Profoundly Promiscuous and Totally Tasty Tomato Project

William Schlegel wrote:Discovered that in the patch from my home saved half habrochaites seed the ripe fruit is way down in there.

Found a pretty one in there. The inflorescence they were on had mostly aborted so probably an intact S allele.

Processed these two tomatoes for seed today. Only a couple super viable looking seeds in them. Wish I could tell which plant it was from. There are a few smaller ones if about the same color. Oh and I tried them. They tasted alright, edible. Only half wilds I've processed so far.

Collected the first of the year Solanum arcanum and S. Peruvianum berries today. Also got berries from the three way penellii hybrid, the F1 penellii hybrid and the pureish penellii. Ended up mixing them in the same sandwich bag. Probably all interhybridized anyhow due to proximity. Also found a berry on another F3 penellii.

Hoping to get back into this next spring.  Keep up the good work!

Joseph, I am really excited about the F2 of Big Hill x W4. I'm thinking about skipping a year on many other things to give the F2 growout a lot of space. Maybe 3 steps apart ~ 9 ft or perhaps 4 steps 12 ft? Figure I should be able to grow ~300 plants from the seed I saved.

Does that sound like a good plan?

William Schlegel wrote:Joseph, I am really excited about the F2 of Big Hill x W4. I'm thinking about skipping a year on many other things to give the F2 growout a lot of space. Maybe 3 steps apart ~ 9 ft or perhaps 4 steps 12 ft? Figure I should be able to grow ~300 plants from the seed I saved.

Does that sound like a good plan?

I loved the BH X W? Populations. One of my favorites came out of W4.  BH X W4 produced some lovely flavors, and also some fruits which were described as acidic.

A constant worry for me is maintaining/increasing the s-alleles in this population. Therefore, I would include some of the pennellii crosses in the grow out as well, just for the s-alleles. Also, the descendants of S pennellii, are tending towards being more fruity than those from S habrochaites.

At 9 feet separation, the plants would still be likely to grow together, but it would be easy to tell which is which. I'll probably plant 3 feet apart with rows separated by 9 feet. That'd give me enough separation that I could cull shortly after they start flowering. One year, I grew them in pots until flowering, then only planted out those with promiscuous flowers.

Thanks Joseph. I like your spacing better.

I'm attending the Organic Seed Growers Conference in February to speak about these tomatoes. Here are some graphics that I made for my presentation. I want to learn how to talk about them so that they are understandable, and helpful rather than being too technical or obtuse. I'd welcome feedback or questions to help me refine my talking points.

Domestic tomatoes today are highly inbred. When they were domesticated, they lost 95% of their genetic diversity due to founder's effect genetic bottlenecks. Tomatoes originated in the Andes, and a few seeds from there made it to Mexico. In the process, they shed genetic diversity, and their accustomed pollinators. Thus the plants self-selected for more inbreeding types of flowers. The same thing happened again when they traveled to Europe for more vigorous domestication. And again when a few seeds left Europe to travel into the rest of the world. Then, the heirloom seed conservation movement got involved inbreeding varieties for 50 years or more, with a strong bias against crossing, so tomatoes became ever more inbreeding, and ever less capable of cross-pollination.



In 50 years, a domestic heirloom tomato has exactly one parent. It basically remains a clone of itself as long as anyone cares to grow it. Hybrid tomatoes have two parents. Hybrid tomatoes tend to average about 50% more productive than heirlooms.  In ten generations, a promiscuously pollinating tomato has 512 parents. In 50 years it has 600 trillion ancestors. No wonder the promiscuous tomatoes are such a joy to grow.



When genes are dividing prior to reproduction, DNA molecules split in half, unzipping the pairing. In the process, the strands switch sides 2 to 3 times along the length of the chromosome. This allows for the "independent assortment" of genetic traits. After a few generations, the inbreeding tomatoes have settled into near homozygous gene combinations, and lost half their genes. The promiscuous tomatoes conserve their genes, and create more and more genetic combinations. Here's a graph of what that looks like in the 4th generation.

"The same thing happened again when they traveled to Europe for more vigorous domestication"

Europeans got completely domesticated and actually extremely fancy heirloom tomatoes from the indigenous people of the Americas. They took that material in different directions because of cultural and culinary differences. They did that within the constraints of an inadequate sample.

There are some exserted strains of heirlooms. Extremely recent modern breeders strongly selected against.

