Reusing yeast - how many generations?

[specialkayme]

In everything I've read and listened to, from those that advise that it is possible to reuse yeast, they advise you not reuse the yeast past X generations (some say 5, some say 10). When reusing a yeast cake, I get it. But when using an overbuilt starter that is split in two, what risk truly exists from using the yeast 50 generations?

I understand that contamination may occur, and mutation may occur, but breweries in Europe have been reusing the same yeast strain for literally thousands of years without apparent issue. How is it that a farm in Norway can reuse the same yeast strain for a few hundred years (going back before they knew anything about yeast's existence, let alone contamination or sanitation), but as a homebrewer in the US if I reuse the yeast beyond 10 generations I"m likely to have a problem?

What am I missing?

 

[MGamber]

I overbuild with a 1.040 gravity starter, save about 1/3rd of it and pitch the rest and I've gone well beyond 10 generations without a noticeable problem. As long as you're careful with sanitation and storage, I don't see an issue. Reusing yeast from a big beer can be problematic for the reasons you cited and also because the yeast are just pooped when it's done. A local brewpub will do a 5% beer, use that yeast to do a 6% beer and then use it again to do an 8% beer. After the 8% beer they toss the yeast because it's shot.

 

[Vale71]

Any beer before 1880 was more or less a lambic/sour fermentation, so it's not correct to say that breweries used the same strain for centuries. Modern strains were isolated as single-cell culturing techniques were developed, any beer brewed before that was likely very different from its modern version.
The problem with reusing yeast indefinitely is that your "yeast" will veer more and more towards an undefined lambic/sour blend and the beer will gradually lose its character. Worst case of course is you catch something nasty and then you really have to throw everything away.

 

[4of7]

I agree with you. How many times, Do big brewers just throw it all away and start over again.. I bet someone reading this job at the brewery is the fermentor.

 

[yourlastchance89]

From what I've read, most breweries if they do repitch, will repitch a strain around half a dozen times before starting anew. Ronald Pattison was on the BrewSmith podcast not long ago discussing post-war brewing in the UK. In that interview he mentioned a specific brewery that had been repitching the same yeast from batch to batch for over 50 years! It's also not that crazy of an idea when you consider this has been what mankind has been doing basically since the inception of beer.

 

[specialkayme]

Any beer before 1880 was more or less a lambic/sour fermentation

Do you have any type of source for this?

This runs counter to my understanding of the evolution and history of most of the beer styles I know of, including Pale Ales that date back to the early 17th century, IPAs that date back to the mid 1700's, not to mention the Norwegian and Swedish ales that involve the soaking of yeast rings or "magic spoons" traded down in families for literally hundreds of years, some of which test for zero bacterial presence today. I wouldn't consider any of these as "lambic or sour" type beers. But of course I wasn't around when they were brewed, so who knows.

 

[balrog]

I try really hard to be careful with cleanliness and sanitation, but I eventually get a problem after 6-8 overbuilds. I only brew about 15-18x a year, with 3-4 different yeasts so they will sit a while. Since I overbuild starters, saving about 1/3 for next starter, it's a little cleaner than saving from fermenter. But also I don't get the bacteriacidal hop that ferment-saved yeast has. But for me, after 6-8 times I will see a difference in the way the starter, um, starts, or I'll open a saved third overbuild and it smells like the underside of a dead yak, but I'll know it's time for the new stuff.

 

[specialkayme]

I try really hard to be careful with cleanliness and sanitation, but I eventually get a problem after 6-8 overbuilds. . . I'll open a saved third overbuild and it smells like the underside of a dead yak, but I'll know it's time for the new stuff.

Which is in line with what almost everyone else I've listened to or read from says. After between 5 and 10 generations, the yeast goes bad (or, some say chuck it out when you hit that time period to avoid the issue).

But why does it happen?

Clearly your sanitation and cleanliness is better than what was in existence in the 1700 or 1800's. They didn't know what yeast was, didn't have pasteurization, and didn't sterilize a thing. The water they were drinking (in many parts of the world) was undrinkable unless it was part of beer or wine. The possibility for contamination existed probably a thousand fold more a few hundred years ago than it does today. But somehow, today you can only make it 6-8 generations before it "goes bad" but the underlying strains survived, generation after generation, for thousands of years.

How is it that dipping a wooden block in a yeast cake and drying it out is more sterile, and produces more generations of untainted, unmutated yeast than a sterile yeast starter with an airlock?

 

[Vale71]

How is it that dipping a wooden block in a yeast cake and drying it out is more sterile, and produces more generations of untainted, unmutated yeast than a sterile yeast starter with an airlock?

Because it simply doesn't. The only thing that's changed since then (besides technology) is consumer expectations.
Nowadays we are so used to modern beers that most people will find a lambic beer simply disgusting and undrinkable. Back when that was the only beer available and people had no choice nobody complained about it because there simply was nothing else to drink.
Just an example: in the Kingdom of Bavaria brewing of lager beers was only allowed between the months of October and March. In the warmer months the only beer that was brewed was wheat beer, which was the only top-cropping beer brewed in that region. The reason for that is simply that in the warmer months the top-cropping yeasts and other assorted beasties prevailed on the bottom cropping yeast and the beer changed character drastically. Beer brewed in the colder months had a much cleaner character but was still far from what we expect today of a lager. And this is how the Märzen beer was born, it was simply the last lager beer brewed in the month of March (=März), lagered in the ice cellars and then imbibed in the warmer months as the only alternative to Weizenbier.
Until pure strains where made available thanks to the work of Pasteur every beer was fermented by a mixture of yeasts and other organisms, ambient conditions and process did influence fermentation but to a much smaller degree than what we can achieve today. Beer styles were mostly defined by grist, process and water composition and much less by fermentation.

