Long Technical Post 1
Daniel Wing
Some members of this newsgroup will remember that I have posted some of
the content of my correspondence with Michael Ganzle, a German sourdough
researcher. He has recently reviewed a proof of a book I have written
about masonry ovens and naturally fermented bread, and has commented in
detail. Those comments will interest those of you who are interested in
the science and technology of sourdoughs. This post and others that follow
are for you. If you ARE NOT linterested in the subject, stop here, and
save yourself from confusion and frustration.
In each section Michael quotes a sentance from the book, and then responds:
------------------------------------
“witness the profusion of instant yeast brands-- while the opposite is trueł
**
I strongly appreciate the notion that the “time equals money equationł is
not true for sourdough bread or any kind of other fermented foods-- wine,
soy sauce, cheese, vinegar, fermented sausage: they usually get better if
they are fermented for a long time (the definition of “longł varies,
though, with the different foods).
-------------------------------------
“I triple it by mixing it with its weight of water and its weight of flourł
**
There is a microbiological explanation for the three stage sourdough
processes. Microbial growth can be divided in three stages. When the
organisms are transferred to a new environment (e.g. by refreshing a
sourdough that has been in the refrigerator), they take some time to
adapt; no growth occurs (“lag phaseł). Once the organisms are familiar
with the new environment, they start to grow exponentially, meaning one
doubling of cell counts in a given time (generation time), so called “log
phaseł. Eventually, the culture will become stationary, i.e. the organism
have run out of food, or are inhibited by the metabolic end products. For
effective sourdough fermentation, one needs a lot of metabolically active
cells. After three or more refreshments, the organisms will reliably start
to grow soon after inoculation and will produce enough carbon dioxide.
Things are different with yeast dough, though: there simply are so many
cells that these have to cough only once to raise the dough.
--------------------------------------------
“the time it was inoculated and to the temperature at which it is kept
than with the size of the inoculation.] Let's call this the second leavenł
**
Comment No1: Weąve been doing quite some work to figure out which factors
affect microbial growth in sourdough. Iąve done some work in vitro (which
is about to be published: Gänzle et al., Modeling of growth of
Lactobacillus sanfranciscensis and Candida milleri in response to process
parameters of the sourdough fermentation, Applied and Environmental
Microbiology, July 1998); and a colleague of mine, Markus Brandt, has
tried to figure out how my “model predictionsł work out during the actual
dough fermentation. Taken together, one can state the following:
A) The optimum temperature for sourdough lactobacilli is 32 - 33°C. At
37°C and 20°C, the generation time is twice as long.
B) At 39 and 15°C, the generation time is four times as long.
C) At 41°C and 4°C, no growth is observed.
For the yeasts, the figures are as follows:
A) 28°C(optimum growth)
B) 32/20 (double generation time)
C) 34/14 (fourfold generation time)
D) 35°C, 8°C: no growth.
So: if several refreshments are done above 32°C, the yeasts will drop out
eventually. The optimum pH for lactobacilli is 5.0 - 5.5 (which is the
initial pH of a sourdough with 5 - 20% inoculum), the minimum pH for
growth is 3.8 (they usually produce acid until pH 3.6 is reached).
Lactic or acetic concentrations donąt affect growth of lactobacilli very
much: this is the reason why the buffering capacity of the flour is so
important for the organism (a high buffering capacity in high ash flours
means that the lactobacilli produce much acid until the critical pH is
reached). It also means, that in doughs that are continuously operated
with a high inoculum (more than about 30%), youąll find more yeasts and
fewer lactobacilli. Eventually, the lactobacilli flora may change, with
more acid tolerant lactobacilli (e.g. L. pontis) prevailing. Such a
sourdough is found in the Vollmar and Meuser continuous sourdough
fermentation machines (there are 6 operating in Germany, and a diploma
candidate in our department characterised the microflora of several of
these: as the machine is operated with a 50% inoculum, the pH is never
above 4.1 - 4.3, and no L. sanfranciscensis is found in those doughs).