Most species in the ~14 species tomato complex were never part of domestication at all. Adding them in creates incredible new opportunities such as your auto hybridizing tomato project. However I would not characterize this as a restoration because the species involved diverged long before humans intervened with any of the species in the species complex. I would instead characterize it as an incredible new opportunity to create a new tomato type with the heightened genetic variation available in the obligately out crossing species.

My story might go thus:

~14 tomato species diverged before human intervention.


Speciation:
One of these a relatively inbreeding species that was not obligate outcrossing coevolved into a domesticated state in ~Equador.

1st bottleneck:
Then a less domesticated subset of that moved to Mexico where it reached ~modern or greater heights of domestication.

2nd bottleneck:
Then ~500 ybp transfer of a subset to Europe with extensive cultural modification. Then return.

3rd bottleneck:
Then ~100 years to present increasingly boring factory type tomatoes suitable for food preservation and shipping produced.

Boring status quo maintenance:
Professional breeders introgress very limited traits to maintain this status quo from ~14 species tomato complex.

Finally some excitement:
Now amateur breeders can create incredibly creative projects with same species complex:

Joseph's neat project:
Que Joseph Lofthouse's new class of tomatoes with obligate out breeding like tomatillos which should be much more resilient.

Anyhoo, that's how I'd spin it.

Also interesting new articles about a new study:

https://phys.org/news/2020-01-evolution-vine-history-tomato-domestication.amp

And

https://www.sciencedaily.com/releases/2020/01/200107170110.htm

And

https://cosmosmagazine.com/biology/unveiling-the-tomato-s-evolution

All for same study.

Hi Joseph,
nice drawings, shows the process very well!

About the following part:

Joseph Lofthouse wrote:
...
In 50 years, a domestic heirloom tomato has exactly one parent.
...
In ten generations, a promiscuously pollinating tomato has 512 parents. In 50 years it has 600 trillion ancestors. No wonder the promiscuous tomatoes are such a joy to grow.
...

That a heirloom has exactly one parent, fully agree, and it is important that it is undrestood.

But the second part, while being theoretically possible at least till the 10th generation, is not true in further generations, as there is some crossing occuring among cousins, great-cousins, inter-generational crossings, etc. Houndreds of of parents 50 years ago, or even thousand, seems more accurate.

And you maybe want to add that as soon as you reach around 1000 individuals per generations, the loss of alleles through genetic drift is in balance with new mutations occuring in the population.

Just to affirm that locality of any breeding population will exert its influence, irrespective of whether or not you have the more heterozygous population scheme of the promiscuously pollinated approach (PPA) or the inbred heirloom approach (IHA). As noted, the former will have, and be likely to maintain, greater genetic diversity than the latter.  Yet at the end of the day, a person maintaining a PPA population in, say, Yuma, Arizona will likely have different allele frequencies after several generations than a person does using the same approach and same founder population if they have been doing their project in Winnipeg, Manitoba, even if both have been diligent about samples size during pollination.  Also, I remain fascinated by what constitutes (genetically speakding) any given heirloom variety today.  If one were to take a small seed sample from all of the independent growers/gardeners around the world of what they call, for example, 'San Marzano' tomato, how genetic diversity would exist *within* each seed lot and how much diversity would exist *between* the different seed lots?  Moreover, even if a population is isolated from others of its species and is inbreeding, to what extent is mutation AND acceptance of DNA from other species in the same Genus driving the creation of new alleles, some of which may offer selective advantage in that locality and/or new traits to the population?  

Great project, Joseph, and good luck at the conference!

John Weiland wrote:Also, I remain fascinated by what constitutes (genetically speakding) any given heirloom variety today.  If one were to take a small seed sample from all of the independent growers/gardeners around the world of what they call, for example, 'San Marzano' tomato, how genetic diversity would exist *within* each seed lot and how much diversity would exist *between* the different seed lots?  Moreover, even if a population is isolated from others of its species and is inbreeding, to what extent is mutation AND acceptance of DNA from other species in the same Genus driving the creation of new alleles, some of which may offer selective advantage in that locality and/or new traits to the population?  

Joseph's project is based partly on the results of research which has found that a single population of Solanum habrochaites has greater diversity at the genetic level than All domestic tomatoes combined.

He has reasoned that this is due to breeding system and has proposed transferring the habrochaites obligate outcrossing system to domestic or a few domestic genes to habrochaites or penellii.

This system should maintain a much higher level of heterozygosity, but also more alleles in general.

So the Marzano metapopulation population is always dramatically less genetic diversity than a small population of habrochaites even with mutations and ocassional outcrossing.

