.   As another commenter pointed out, it is a dizzyingly complex system.  Bottom line:  You send me your starter and within a week or two of refreshments chez moi, it no longer resembles your starter at all.

Finally, here's a recent, somewhat illustrative and, to me, fairly stunning anecdote:  Last week I did my three pre-bake starter refreshments differently than usual.  Same ratios, feed and hydration but different mixing order.  I usually take a few grams of starter from a recently matured culture, vigorously disperse it into water, add fresh feed flour, mix, cap and incubate @ 78˚F.  This time, because I was running low on starter feed and wanted to make sure I had enough for all three refreshments, I first measured out the feed flour mix into three dry jars.  I stored two and added water to the third, mixed it thoroughly and then mixed in the starter inoculum last.  Not only did the cultures feel as if the hydration was 10-20% higher than it was, but they grew out relatively slowly compared to ones mixed by my usual method for the past decade, including just the week before.  The inoculum was very thoroughly mixed in -- that wasn't a problem.  But something about fully hydrating the fresh feed flour first results in a totally different behavior of the culture.  So mixing order constitutes an element of the overall conditions that I create for my starter and changing it changed the culture's behavior.  If I adhere to this change in my practice, I expect I'll have a manifestly different starter in a few weeks.  Like I said, blissfully oblivious.

All of the above can be seen as a corollary of sentiments shared in Is the Bread Terroir Code Crackable?  

Enough.  This sheltering in place sure brings out the verbiage.  Sorry.

Tom

Doc.Dough's picture
Doc.Dough

I am wondering if you continued your accidental experiment to verify repeatability and to see if you wound up with a manifestly different starter after a few weeks (March to May being defined as a few weeks). Your result is very interesting and somewhat surprising but as you point out there is a lot going on and it is trajectory dependent (as I tried to express in the paper I pointed to above).

Our Crumb's picture
Our Crumb

Hi Doc,

No, that unexpected behavior put me off that mod of my routine.  Now, if my starter had grown visibly better after that mixing protocol, I likely would have pursued it.  But baking good bread tends to be more important than experimentation chez nous.  Maybe when things settle down, I'll be more inclined to an experiment like that.  We're moving across the country in a few weeks with many things on our plate, all coated with a gooey pandemic glaze. 

But if and when I do get back to that, you'll see the results here.

Stay safe,

Tom

Doc.Dough's picture
Doc.Dough

Dan Ayo sent me a sample of his "new" starter for testing and I found that it was not as fast as his results appeared to show.  But more recently I sent a dehydrated sample across the country in a very slow USPS truck and held back a bit of the sample for comparison purposes.  When I revived the starter on my end and compared it side by side with the original in how fast they produced CO2 (or at least how fast they lose weight during growth) the recovered starter will burn through 2% of the weight of added flour in half the time that the original does (I ran three sequential tests because I was questioning my data). So the process of dehydration and rehydration has done something to the community of yeast/LAB/other to select for a very fast dominant yeast strain.  No further insights at the moment but I am thinking about that result.  The pH of the starter at the 2% weight loss point was initially somewhat high (4.2) but over three refresh cycles came down to 3.7 which is a healthy operating point which I attribute to slow recovery of the LAB from dehydration. Still work to do to understand it (or not).

Good luck on your trip.  How do you plan to transport your starter?

DanAyo's picture
DanAyo

Doc, I think I told you the following, but thought it good to share it for public record.

The starter in question can be seen HERE. During the fist few feedings the starter grew abnormally fast and rose as much a quadruple. My typical starter will rise ~3X or so. BUT after a short period of time (less than 2 weeks)  the starter was acting exactly like one of my typical starters.

Guess what! While working with the starter above, I forgot to label which starter was the new one and which was the older one. For the life of me I couldn't tell which was which. To this day I can't tell you if I'm baking with the latest starter or my old stand by. I do keep dehydrated backups, but the present one is doing fine. That scenario has happened to me so much with different starter is the past, that I have become a firm believer that if 2 very different starters are maintained identically and feed the same flour, that in short order they will become indistinguishable. Many starters and much hands-on experience have brought me to that conclusion. I know many will disagree but at this time this is exactly what I believe.

Debra wrote me this week about the starter linked above. She read that according to Robb Dunn's study that bakers that bake a lot of SD carry the microbes of their starters on in the skin on their hands. Originally it was thought that the bakers supplied the microbes from their hands for the starter, but the study showed that the microbes from the starters were found in the hands of the bakers.

Doc.Dough's picture
Doc.Dough

Danny - have you ever run two starter refresh cycles side by side with the same amount of seed starter, same amount of flour, but different hydrations and compared the volume increase when they peak?

It will be interesting to see how this rehydrated starter behaves over time.  I maintain two (other) starters which sometimes behave like twins but sometimes exhibit variations in time to achieve the same weight loss. But that does not explain where the "new" vigor came from in the rehydrated sample.  If it came from my hands it did not come from either of the other samples.  Weight loss data below. DA' is the original, DA" is the rehydrated version.  The difference is not small, and it is consistent.  We will see how long that lasts.

DanAyo's picture
DanAyo

“ Danny - have you ever run two starter refresh cycles side by side with the same amount of seed starter, same amount of flour, but different hydrations and compared the volume increase when they peak?”

I have, but unfortunately don’t remember the results :-(  Common sense tells me the drier starter will rise higher because of the strong gluten, but I don’t have the data to back that up.

My best explanation of the run away new starter and possibly your rehydrated starter is that the yeast population is not hindered by an abundance of LAB. Once the LAB gain a foot hold and the starter begins to settle in and equalize, the yeast become less dominant and consequently not as powerful. What are your thoughts on this?

Doc.Dough's picture
Doc.Dough

That is what I expected too, in part because it took a few days to develop any substantial sour flavor, but even after the pH at end of refresh was down below 3.8 (where half of the lactic acid is ionized and half is not) there was/is no slowing down. 

The reason I asked about the starter hydration is that a 4X volume increase for your starter does not necessarily correspond to more active yeast and is more likely to reflect a stiffer starter, you can only measure yeast activity directly by measuring the weight loss (and even that misses the CO2 that is required to saturate the liquid in the dough). If I make a stiff levain, the volume growth can be quite large if I develop a little gluten at the start (4x-6x before it deflates on its own).

DanAyo's picture
DanAyo

When I was reporting about THIS starter, it was 100% hydration.

Your report of 4x-6x is mind blowing. I have never seen 6x and can’t recall 5x. 6x is super impressive no matter what you did to get it. I’ve tried everything I can think of.

leslieruf's picture
leslieruf

I want to share what is happening with new starters I made for Rob Dunn Lab experiment.  2 starters fed the same flour but the water was different. Corie N was made with bottled spring water, Corie T was made with tap water, filtered, boiled then left out overnight.  Apart from the first few days when Corie T was very active, Corie N is way more active, even though now that I have submitted my data and run out of bottled water both are fed and treated identically.  They were started on 18th April 

Leslie

 

DanAyo's picture
DanAyo

Great post, Leslie!  ...and so timely.

Just yesterday I called our tap water provider to learn more about our water. They informed me that it did contain chloramines. The lady said that if the water was allowed to sit open for several days or more that it would dissipate like chlorine. I'm not sure I believe that from the things previously read.

Water quality would be a reasonable answer to why certain new bakers produce great bread very early in their learning process.

You’re experience is intriguing. Let’s get together and discuss that more.

Danny