Yeasts are different: they donąt mind the pH at all, but are strongly
inhibited by acetic acid, and to a much lesser extend by lactic acid.
Increasing salt concentrations inhibit growth of lactobacilli, but yeasts
tolerate more salt. No salt is added to the sourdough until the final
bread dough, but the dough yield affects the salt concentration: with a
low dough yield (little water), the salt (ash) is dissolved in a smaller
water volume, and the salt concentration goes up: resulting in a slower
fermentation.
So much for the “in vitroł theory. Surprisingly, Markus has found most of
the predictions to come true when he was looking at the cell counts at
different temperature, size of inoculum, salt concentration, and pH in rye
dough. The variation of the inoculum size was interesting: If he reduced
the inoculum size by 2, he had to wait almost exactly one generation time
(one doubling time of the lactobacilli) longer until the dough has reached
the same cell counts, pH, titrable acidity, and so on as the dough with
the higher inoculum. This was true for inoculum sizes between 1% and 20%:
at 50% inoculum, the pH is so low that the lactobacilli donąt really grow
well, and at an inoculum size of 0.1%, the pH and/or the oxygen pressure
in the dough are so high that the cells have a lag-time (see above) of an
hour. Thus, a scanty inoculum means one generation time longer
fermentation.
The generation time of L. sanfranciscensis in rye dough at 28°C is a
little less than an hour (figures may vary with different strains in
different flours, but itąs not much more or less than that), so if the
inoculation size is reduced from 20 to 2.5%, itąll take about three hours
more until the dough is ripe.
The question is, whether these findings are true for all flours and for
all organisms. The strain isolated by Kline and Sugihara does not differ
very much from the two strains Iąve been looking at. All the literature
available tells me that - as long as weąre looking at sourdoughs with a
tradition of continuous propagation - the system behaves the same way.
Differences may be between rye flour and white wheat flour: in white wheat
flour, the enzyme activities are so low that the organisms may run out of
food before the critical pH (lactobacilli) or the critical acetic acid
concentration (yeasts) is reached.
=============================================================
This discussion continues in the next post-- DCW
--
Dan Wing
Wag...@connriver.net
slki...@aol.com
Dan,
Thanks for the great info! I'm still digesting much of it
(if you can forgive the pun), but I had a few thoughts.
I found Michael Ganzle's "comment #1" especially informative.
wag...@connriver.net (Daniel Wing) wrote:
>
> Taken together, one can state the following
> [condensed by Sam]:
>
> For sourdough lactobacilli
> o 32 - 33C = optimum growth
> o 37C and 20C = double generation time
> o 39C and 15C = fourfold generation time
> o 41C and 4C = no growth observed
>
> For sourdough yeast
> o 28C = optimum growth
> o 32C and 20C = double generation time
> o 34C and 14C = fourfold generation time
> o 35C and 8C = no growth observed
This, I think, is very useful and enlightening information
that has direct bearing on some of the recent r.f.s.
discussions, as well as some of the FAQ entries. So I
thought I'd talk about some of them...
First there is Dick's time/temperature formula, which
is based on the assumption that "it is usual for biological
reactions to double in velocity for each 10C rise in
temperature, between freezing and heat death." This
formula works well for rising times, and 10C is a
decent approximation of the combined data we see above.
What the assumption doesn't explain is the steep drop in
population growth at (sub-cell-death) temperatures above
"optimum," and that "zero growth" is observed above freezing.
That said, it should work very well between ~10C and ~32C.
So we see that most of the interesting growth phenomena
(particularly with respect to yeast versus bacterial growth)
happen at extreme temperatures. This has direct bearing on
certain suggested methods of starter maintenance:
High Temperature Maintenance
The FAQ contains an interesting entry on how one might
manipulate the microbiological population of a starter by
refreshing it at high temperature. This is summarized in
http://www.nyx.net/~dgreenw/whattemperatureshouldmysta.html
One should carefully note (and I would suggest that the FAQ
be annotated to reflect) that "if several refreshments are
done above 32C, the yeast will drop out eventually." This
would be especially true at, say, 35C/95F when the lactobacilli
are still multiplying at a high rate while the yeast are at
"zero growth."