 

[stz]

Few different ways to look at it ...

Until there is a problem there is not a problem. Problem is subjective! I work in a commercial brewery and I've pitched the same yeast over 200 times before renewing. The reason why it was renewed was it just started to struggle to fully attenuate over the required timespan. Leading up to that decision we were rousing, increasing the temperate, slightly increasing the pitch rate etc to get the beer finished in time. Probably could have still used it, we make a lot of different products and if everything stays pretty much the same you don't get quite the variation over time.

What problems occur? Your yeast can pick up contamination which is pretty much impossible to avoid. You then grow that contamination up alongside your regular yeast until it becomes a big enough contamination is cause quality issues within the desired shelf life. I used to think that one lactic will rapidly multiply to cause problems, but they kind of don't. A small amount are always present, but don't really take over, all the usual stuff keeps the population fairly suppressed even over many generations. Same with wild yeast. Nothing can compete with the pitch rates we use and cask beer is just ... drunk super quick. Under a microscope often find the odd lactic, not enough to cause a problem until 6-9 months.

Genetic drift is a thing, but also kind of not. Some traits are gone from the strains genome and aren't coming back. There is also this self selecting thing where a quantity of yeast will contain 50% of generation 0, 25% of generation 1, 12.5% of generation 2, 6.25% of generation 3, 3.125% of generation 4 and so on out until generation 6 where the numbers start to get so small. Yeast leaves a scar on its cell wall each time it buds and out past generation 6 the yeast has such a damaged cell wall it struggles to correctly transport things into and out of the cell walls. This means a population eventually becomes self limiting in regard to number of generations. Each time you pitch most of the cells are from the first few generations and they produce the majority of the next generations so drift isn't as bad as you think.

People toss yeast at generation 6 because of the cell membrane thing, not because it has genetically drifted by this point. As long as you keep your culture clean you can theoretically use it indefinitely, until you have problems, then you have problems.

Anyway. What else? You can acid wash yeast occasionally. pH 2.2 for 20m with phosphoric acid is standard. We do cell counts and viability on every beer on package, but not on pitch unless we are investigating problems. Our process is the same every single time and we know how many cells are in 1g of slurry, what the viability is going to be based on storage (tends to lose 5% per day under barm ale, 2-3% under distilled water @ 4C) and so on.

If I've been using an interesting yeast I'll run off slurry into a sanitised drum and top off with distilled water. I've got a bunch of these hanging around the brewery. Occasionally I want some for something and I think ... man this is 2 months old, the viability is shot, but more often than not it isn't that bad from the fluid layer, the stuff at the bottom is dead, but a slight swirl, take off some liquid, typically 100x10^6 cells per ml at 70-80% viability, usually put 100ml into a flask with 2L of fresh wort and spin it up overnight if I want it for a homebrew.

 

[specialkayme]

Because it simply doesn't.

Oh, but it does. Samples taken of various wooden yeast rings in Norway that contain Kviek, yeast that has been handed down from family to family, generation to generation, and neighbor to neighbor, showed ZERO bacterial or other infections. Not all of them were that clean, but some were. The fact that any were completely devoid of bacteria or other infections is nothing short of astonishing. That's literally hundreds of years of no pasteurization, and pure yeast strains that continue to thrive and produce great beer, with no off flavors. 10 generations of overbuilt starts can't do that. That doesn't make sense.

Your yeast can pick up contamination which is pretty much impossible to avoid. You then grow that contamination up alongside your regular yeast until it becomes a big enough contamination is cause quality issues within the desired shelf life.

And it makes sense. All it takes is one single cell of something off to create an eventual chain reaction. But if that's the case, how did yeast survive from 9,000 BC to 1880? It seems that 11,000 years of opportunity for contamination to occur should have destroyed any opportunity yeast had to survive. Or should have all but guaranteed that what they were brewing with in 1750 was 99.9999999% not yeast. And yet . . .

There is also this self selecting thing where a quantity of yeast will contain 50% of generation 0, 25% of generation 1, 12.5% of generation 2, 6.25% of generation 3, 3.125% of generation 4 and so on out until generation 6 where the numbers start to get so small.

I may not be understanding this right, but wouldn't that be the other way around? Shouldn't it be 50% generation 6, 25% generation 5, 12.5% generation 4, and so on? When I make a yeast starter, I start with a small pile of yeast. At the end of the starter, I have a big old pile of yeast. The original yeast that began my starter are still present, but the rest were created in the starter. So there is more new yeast than old yeast. Or am I not understanding something?

People toss yeast at generation 6 because of the cell membrane thing, not because it has genetically drifted by this point.