So the innovation inherent in the new proposed system is extreme. It's like comparing a biplane to a jet engine. With Marzano the biplane and a habrochaites/penellii obligate outcrossing breeding system with palatable fruit the jet engine.

Though to keep us grounded, tomatillos already work like the latter.

In my garden tomatillos are the most capable tomato relative. They reliably volunteer and set fruit. If my tomatoes could be tomatillo level I would be satisfied.

William Schlegel wrote:

In my garden tomatillos are the most capable tomato relative. They reliably volunteer and set fruit. If my tomatoes could be tomatillo level I would be satisfied.

Agreed!......It's just amazing what tomatillos tolerate and still come back year after year with generous output, flavor, and hardiness.

No question that Joseph's and like breeding schemes would produce much greater diversity overall than the standard planting and seed saving schemes for most heirlooms.  One day in the future, what would be interesting to add to these studies are the growing of ostensibly homogeneous (genetically) varieties in different regions around the globe and then examining them during different points throughout the season for epigenetic differences, some of which might be specific to that particular growing year and time-point, but others of which may be pointing to an epigenetic 'fingerprint' for that variety in that region.  It may be another factor, along with greater diversity in general, for more rapid adaptation to a potentially more rapidly changing climate.

https://www.sciencedaily.com/releases/2018/11/181108105932.htm

(....this discussion is making me crave spaghetti sauce for dinner!.... :-P  )

Important to bringing epigenetics into this discussion is that the outbred population would have the same epigenetic advantages as the inbred when grown regionally.

So my subset F2 strain of Joseph's Big Hill x W4 I plan to grow out in 2020 has the obligate outcrossing trait, though that may segregate. It also will have two years 2019 and 2020 of epigenetic adaptation to my growing conditions. Notably Joseph's growing conditions have a lot of similarities to my own and I think his stuff comes to me epigenetically preadapted.

The peruvianum population that's been growing in my garden for several years also has both obligate outcrossing and epigenetic change.

Since both the obligate out breeding and inbreeding populations have epigenetics in play the epigenetic question is somewhat neutral to the discussion of differences between the two.

I suspect that while the epigenetic adaptation is useful, longer term adaptation will come from segregating alleles. So the genetic diversity point is central here because it gives us more alleles to work with.

That said. It occurs to me that just like with tomatillos relatively inbred strains of the outbreeding tomatoes can and will be developed. These would probably be uniform for fruit traits like a yellow strain or a bicolor strain or a pineapple flavored strain. However when we grow the different strains together again, just like with tomatillos they will quickly dissolve into a variable grex. Should be a powerful breeding tool. Also, when grown with regular tomatoes, particularly those with exserted stigmas, the regulars should up take a lot of DNA.

Since both regular exserted tomatoes and obligate outcrossing half wild tomatoes were both very common and closely interplanted in my 2019 garden it's possible I could grow out flats of exserted seedlings and pick out by leaf traits 75% hybrids and thus add more to my 75% domestic gene pool. Though I think I have plenty of 75% W4 x Big Hill seed so that might not be worth the effort. I may even have some envelopes in particular earmarked towards this. Might even have done some pollen transferring on likely exserted plants from obligate outcrossers and have those envelopes labeled....

Getting back to Joseph's question of how to present this project.

I think there are several extremely simple and compelling stories here:

When I got a seed savers exchange catalogue for the first and only time for 2006 the tomato section was the largest. Since then tomato color diversity has exploded. However at the genetic level the tomato section is one of the weakest because of inbreeding depression. So that diversity is only surface deep.

However this need not be the case, because of the deep diversity in the related wild species. Some of the wild species have to outcross. Those that do are incredibly diverse at the genetic level.

We should be able to get a uniformly tasty offspring of wild and domestic that retains this need to outcross.

A domestic tomato that has to outcross will be a much more resilient tomato at the genetic level.

I think that's about as general population accessible as I can say it.

I would say figures shown are generally accurate and useful for those who quickly absorb info from figures.

Planted seeds for this project yesterday 3-17-2020 may plant more today of wilder strains.

Planted the remaining trays. Project is back in action as of today in my garden. I need a seventh isolation spot so I guess that will be the back yard.

A Starling forced its way into a gap in the basement window the sump pump hose goes out and I found it helping itself to my promiscuous tomato seedlings. It didn't completely decimate any though the pure penellii are much reduced.

