In the spring of 2011, about eighteen months into making cheese, I took a sample of whey from a batch that had acidified particularly well and used it to start a propagation culture rather than discarding it. The logic was simple: the milk from our particular herd, grazing our particular pasture, had fermented in a way that tasted right. If that character was partly a function of the bacteria in that milk, keeping them alive seemed worth trying.
That culture is still running. We use it as the primary starter for every belegen and oud batch we make. I have not been able to prove scientifically that it produces a better cheese than a commercial mesophilic culture would — the variables are too numerous and I do not have a laboratory. But in fifteen years of comparison tastings, it consistently produces a cheese with more complexity in the middle notes than the commercial culture we keep as a backup.
What a back-slop culture is
A back-slop (Dutch: terugzetten) is simply the practice of reserving a portion of active whey or curd from one batch to inoculate the next. It is one of the oldest methods in fermented food production — the same principle underlies traditional sourdough and many kefir and yogurt traditions. The culture evolves over time as it adapts to the specific substrate, temperature regime, and environment where it is maintained.
In commercial cheese production, back-slop fell out of use because it introduces variability. A commercial mesophilic culture — typically a defined blend of Lactococcus lactis subsp. lactis and cremoris — acidifies predictably, hits known pH targets at known times, and behaves consistently from batch to batch. For a factory producing thousands of wheels a week, that predictability is worth more than any flavour advantage a house culture might offer.
For a small operation making fifteen to twenty wheels per month, the variability is manageable if you pay attention.
The propagation routine
The culture is maintained in two parallel lines — a working pot and a reserve. After each cheese make, we transfer roughly 200 mL of whey from the vat into a sterilised glass jar, seal it, and refrigerate. This becomes the inoculant for the next batch, typically made three to five days later.
The reserve line is refreshed weekly whether we are making cheese or not. We heat a litre of whole milk from our own tank to 72 °C, hold it for fifteen seconds (a minimal pasteurisation — the culture needs a clean substrate to propagate into, not a competing microflora), cool it to 28 °C, inoculate with 2% from the reserve, incubate at 28 °C for twelve hours, then refrigerate.
The working culture is not pasteurised — it propagates directly into warm whey from the previous batch. This means it carries whatever was in the vat that day, which is occasionally a source of surprises.
Two failures and what caused them
We have lost the working culture twice. Once in 2015, when a bacteriophage contamination wiped out most of the lactococcal population in about four days — the tell was a dramatically slowed acidification that I initially attributed to cold weather affecting the cellar temperature before I tested the pH properly. We fell back to the reserve, which had not been exposed to the phage because it is propagated on pasteurised milk.
The second failure was in 2020, entirely my own fault. I used tap water instead of borehole water to rinse the propagation jar before one of the reserve refreshes. The tap water in this part of Noord-Holland is chlorinated at a level that is fine for drinking and lethal for a lactic acid culture within about six hours. The jar that week produced almost no acidity. We fell back to the commercial backup starter for two batches while the reserve recovered.
Both failures were recoverable because of the two-line system. If I were maintaining only one culture I would have lost it permanently in 2015.
How it tastes different
I am cautious about overstating this. Flavour perception is subjective and our tasting conditions are not controlled. But consistently, over many years, the cheese made with our house culture at 16 weeks shows:
- A more pronounced development in the first third of the tasting — what I would describe as a grassy-sweet note that fades into the familiar Gouda butter
- A slightly slower acidification curve during the make — we typically hit our target pH of 5.35 at pressing about 20 minutes later than with the commercial culture, which means a marginally longer make day but possibly more complex secondary fermentation
- Greater batch-to-batch variation, particularly in summer when the milk composition changes with the grass — some batches are noticeably sharper than others at the same age
That last point is the honest admission. The house culture produces less consistent results. In a summer with variable rainfall and rapidly changing pasture composition, some batches age differently than expected. We have learned to accommodate this, but it requires more attention during the aging process than a commercial culture would.
"Keeping a culture alive for fifteen years is mostly a matter of not being careless twice in the same week."
The commercial backup sits in the freezer. We use it two or three times a year when the house culture is behaving unusually or when we are recovering from a disruption. The cheese it produces is good. It is not our cheese.