So how do yeast companies do it? I'm assuming when you buy a yeast sample, you aren't getting old yeast that has worn out cell membranes. I'm assuming you're getting mostly younger, healthy yeast capable of budding new yeast cells. But the yeast companies are doing just what you're doing, only on a scientific scale. They take yeast cultures, develop them into slants, turn slants into starters of various and increasing size until they have buckets of yeast. Then they take some samples from the buckets of yeast and create new samples from it and restart the process, right? How is that different from what the homebrewer is doing, other than scale (and sanitary procedures)?

 

[Vale71]

Generation 0 means 0 buds, gen 1 one bud, gen 2 two buds and so on and so forth. The concept of generation differs when you reproduce asexually.
The difference between the starters that we make at home and what White Labs and all the other labs do is that they always start with a single cell and then work in a very sanitary environment in order to guarantee very low levels of contamination.

 

[cubalz]

At the Brewery I work at, we typically use most yeast for 4 generations. At that point most start to underperform. When I brew at home, I will only go to 3 generations.

 

[stz]

How do yeast companies do it?

Ok... lets say I have a billion cells on the cusp of death. They bud and every single one of them drops dead. I have a billion cells again, except these ones have fresh cell membranes and are good to go another 6 generations. That billion becomes 2 billion, 1 billion at generation 1, another billion at generation 0. Those 2 billion bud and become 4 billion, 2 billion at generation 0, 1 billion at generation 2 and 1 billion at generation 1. That 4 billion bud and become 8 billion, 4 billion at generation 0, 2 billion at generation 1, 1 billion at 2 and 1 billion at 3. That 8 billion bud and become 16 billion, 8 billion at generation 0, 4 billion at generation 1, 2 at generation 2, 1 at 3, 1 at 4 and so on and so on. So you can see that if they die off at generation 6 you don't lose the majority of the population, it naturally attenuates off past this point and the largest part of the population is always fresh yeast.

The hard part is working from a slant. Once you've got a pretty decent population it is WAY easier. Typically slant to brewery pitch takes about two weeks to prepare. Propagating yeast is quite different to making a starter, you keep it aerated, aim to keep it in a constant growth phase, best way is to slowly add fresh aerated wort throughout.

 

[TallDan]

Oh, but it does. Samples taken of various wooden yeast rings in Norway that contain Kviek, yeast that has been handed down from family to family, generation to generation, and neighbor to neighbor, showed ZERO bacterial or other infections. Not all of them were that clean, but some were. The fact that any were completely devoid of bacteria or other infections is nothing short of astonishing. That's literally hundreds of years of no pasteurization, and pure yeast strains that continue to thrive and produce great beer, with no off flavors. 10 generations of overbuilt starts can't do that. That doesn't make sense.

Survivorship bias.

https://en.wikipedia.org/wiki/Survivorship_bias

Over hundreds of years, they threw away millions of pieces of wood that made bad beer and kept the few that made good beer. They decided that those few pieces were "magic" or "blessed" and were very careful not to lose or damage them.

Over hundreds of years, there were probably an uncountable number of failed farmhouse breweries. Of course, they didn't understand why they failed, but while Weihenstephan was turning out tasty beer close to a thousand years ago, you can be sure that they weren't the only ones trying. They were just the only one that by luck, by the blessing of god, or by better brewing practices, happened to make better beer and stuck to their traditions.

 

[TEWNCfarms]

From what I've read, most breweries if they do repitch, will repitch a strain around half a dozen times before starting anew. Ronald Pattison was on the BrewSmith podcast not long ago discussing post-war brewing in the UK. In that interview he mentioned a specific brewery that had been repitching the same yeast from batch to batch for over 50 years! It's also not that crazy of an idea when you consider this has been what mankind has been doing basically since the inception of beer.

Gonna have to check that out

 

[Northern_Brewer]

Any beer before 1880 was more or less a lambic/sour fermentation

Lambics typically have total acidity of around 1.1% - here's some beers from before Hansen isolated single cultures in 1883 -German wheat beers of 3% ABV or less below 0.2% TA and dunkels averaging <0.15% lactic, barely more than you'll see today.

On the other hand, some of the stories about repitching for decades need to be taken with a pinch of salt. Yes there are some - Lees are approaching generation 5000 with their yeast, and there's a couple more like that. OTOH, there's a lot of British breweries whose marketing departments talk about using "the same yeast" for decades, but in fact they mean that they periodically get it out of the freezer/NCYC and grow it up from scratch.

There's two main risks to yeast - infection and mutation. Infection seems to be relatively rare in farmhouse brewing - things like lactobacillus are super-sensitive to hops (dying at 5-10 IBU) and the natural acidity and alcohol of beer takes care of most of the rest - it's a pretty hostile environment. Things like juniper will also contribute anti-microbials, and if the yeast express killer factors that will also help keep wild yeasts at bay.

Crucially, if a traditional brewer does get an infection, he just goes to his neighbour or someone known locally for having a "good" yeast and borrows some from them. It seems to happen reasonably regularly among the kveik brewers - maybe once every 20-30 years? So the good yeasts spread and the less good ones tend to die out - selection in action. Something similar happens among commercial brewers - for instance it seems that the Edinburgh brewers exchanged yeasts so often that you can't really talk of a distinct eg "Youngers" yeast, there was just a collective pool of yeast for the whole city.