Low Temperature Maintenance
A lot of us keep our starters in the refrigerator. Many
of us keep them exclusively in the refrigerator. The
latter practice may not be such a good idea. The data
above seems to indicate that yeast may drop out over time
if a starter is refreshed at temperatures between 4C/39F
(bacteria "no growth") and 8C/46F (yeast "no growth").
I imagine that a 10C/50F temperature would be firmly
within the margins of safety, but that seems awfully warm
for a refrigerator. Unfortunately, an observation of "froth"
or other visible starter activity cannot necessarily confirm
the presence of yeast. It is possible to ferment/rise a
dough exclusively with lactobacilli.
On the bright side, I imagine it is possible that the yeast
in a constantly-refrigerated culture may mutate to become
more low-temperature-tolerant, or (more likely) a low-
temperature-tolerant yeast may "invade" the culture and
take the place of the original yeast. Either way, although
you may find yourself with a perfectly good culture, it will
not be the same culture with which you began.
This shouldn't affect people who store their starter
in the refrigerator but generally refresh it at room
temperature.
> In doughs that are continuously operated with a high
> inoculum (more than about 30%), you'll find more yeast and
> fewer lactobacilli. Eventually, the lactobacilli flora may
> change, with more acid tolerant lactobacilli (e.g. L. pontis)
> prevailing.
Many of us feed our starters by "doubling." This seems a very
good reason to switch to "tripling!"
> In white wheat flour, the enzyme activities are so low that
> the organisms may run out of food before the critical pH
> (lactobacilli) or the critical acetic acid concentration
> (yeast) is reached.
I wonder if this is true for our typical US white wheat flour,
which includes ~.1% barley malt for the enzymatic content.
I never supposed that this might be so... I'll have to
reconsider my thoughts on the use of diastatic malt syrup
as a worthwhile sourdough dough additive.
Sam Kinsey
slki...@aol.com
-----== Posted via Deja News, The Leader in Internet Discussion ==-----
http://www.dejanews.com/ Now offering spam-free web-based newsreading
Larry Caldwell
> The FAQ contains an interesting entry on how one might
> manipulate the microbiological population of a starter by
> refreshing it at high temperature. This is summarized in
[ ... ]
> A lot of us keep our starters in the refrigerator. Many
> of us keep them exclusively in the refrigerator. The
[ ... ]
> take the place of the original yeast. Either way, although
> you may find yourself with a perfectly good culture, it will
> not be the same culture with which you began.
If you really are attached to a particular starter, it's simple
enough to freeze a sample. Constant characteristics may be important
to a commercial baker, but I don't see any reason for the sourdough
hobbyist to be so limited.
I play with my starter. I particularly like to change what I feed
it for a month or so and see what it does. I started it with
half a dozen lactobacillus sources, and I bet all of them are still
in there, in varying concentrations that feeding can change.
I commonly feed potato starch. I also have tried rye flour, oat flour,
barley flour and corn flour. I was using milk for a while, but it
started to get to "limburgery" to the point that the cheese flavor was
coming through in the finished bread, so I dropped the milk and the
flavor shifted back. The corn flour made a real pucker power sour
mash. I didn't feed 100% corn, but mixed it half and half with wheat
flour.
I never refrigerate the starter. I keep it on top of the refrigerator,
where the temp is 23 degrees C. When I feed it, I use hot water,
though not so hot as to kill the culture. If I'm only feeding a
little, I use almost boiling water, if I'm doubling or tripling I
use hot tap water. Except sometimes I don't feel like it and use
regular tap water.
I haven't had a bit of trouble attaining a fine tasting sourdough.
Some types of starter do better in some recipes than others. The
milk based starter made the best double sponge Swedish Rye, but was
too strong flavored for sourdough french.
It's all in how you cook. An amateur can afford to take more risks
than a professional.
-- Larry
slki...@aol.com
Larry,
I agree that one should be open to experimentation.
