I've been trying a couple of things: increasing sourness (based on what I've learned from Debra Wink, and other online references, varying hydration; and feeding portions of my favorite starter different flours, and developing it at different temperatures (part of the sourness investigation.). I've been doing these things one step at a time, so the results don't get clouded.
For the sourness experiments, along with Ms. Wink's super TFL postings, my other main source of information is:
http://aem.asm.org/cgi/reprint/64/7/2616
"Modeling of Growth of Lactobacillus sanfranciscensis and Candida milleri in Response to Process Parameters of Sourdough Fermantation"; Michael G. Ganzle, et al; Applied and Environmental Microbiology, July 1998
and http://www.nyx.net/~dgreenw/whatistherelationshipbetwe.html
an answer provided by the above author to the question, "What is the relationship between temperature and activity?" in a Q and A blog relating to sourdough.
Sourness: flour and temperature
First, an apology, and a plea. Although I am educated as an engineer and scientist, microbiology is far distant from my underwater acoustics speciality. I've struggled, mostly with the subject-specific techincal language, in my effort to understand what I've read. Nonetheless, I think I've acquired the background of knowledge that a home baker, obsessed with sourdough, can use in his or her non-laboratory kitchen to effect the flavor profile of their sourdough breads. Please, if you find my efforts have been based on faulty premises, wrong information, or misdirected experimentation point out the errors, and, more importantly the correct assumptions; accurate, alternative references; or suggest appropriate action--including, "Stop your silly mucking around!"
Debra Wink, in one of her postings, commented that that a flour's ash content contributed to the degree of sourness one might achieve in a starter, but didn't explan how. The first of the above references shows that the activity (reproduction) of Lactobacillus is strongly linked to the the starter's pH ( a measure of acidity). As the acidity increases. or decreases, above or below a most activity-advantageous value (approximately a pH of 4.2) L. Bacillus reproduction decreases. Assuming, for the moment, the temperature of the starter remains steady, and the activity-advantageous pH can be preserved, the amounts of acetic and lactic acid produced is proportional to the concentration of L. sanfranciscensis. However, in any solution the more acid the lower the pH. Some molecular components of the starter's mix may neutralize a portion of the acidity, while maintaning its sourness contribution. In flour and water mixtures that neutralizing (buffering) quality is supplied by the flour's ash content. Simplistically, I thought, the higher the ash content in the feed, the greater the buffering quality of the flour, and, therefore, the more acids produced before the bacteria activity slows down.
With that in mind, I fed a portion of my favorite starter, at room temperature, for three days a steady, every-twelve-hours diet of first clear flour, known to have high ash content. This became my seed starter for three formula-ready levains. In general, this starter, aledged by the vendor to be authentic San Francisco sourdough starter, doesn't produce much discernable sourness, if any at all. On a few occasions, we (my wife and I) have detected some sourness, which has allowed me to conclude there's some L. bacillus in there, maybe.
After 72 hours I built 500g of formula-ready levain,at 100% hydration, using first clear flour; it contributed 28% of the total flour weight. The balance of the dough's flour consisted of 10% rye flour, 31% all purpose flour, and 31% bread flour. The final dough contained 2% salt, at 70% hydration. This formula was used three times; each bake consisted of two loaves, formed into approximately 750 g batards. Every loaf was processed as indentically as possible in a home kitchen: two and one-quarter hour bulk proof with two S&F at 45 minute intervals, followed by an additional 45 minutes. Subsequently, the dough was divided, preshaped. rested for 10 minutes, shaped, final proofed for two hours, slashed and baked at 450*F, with steam for the first 15 minutes. The remaining seed starter was stored in the refrigerator at 37°F.
The only intentional variable was in the levain constructions.
First levain: 20g seed starter, three 1:1:1 feedings of first clear flour, initially and at eight hour intervals. Harvested 500g of levain after 24 hours. The developing levain remained at room temperature (68°F to 72°F) for the entire duration.
Second levain: 20g seed starter, three 1:1:1 feedings of all purpose flour, initially and at eight hour intervals. Harvested 500g of levain after 24 hours. The developing levain remained at room temperature (68°F to 72°F) for the entire duration.
Third levain: 20g seed starter, three 1:1:1 feedings of first clear flour, initially and at eight hour intervals. Harvested 500g of levain after 24 hours. This levain was held at room temperature for the first eight hours, approximately 82°F for three hours, and 89°F for the remaining 13 hours. These temperature choices reflect the findings reported in the first reference: optimum yeast activity occurs at approximately 82°F; optimum bacteria activity occurs at approximately 89°F. Additionally, yeast and bacteria activity are approximately the same at room temperatures, yeast activity falls dramatically at 89°F.
Subjective Results:
First of all, these were not meant to be controlled, scientific experiments. To the contrary, what i wanted to explore was, "Can a home baker influence the flavor profile of his or her doughs, guided by scientific results, with only those tools common to a baker's home kitchen?". In my case, a small, lidded plastic box,for the developing levain; placed inside a larger, lidded plastic box (my dough proofing box) to minimize the effects of drafts; all placed inside an oven with a manually controlled oven light, to vary the oven's temperature); and a thermometer, aledged by the manufacturer to be accurate to +/- 1°F).
Furthermore, the only way I could test a finished bread's sourness was by tasting it. (in the laboratory they measured the amount of lactic and acetic acid produced.). My taste would be suspect: I was hoping for discernable sourness with the first and third levains; I would taste discernable sournesss with the first and third levains. So, I asked my wife to taste the finished breads. She had no knowledge of the differences in the levain, nor what my expectations were.
The results are a bit anticlimactic:
We both found breads made with the first and third levains had discernable "tang"; in part because we didn't taste them side-by-side, niether she nor I could state with any certainty one was "tangier" than the other.
The bread made with the second levain, fed with all purpose flour, didn't have any "tang". Good bread, but no sourness.
Next steps:
I'm building a proofing box, wherein I can control temperature better than with the oven light. When its finished, I'm going to push a levain to favor only bactieria growth, and add commercial yeast to the dough for gas production.
Here's a picture of the most recent (third levain) bread.