Mutation is harder to manage, but the main counters to it are having a nice diverse population of yeast and beer. The typical homebrewer is very vulnerable to mutation as they're typically growing just a single strain, which when it goes wrong tends to go wrong as a population, all at once. That makes it really obvious at a bulk level. In fact what seems to happen is that you appear to get waves, as all of the population goes "wrong", but then comes back, then goes wrong again. That's less of a problem in a typical British brewery multistrain, as you have different strain within it that don't mutate in sync in the same way so the bulk behaviour is steadier.

But there are some really interesting genetics going on during repitching, and as genome sequencing has come down in price people like this guy are investigating it : https://scienceinseattle.com/2018/11/28/chris-large-talks-brewing-yeast-evolution/ He's seen chromosome duplications and a particular mutation happening in hundreds of separate sublineages as it meant losing a "bad" bit of DNA.

So homebrewers don't have the balancing effect of a multistrain, almost all homebrew beer strains don't have killer factors, and homebrewers tend to buy new when there's a problem rather than using storage or borrowing from friends. Sanitation may be better in general, but "contamination" by other unwanted Saccharomyces may be more of a problem for yeast made in yeast labs - the White Labs diastaticus scandal is one example, but more generally dry yeast in particular seem to have "friends" in the pack - or may be deliberately mixed with poor-performing yeast that take over after a few generations.

So there's a lot going on - but there's a few suggestions.

Adnams lost their yeast during WWII and got a replacement from

 

[balrog]

...Adnams lost their yeast during WWII and got a replacement from

@Northern_Brewer--a great read until then end, like a French movie, just stopped! I'm sitting here in feverish anticipation of the rest of that sentence!

 

[JohnSand]

Adnams lost their yeast during WWII and got a replacement from

Tune in next time?

 

[Northern_Brewer]

Ah, sorry - I was trying to keep it brief and left a fragment below the line. But Adnams are a very typical example of a regional brewer from a yeast point of view. They lost their yeast to an infection in 1942 and replaced it with (allegedly a Whitbread B derivative) from Morgans of Norwich.

In the 1970s they cut it down from a 5-strain to a 2-strain. I assume this was part of a move to conicals - bottom-cropping makes it much harder to manage a multistrain and many breweries went to a single strain when they moved to conicals around this time (eg Fuller's in 1976).

If WLP025 Southwold and 1335 British Ale II are both from Adnams as the internet claims, then they look like the two halves of the 2-strain - the former more flavourful, the latter having better brewing performance for attenuation/flocculation. That's a pretty common arrangement in British breweries - supposedly Windsor and Nottingham came from the same brewery and would represent a similar best-of-both-worlds combination. It's really quite unusual to have One Yeast To Rule Them All, but homebrewers tend to think in terms of single strains and just put up with the fact that eg Chico has pretty crappy flocculation.

 

[specialkayme]

Survivorship bias.

If we were talking about a single success story, I'd completely agree with you. Maybe even a dozen isolated situations. But we aren't. We're talking about SIGNIFICANTLY greater quantities of success stories. How many yeast strains does White labs sell? Maybe a hundred? Then there are all the strains in their yeast vault that they maintain, but don't sell, or sell to private clients and breweries but not the general public. How many are there, maybe two hundred more? Probably more, but still. So that's three hundred individual success stories. That's three hundred strains of yeast that likely weren't "created" after 1880. That doesn't count the several thousand cases where White labs received a strain that survived several thousand years of selection, but the yeast just produced off flavors and wasn't suitable to maintain. The yeast collected off the islands of Estonia is a perfect example. That also doesn't count the several thousand yeast strains not maintained by White labs, or their competitors. But lets ignore all that for a second. Lets say in the history of man-kind we have 300 successful "survivors."

What are the odds of having one survivor? Lets say, at home and under somewhat reasonable sanitary conditions (but not laboratory conditions), I can reuse a yeast 6 times before something goes wrong (either infection, mutation, lazy yeast, whatever). That's roughly a 15% chance that each batch will have something go wrong. Or put another way, an 85% chance that I can successfully reuse a yeast from one batch to another. Now lets also assume I brew once a month, with the same yeast strain. After two years (24 batches) there is a 2% chance my yeast is still healthy. After 315 brews, there is a 0.00000000000000000000001% chance my yeast is still healthy. That's one in ten sextillion odds. 315 brews represents 26 years of brewing, at one brew per month.

Now I can point to Weihenstephan and say they produced beer for a thousand years. Lets assume that is our outlier. But how many breweries, in the past 300 years, operated consistently for more than 26 years? Maybe a thousand? I'd wager it's SIGNIFICANTLY more than that, considering the number of yeast strains we have, the fact that these yeast strains were developed all around the world, independently, in households and breweries alike. But let's say it is a thousand.

Each one of those thousand situations had a one in ten sextillion odds of success over any one 26 year time period. The odds each could do it over a 26 year time period? Then the odds that each could do it repeatedly over a 300 year time period? Well frankly we're getting beyond my ability to calculate odds, it's so incredibly small.