Starling thinning hmpph

I've requested officially seed from TGRC and GRIN for what are supposed to be Solanum peruvianum hybrids. They will probably have lots of sterility issues and will most likely need a healthy bee population that can cross with compatible S alleles from other wild tomato populations (or by hand). If i get plenty of seed i am planning on splitting it three ways. Some for William and some for Joseph. Hopefully like the pennellii germplasm, we can get peruvianum into this project fully and benefit from those genetics.

I'm thinking of dropping my watermelon breeding and most of my pea breeding next year to devote entirely to wild / domestic tomatoes.

For this year in terms of wilds i think i planted joseph's best wilds and fairy hollow together. I also planted joseph's improved peruvianum, S. galapagense and S. cheesmaniae interplanted, and my suspected pimpinellifolium x habrochiates cross. We will see what grows best and shakes out this year.

I've focused this year on the 3/4 domestics that seemed so promising in Joseph's garden last year. The best of them are going into the fenced garden. Have some planted and some more seed planted (G3) and garden space held for those. Waiting for germination!

The seed I grew of 3/4 wilds went into the general population as a frost test, a regular plant out, and a direct seeding. I expect it to segregate into edibility at a higher rate than before but not as high as the plants that were best from 2019 in Josephs larger growout.

Then I also in the general population have some pure habrochaites of a new strain, pure peruvianum, and pure arcanum. I have what I believe to be high percentage penellii and high percentage habrochaites from wild maternal cytoplasm as well. Also a three way cross Joseph sent that has domestic maternal cytoplasm.

Then I have a mini population with lots of penellii and one peruvianum. I hope that it will result in a good amount of presumptive penellii x peruvianum seed with peruvianum as the mother for 2021.

I haven't grown any more chilense this year which is probably the most promising germ plasm for crossing with peruvianum I have. It's also the tomato hardest for me to cultivate based on my 2019 results which was plants that died young before blooming.

I also have all thoughts about making embryo rescue peruvianum on hold till a time when I have more time.

I'm very curious about the crossed accessions peruvianum x domestic Andrew mentions. Particularly in terms of generation?

I intend to plant an exserted domestic right next to the Arcanum but not sure if I'll get it done or be able to grow the resulting seed enmasse if I do. A controlled Arcanum x domestic cross should be possible and would result in another interesting line of its own. However is not reportedly a good bridge after all but rather something interesting in its own right.

Roads not travelled this year include not growing many of the exciting penellii germ plasm from last year yet. Though there is penellii in the 3/4 domestic lines and Joseph thought it produced some of the best 2019 flavors. I counted about nine plants with penellii leaf surface dots in the best flavor population. I meant to sort them but didn't. Small possibility there that crossing behavior may be directional based on something Andrew and Joseph were pointing out the other day. Penellii can in theory pollinate domestic, habrochaites, and peruvianum. However perhaps not the reverse. If the hybrids behave at least partially in the same way we should expect penellii hybrids to pollinate: penellii hybrids, perhaps back pollinate pure penellii (we think we have such a population, domestics, habrochaites and hab hybrids, and maybe peruvianum. However habrochaites hybrids probably cannot pollinate penellii hybrids. Though this thought may be taking that thinking to far. It's also possible that complex penellii x hab x domestic hybrids can pollinate in both directions depending on which genes are present and from whom. Though I do suspect that in our multiple wild species promiscous lines there is probably some odd crossing behavior that may be undetectable. If we add more species it may further complicate that within a general multi species hybrid zone population. Which makes me wonder if we need multiple populations with 150 feet of separation between. Even in the 3/4 domestic population wonder if we should separate out the penellii influenced portion and grow it by itself? Also makes me wonder if we could ultimately have multiple promiscuous populations derived from different parental stock that still behave like separate species. I.e. penellii based, habrochaites based, peruvianum based, and arcanum based? As opposed to a universal hybrid zone in which you might end up with plants with small introgressions from all species like modern domestic tomatoes but uncertain pollen flow behaviors.

I'm leaning towards wanting the various hybrids and populations to be at least somewhat compatible with each other. Basically hoping that the species barriers break down and we achieve something closer to what they were originally long ago before they differentiated into different species. Not sure if that is possible or even fully possible, but i think it would be good for gene flow. In my mind it would be good if the lines became so blurred that it was hard to tell which species was which because they all melted into the same homegeneous but genetically diverse population. But that's just me. i like population genetics and messing with those kinds of things. I'm kinda a mad scientist when it comes to things like that.