Also, if what you're doing makes bread that you like,
more power to you. That said, there are some comments
I'd like to make on your post for the sake of clarity.
In article <tDbb10O5...@teleport.com>,
lar...@teleport.com (Larry Caldwell) wrote:
>
> If you really are attached to a particular starter
> it's simple enough to freeze a sample.
Actually, it's not so simple. There is fairly strong
evidence that many sourdough organisms will not survive
freezing. On the other hand, Carl has been drying and
freezing his sourdough culture for years (see
http://www.nyx.net/~dgreenw/canifreezeordrymystarter.html).
However, it is likely that Carl's starter has responded
to this treatment by evolving freeze-tolerance. Such
tolerance cannot be assumed for all sourdough cultures.
Anyone who plans to freeze a sourdough culture (other
than Carl's) would be well-advised to "test-freeze" a
sample and make sure that it recovers from freezing
with its unique characteristics intact.
> Constant characteristics may be important to a
> commercial baker, but I don't see any reason for
> the sourdough hobbyist to be so limited.
Well, there are limits and there are limits. I would
never suggest that home bakers abandon experimentation.
However, unpredictible starter behavior is often cited
as one of the most limiting aspects of sourdough baking.
Mistrust (or misunderstanding) of one's starter is a
primary reason why some bakers chose to rely on added
bakers' yeast to leaven the dough.
> I play with my starter. I particularly like to change
> what I feed it for a month or so and see what it does.
> I started it with half a dozen lactobacillus sources,
> and I bet all of them are still in there, in varying
> concentrations that feeding can change.
I'll take that bet, because I bet that you're wrong
there. As I showed in my previous post, temperature
(especially >35C) can have a profound affect on
the microorganisms in your starter. Dan's recent
"technical" posts also showed how the percent
innoculation (and consequent pH effect) can cause
certain lactobacilli to drop out while others
become dominant. Etc. Etc. Etc.
A starter is basically a little "Darwin Machine."
As the environment changes, the microorganisms living
in it will either change and adapt or they will be
replaced by better-adapted microorganisms. The pre-
existing microorganisms do have a survival advantage,
because they have established a mutually beneficial
symbiosis. But a big change in environment could
easily take away that advantage. It is true that
certain sourdough cultures have been scientifically
shown to retain the same microorganisms over many,
many years -- but these are consistently-maintained
cultures.
> I was using milk for a while, but it started to get
> too "limburgery"
I imagine that the milkfat was turning rancid.
> I never refrigerate the starter. I keep it on top of
> the refrigerator, where the temp is 23 degrees C. When
> I feed it, I use hot water, though not so hot as to kill
> the culture. If I'm only feeding a little, I use almost
> boiling water, if I'm doubling or tripling I use hot tap
> water.
Like I said, if it works for you -- great. I would never
recommend this kind of starter maintenance to anyone.
At the temperature you are keeping your starter, one would
have to triple it at least once a day to keep the pH low
enough for L. sanfransisco (assuming that you want that
kind of lactobacillus) and to keep whatever microorganisms
you have in there healthy. To come along and add "almost
boiling water" to this starter will surely kill off plenty
of the remaining yeast/bacteria. From what you describe,
I would imagine that the microbial population of your
starter is in almost constant change.
Larry Caldwell
> > I was using milk for a while, but it started to get
> > too "limburgery"
> I imagine that the milkfat was turning rancid.
I guess I forgot to mention that I used nonfat milk. I think it
was actually penicillin, and I was getting a true "limburger"
culture going in there with the sourdough. It was certainly
interesting, and actually quite tangy, but it did have an unfortunate
aroma that didn't cook completely out. Dropping the dairy products
solved the problem.
It may have been a yeast strain. Anyone who has ever worked with
salt rising bread knows that yeast can make some pretty pungent
odors. Some day I'm going to develop a saline starter and see what
it tastes like. Summer is coming, and I could set up a rising box
in the garage when it gets warm enough. Making salt rising bread
in the kitchen is rude. :)
-- Larry