But we aren't talking about a 26 year time period. We're talking about a few hundred, or a few thousand year time period. But lets just look at that 26 year time period. Each brewery, each house, each community had a one in ten sextillion odds of success, but each one overcame the odds. Except they did that without pasturization, without sanitation, without having a clue what yeast was, and while (in many situations) tossing literal **** into beer (although realistically speaking those probably didn't make the cut). That 85% chance of success is an unreasonably high estimate. But still. All of these breweries succeeded, and yet I can't make it past 6 months of continuous brews without things going bad. How is that possible?

Now you look at the 1,000 breweries I claim as successful, and say there are countless others that were not successful. But if it is one in ten sextillion odds of success, and we have 1,000 cases of success, you should have somewhere around a hundred sextillion (at the very least) cases of failure. That amounts to roughly one trillion failed attempts for every single man, woman and child on the face of the earth today (or one trillion times ten billion) in each 26 year time period of human history. One trillion failed attempts in a 26 year time period accounts for one thousand two hundred failed brews per person alive today PER SECOND.

So what does that mean? History of human (and yeast) evolution tells me the odds of actual success aren't 85%. It should be something much higher. Maybe 99%. Maybe more. But whatever the odds are, I should be able to go longer than 6, longer than 10, longer than 100 brews before something goes wrong.

I wouldn't consider that survivorship bias.

 

[Northern_Brewer]

That's three hundred strains of yeast that likely weren't "created" after 1880.

Actually a lot of them are way more recent than that. Just think of the way that WLP001, 1056, US-05, WLP090, Pacman have all diverged from Sierra Nevada within the last 30 years or so. There's been an even greater radiation from Whitbread B in the UK as it was one of a few strains that coped well with the move to conicals. Most beer these days is lager, which is a) a relatively recent development and b) hugely dependent on just a handful of strains, particularly 34/70.

Looking at recent genome sequencing, the current range of homebrew Saccharomyces probably only have mebbe 10 "roots" before 1880.

Your 1 in 6 thing assumes you're on your own, which historically you wouldn't be. If you share your yeast with 5 friends all with a 1in 6 failure rate then your yeast only dies out if all 6 of you have a disaster - a 1 in 46,456 chance. Assume more friends or a lower failure rate (ie closer to what is observed empirically with modern farmhouse brewers) and the odds of a complete loss become ever more remote.

 

[balrog]

They lost their yeast to an infection in 1942 and replaced it with (allegedly a Whitbread B derivative) from Morgans of Norwich.

Thanks for the resolution.

... the former more flavourful, the latter having better brewing performance for attenuation/flocculation. That's a pretty common arrangement in British breweries - supposedly Windsor and Nottingham came from the same brewery and would represent a similar best-of-both-worlds combination. It's really quite unusual to have One Yeast To Rule Them All...

Soooo, if I like WY1450 Denny's but want to get fast floc I throw Notty or something in with it, at start or end of 2-3d typ ferment?

 

[Northern_Brewer]

That's the general idea, you put the "flavour" strain in first, then the "performance" strain in maybe 48h later and it should drop out the flavour strain with it. Windsor followed by Notty is a classic example.

 

[specialkayme]

Actually a lot of them are way more recent than that. Just think of the way that WLP001, 1056, US-05, WLP090, Pacman have all diverged from Sierra Nevada within the last 30 years or so.

Are any of those new strains though, that didn't exist 30 years ago? Or are they isolations of strains that previously existed, but not empirically cataloged?

No doubt some new strains surfaced since 1880. But I would venture to say not most. But a wild guess, for sure.

Your 1 in 6 thing assumes you're on your own, which historically you wouldn't be. If you share your yeast with 5 friends all with a 1in 6 failure rate then your yeast only dies out if all 6 of you have a disaster - a 1 in 46,456 chance. Assume more friends or a lower failure rate (ie closer to what is observed empirically with modern farmhouse brewers) and the odds of a complete loss become ever more remote.

Actually it doesn't. The 1 in 6 thing assumes that you can make it 315 consecutive brews without losing the strain. It doesn't have to be done by one person. You can have 50 brews done by person A, then they hand the strain over to person B who does 50 more brews before person A's yeast dies out, then they hand that strain back over to person A who does 215 new brews. The odds are still one in ten sextillion that the yeast can make it that far without a problem (contamination, mutation, weak yeast, whatever).

Now, that's one in ten sextillion that that specific strain survives. Not that any yeast survives.

If person A gives his yeast out to 10 people, that strain still has a one in ten sextillion chance of making it 315 consecutive brews with any of those 10 people. If 9 lose the strain after 100 brews, and the sole survivor spreads the yeast back out to the other 10, it doesn't impact the survival odds from brew one. It just impacts the survival odds from brew 101 to 315. The odds of that strain surviving another 215 brews is then 0.0000000000000668%. But that was true regardless of whether that sole survivor reshares the yeast or not. If that sole survivor lived in a bubble, after 100 brews the yeast still had a 0.0000000000000668% chance of surviving the next 215 brews. Resharing impacts future brews, but doesn't change the overall odds.

 

[Northern_Brewer]

Are any of those new strains though, that didn't exist 30 years ago? Or are they isolations of strains that previously existed, but not empirically cataloged?

No doubt some new strains surfaced since 1880. But I would venture to say not most. But a wild guess, for sure.

You guess wrong. Looking at the DNA, much of the apparent diversity of homebrew strains appears to have come about since WWII. The Chicos are a good example - they all appear to have their origin in slants sent from the Seibel Institute to Sierra Nevada in the 1970s. Since then people have harvested yeast from different bottles and they've evolved and we can now see measurable differences in brewing performance. These are new strains that didn't exist before Sierra Nevada (as far as we can tell). The Whitbread family is even more diverse.

Actually it doesn't. The 1 in 6 thing assumes that you can make it 315 consecutive brews without losing the strain. It doesn't have to be done by one person. You can have 50 brews done by person A, then they hand the strain over to person B who does 50 more brews before person A's yeast dies out, then they hand that strain back over to person A who does 215 new brews. The odds are still one in ten sextillion that the yeast can make it that far without a problem (contamination, mutation, weak yeast, whatever).

You're assuming a sequential process - 315 brews one after another by a single holder of the yeast.

Now imagine that you have six people with a 1 in 6 chance of losing their yeast. They brew once a year, on Christmas Day. The first year with average luck, one of them will lose their yeast. In your model, that means only 5 people brew the next year. In reality, one of the 5 give their yeast to the "failure" and you're back with a team of 6 the next year. The system has reset itself. Only if all 6 of them fail on the same day will there be no yeast the next year, the system is more resilient than you make out.

And in reality they might not all brew on the same day so if 3 brew on Christmas Day and 3 on Easter Day, even if they all have a failure that year, some of the Easter failures could scrounge some yeast from the Christmas brewers during the year, before the Christmas brewers all fail.

 

[specialkayme]

You guess wrong. Looking at the DNA, much of the apparent diversity of homebrew strains appears to have come about since WWII.

I'll have to take your word for that. I have no idea about the DNA research of current yeast strains.

Not derogatory at all. I appreciate your contribution and am just saying I'm going to have to assume you're right, as I have no independent knowledge of that.

You're assuming a sequential process - 315 brews one after another by a single holder of the yeast.

No, I'm not. See post #25. I'm not talking about 315 brews by one holder of the yeast. I'm talking about 315 consecutive brews of the yeast, regardless of who's brewing it or how many hands it changed.

Now imagine that you have six people with a 1 in 6 chance of losing their yeast. They brew once a year, on Christmas Day. The first year with average luck, one of them will lose their yeast. In your model, that means only 5 people brew the next year. In reality, one of the 5 give their yeast to the "failure" and you're back with a team of 6 the next year. The system has reset itself. Only if all 6 of them fail on the same day will there be no yeast the next year, the system is more resilient than you make out.

We're comparing apples to oranges.

So lets say you start with six people. They brew. One loses their yeast. Five still have theirs. What are the odds that one of those five can continue to brew 314 more batches without their yeast going bad? About one in nine sextillion. Significantly better than the one in ten sextillion that existed before brew 1. Why the difference? Because that strain now has to go 314 brews without an issue, rather than the original 315 brews. But the new odds doesn't change the odds that existed on brew 1. It calculates new odds for brew 2. The odds decrease with each successive brew, but they are still large, and never go to zero for that particular yeast line.

So lets look at some new math then. Lets say that we have six brewers, all sharing the same yeast. They brew a batch on the same day, the first of every month. Each brewer has an 85% chance of keeping healthy yeast alive at the end of the brew. If statistics worked the way we hoped they would, one brewer each time would lose their yeast, and would then get replenished by one of the other five. And all is great in the world. But statistics doesn't work that way. Sometimes no one loses yeast. Sometimes one loses yeast. Sometimes four loses yeast.

So what are the odds that no one loses their yeast in that scenario? 37.7% (.85^6). What are the odds that they ALL lose their yeast in the same brew? 0.00113%. Small, yes. But not zero, right? Now lets say all six families brewed on the same day of the month in perpetuity. If statistics worked the way we wanted them to, it would take those six families 8,333 years to come across one of those rare occurrences where all six lose their yeast. But again, statistics don't work that way. It can happen 5 times this year, or not at all for 100,000 years.

But when we were trading yeast back and forth in tribal villages of Germany, they didn't have a 15% chance of infection. Lets be honest. They should have had a 10% chance of not getting infected, especially when you consider that you dip your "magic spoon" into some goo (that wasn't always boiled) and hang it up on the rafters of the barn when you're done. Now lets say you have those same 6 families brewing, again once a month. What are the odds that all six will walk away with their yeast when you have a 90% chance of failure? 0.0001%. What are the odds that all six will walk away with no yeast? 53%. More often than not, those six people would walk away without yeast on brew day. If you increase your pool size to 20 people, the odds of having no viable yeast is now 12%. Still, on average once a year you're losing your yeast. And how many families brewed in tribal Germany in the year 900, in one village? Keep in mind, if they lost their yeast, they didn't know what yeast was, let alone that they needed to travel to the next town over to take their new "magic spoon" and dip it into the next tribal villagers "goo." Assuming that they could get authority to do it, which the probably couldn't, or that they didn't take the same "magic spoon" they had before that was already infected and used it to infect their neighbor's beer.

Now there are a ton of assumptions in that exercise. Yes, not all six would brew on the same day, decreasing the chances of a total loss. But there are a ton of things that happen that increase the chances of a total loss. If anything happens to an individual brewer, the yeast is lost. Lifespan was 34 years. Maybe 20 of them would be spent brewing. War, disease, crime, floods, famine, all would negatively impact a brewer which would remove them from the pool of available yeast, in addition to the possibility of infection if they did brew. Or maybe one of the six loses their yeast, but waits a month to get a new yeast culture. In the meantime each of the other 5 performs a brew. The odds of loss are the same as if they all brewed on the same day.

The point is, the odds the yeast survived is small. Very small. Like ridiculously small over a thousand years. One in ten sextillion may be an over estimation. But it's still very small. And yet it occurred.

 

[jshell55]

Few different ways to look at it ...

Until there is a problem there is not a problem. Problem is subjective! I work in a commercial brewery and I've pitched the same yeast over 200 times before renewing. The reason why it was renewed was it just started to struggle to fully attenuate over the required timespan. Leading up to that decision we were rousing, increasing the temperate, slightly increasing the pitch rate etc to get the beer finished in time. Probably could have still used it, we make a lot of different products and if everything stays pretty much the same you don't get quite the variation over time.

What problems occur? Your yeast can pick up contamination which is pretty much impossible to avoid. You then grow that contamination up alongside your regular yeast until it becomes a big enough contamination is cause quality issues within the desired shelf life. I used to think that one lactic will rapidly multiply to cause problems, but they kind of don't. A small amount are always present, but don't really take over, all the usual stuff keeps the population fairly suppressed even over many generations. Same with wild yeast. Nothing can compete with the pitch rates we use and cask beer is just ... drunk super quick. Under a microscope often find the odd lactic, not enough to cause a problem until 6-9 months.

Genetic drift is a thing, but also kind of not. Some traits are gone from the strains genome and aren't coming back. There is also this self selecting thing where a quantity of yeast will contain 50% of generation 0, 25% of generation 1, 12.5% of generation 2, 6.25% of generation 3, 3.125% of generation 4 and so on out until generation 6 where the numbers start to get so small. Yeast leaves a scar on its cell wall each time it buds and out past generation 6 the yeast has such a damaged cell wall it struggles to correctly transport things into and out of the cell walls. This means a population eventually becomes self limiting in regard to number of generations. Each time you pitch most of the cells are from the first few generations and they produce the majority of the next generations so drift isn't as bad as you think.

People toss yeast at generation 6 because of the cell membrane thing, not because it has genetically drifted by this point. As long as you keep your culture clean you can theoretically use it indefinitely, until you have problems, then you have problems.

Anyway. What else? You can acid wash yeast occasionally. pH 2.2 for 20m with phosphoric acid is standard. We do cell counts and viability on every beer on package, but not on pitch unless we are investigating problems. Our process is the same every single time and we know how many cells are in 1g of slurry, what the viability is going to be based on storage (tends to lose 5% per day under barm ale, 2-3% under distilled water @ 4C) and so on.

If I've been using an interesting yeast I'll run off slurry into a sanitised drum and top off with distilled water. I've got a bunch of these hanging around the brewery. Occasionally I want some for something and I think ... man this is 2 months old, the viability is shot, but more often than not it isn't that bad from the fluid layer, the stuff at the bottom is dead, but a slight swirl, take off some liquid, typically 100x10^6 cells per ml at 70-80% viability, usually put 100ml into a flask with 2L of fresh wort and spin it up overnight if I want it for a homebrew.

I've been breweing for about 9 years now.

In the beginning I experimented with this as a home Brewer to save a few dollars.

You can indeed get some or all of the sluge off of the bottom of the primary and add it to the next batch make pretty good beer for about 3 generations. After that there is a noticeable reduction in flavor of the final stuff.

I have also grown yeast from commercial beer (some of those German beers that have a lot of sediment in the bottle) but it never tastes quite the same.

I feel that if you had good temperature control and the right equipment you could actually clone the yeast indefinitely in a laboratory situation but not in a batch of beer because only the real hard core home Brewers have good enough temperature control to do so.

The reason some of these breweries can clone yeast is that very thing. They have such good control of their conditions so the end product has the same character. Me with my carboys in my basement can't really do that. There is always some variability because of just seasonality.

Also it matters what you brew. The IPA types are pretty forgiving. If you try to brew the finnicky saisons or heavily yeast-dependent beer like hefeweisen that are more delicate, it is a lot harder.

So nowadays I just spend the extra few bucks on fresh yeast and don't feel that bad about it.

I would say try it. Your experience may be just fine.

 

[SEndorf]

Interesting discussion with lots of didactic information on yeast strains.
For myself as an average homebrewer, 3 gens is the max. Anything further increases the chance of a wildcard. My time is valuable and pushing yeast to save a few bucks isn't worth it. Commercial brewing has more investment and more at stake.

 

[balrog]

While I appreciate reusing yeast to save money, and certainly that appeals to me, I reuse yeast via overbuild primarily out of a personality character flaw--I like having things on hand whether it be grains for next batch, spare headlight bulbs, or tools. I really hate waiting to get yeast shipped. And locally never have what I want unless it's S05 maybe.

 

[wepeeler]

I think the major question at hand would be are we talking about reusing yeast that has already fermented a full batch or are we talking about harvesting yeast from a starter? I would imagine yeast being harvested from a starter and using that to build another starter and harvesting would be reusable for many more than 5 generations, barring any contamination. It's basically fresh unused yeast every time. Makes sense that washed yeast wouldn't be reusable for as long, because it's put in work.

 

[SEndorf]

Good distinction.
I think most of us are talking about fermented yeast, washed and re-used.

 

[wepeeler]

Good distinction.
I think most of us are talking about fermented yeast, washed and re-used.

That's what I'm getting from reading top to bottom. Seems like the discussion keeps going back and forth. Over building a starter seems to be much easier than cleaning and washing, so that's what I'm interested in. I want to push the max and see if I can get 10+ generations out of over building starters. Looks like @MGamber has had some success.

 

[Kiln]

Top crop

 

[jshell55]

Yes actually this happened to me just the other day. I was brewing a stout and a porter at the same time. Similar enough to each other and not finnicky on yeast. I used cheap Nottingham dry yeast in the porter and expensive yeast in the little package where you break the little nutrient sack for the stout.

Sadly the expensive yeast was DOA. Two days after pitching it was not doing squat. The cheap Nottingham was rockin' however. So I did fearlessly transfer some of the live yeast off the top of the porter and it revived my stout no problem.

I believe live yeast off the top could be gathered much better and more consistently than the dormant yeast at the bottom of the primary.

Also clearly you want to use some discretion. The yeast and other sediment in a batch separate out differently and you do want to avoid getting the nasty proteins and leftovers from the cold break. There is a slight color difference which is easy to see. When I harvested it, I poured some of the sediment into a sanitized ball jar with some water, shook it up, and captured the less dense yeast off the top after it settled out.

Sent from my XT1565 using Home Brew mobile app

 

[JAReeves]

Interesting discussion. I have been using a mix of Safale #05 and White Labs #023 since August (6 months). I have brewed a few IPA's, stouts, red rye, blonde, a barley wine (the yeast pooped out on this one at 14% ABV-I added champagne yeast and am hoping for the best), and the yeast seem to be doing fine. I take large sanitized jars (1-2C) of slurry, from the primary only, and let them rest in the back of the fridge (burp them occasionally). If they sit longer than 1 month, I change out the water. I have been getting ~80% attenuation, good beer clarity, and a nice sediment crop at the bottom of the bottle that I can leave behind on a pour with relative success.

To use the yeast, I pour off the liquid, add fresh water, shake vigourously, allow to come to room temp and throw into the fermenter.

I like the taste of the yeast mix, but someday when I don't, or I get a bad batch, then I can start over with my "house culture." I figure that humans have been doing this in a relatively primitive fashion for millennia, and I am just interested to see where the generations of yeast will go with my beer.

Purely for the advancement of human knowledge, mind, the drinking of the experiments is just a necessary evil in the life of an amateur scientist.

 

[Kiln]

Interesting discussion. I have been using a mix of Safale #05 and White Labs #023 since August (6 months). I have brewed a few IPA's, stouts, red rye, blonde, a barley wine (the yeast pooped out on this one at 14% ABV-I added champagne yeast and am hoping for the best), and the yeast seem to be doing fine. I take large sanitized jars (1-2C) of slurry, from the primary only, and let them rest in the back of the fridge (burp them occasionally). If they sit longer than 1 month, I change out the water. I have been getting ~80% attenuation, good beer clarity, and a nice sediment crop at the bottom of the bottle that I can leave behind on a pour with relative success.

To use the yeast, I pour off the liquid, add fresh water, shake vigourously, allow to come to room temp and throw into the fermenter.

I like the taste of the yeast mix, but someday when I don't, or I get a bad batch, then I can start over with my "house culture." I figure that humans have been doing this in a relatively primitive fashion for millennia, and I am just interested to see where the generations of yeast will go with my beer.

Purely for the advancement of human knowledge, mind, the drinking of the experiments is just a necessary evil in the life of an amateur scientist.

I am on #8 of California ale wlp001 for almost a year. Have been top croping with great results. Charlie Papazian author of The Complete Joy of Hombrewing has been using a lager yeast for literally decades. There is a Chop and Brew episode on YouTube that mentions it. So it's for sure infinite if you are thorough approbation and lavish its praise.

 

[Kiln]

This I why I want a good microscope and microbiology degree.

 

[JAReeves]

I am on #8 of California ale wlp001 for almost a year. Have been top croping with great results. Charlie Papazian author of The Complete Joy of Hombrewing has been using a lager yeast for literally decades. There is a Chop and Brew episode on YouTube that mentions it. So it's for sure infinite if you are thorough approbation and lavish its praise.

I will check out chop and brew on YouTube. So many resources! Damn that day job!

 

[SEndorf]

I am on #8 of California ale wlp001 for almost a year. Have been top croping with great results. Charlie Papazian author of The Complete Joy of Hombrewing has been using a lager yeast for literally decades. There is a Chop and Brew episode on YouTube that mentions it. So it's for sure infinite if you are thorough approbation and lavish its praise.

Interesting discussion indeed.
I reuse my yeast but have not gone this far down the road.
My only concern is that most of my beer ends up with friends and family who request duplication of their particular "family favorites."
Doesn't the yeast take on it's own morphed characteristics after this much time?