The main accessions from TGRC are F5.
Mating System: Autogamous-SC
Sporophytic Chromosome Number: 24
Comments: Resistant to PLRV, TYTV, BCTV, and big bud disease. F5 S. lycopersicum x S. peruvianum derivative.
Categories: Disease resistant; Interspecific hybrid

The ones from GRIN range from no known information other than species cross info to some interesting comments such as these:

Pedigree: L. esculentum (Michigan State Forcing) X L. peruvianum (PI 128657)
Improvement status: Breeding material

Pedigree: L. esculentum x L. peruvianum
Improvement status: Uncertain improvement status
Some degenerate plants occur among BC1 and F2 of L.esculentum X L.minutum due to two or more lethal or semi-lethal genetic factors.

Pedigree: L. esculentum x L. peruvianum
Improvement status: Uncertain improvement status
Some degenerate plants occur among BC1 and F2 of L.esculentum X L.minutum due to two or more lethal or semi-lethal genetic factors. Self, seeds not hairy.

Pedigree: L. esculentum x L. peruvianum
Improvement status: Breeding material
Lines that segregate for the mosaic resistance carried by the male parent.

So far this spring, the most vigorous population for me is the three species hybrid... Big Hill lycopersican, pennellii, habrochaites.  I'm currently at about 7 generations into promiscuous pollination. Makes it hard to provide a more accurate pedigree....

i modified joseph's great drawing to include the potential partial one way crossing ability between pennellii and peruvianum

Andrew Barney wrote:i modified joseph's great drawing to include the potential partial one way crossing ability between pennellii and peruvianum

I think the diagram should look more like this.

1. Penellii pollen goes to habrochaites not vice versa. I suspect this has implications for crosses as well.

2. The Arcanum I obtained from TGRC should set some seeds on domestic but not be a good bridge I have a second generation planted. That is ok, it looks useful in its own right.

3. The SC Peruvianum you (Andrew) obtained should set seeds on domestic. Probably won't act as a bridge, but that is useful in its own right.

4. We never had real Cornelio-muelleri and decided it was peruvianum.

5. The seed for Chilense I obtained from TGRC should be a decent bridge- if we could grow it to maturity (this year I didn't plant it, last year no plant made it to blooming). Though the lone putative Chilense I obtained and grew to maturity from Sacred Succulents did not accept pollen from Peruvianum or Arcanum.

Two weeks ago, I made a facebook post about this project, and how I use a vibrator for tomato pollination. It received 1800 shares, and 102,000 views.
https://www.facebook.com/joseph.lofthouse.63/posts/885234225303836

They are writing news articles about the post.
https://www.ladbible.com/news/weird-farmer-reveals-he-tickles-his-plants-with-vibrator-to-improve-crop-20200717

Joseph, do those tomatoes in the circle have a thick, outer part that crunches when you bite it? Is the inner part mostly just juice? I'm not good at describing flavor but do they taste like something you might find on a tree in tropical forest? Several in your picture look a lot like Captain Crunch but those look pretty much exactly like it.

My apologies for swiping your photo.

Mark Reed: I don't know about the circled fruit in particular, but the crunch and flavors you describe are very common in that sibling group. I'm glad that something worthwhile came out of it.

Do they seem self-incompatible at all? Do all the flowers set fruits? I know, wrong time of year to be asking...

I think they are self compatible cause I don't have anything to cross to them except regular tomatoes and pimps. Wouldn't they kind of just fade away if the were only pollinated by those?

I did have a red one volunteer this year that was similar in nearly all ways except not nearly as good flavor, I culled it as soon as identified to keep it from crossing. I haven't really paid good attention if the flowers are open or not but they are quite large for tomato flowers.

This question Joseph posed. I think we can generalize it a little bit. To "have we previously accidentally produced any promiscuous lines".

I've been growing habrochaites tomatoes from Joseph since 2017 alongside exserted domestics. I undoubtedly have newly hybrid embryos amongst my seed stash. No idea what percentage.

In 2019 I grew out a huge F2 growout. Mostly of plants from exserteds I open pollinated. I deliberately daubed habrochaites pollen as well as others onto stigmas of the F1. I included a small portion of known wild hybrids. Percentage wise not great, but I definitely found a few obvious ones in the direct seeded growout. These did not set seed.

There are a few lines that I was uncertain of. However, while these may or may not have had some wild ancestry they produced abundantly. Suggesting that they were selfers.

I direct seeded BH x W4 G2 of wild type parents in 2020. Direct seeding worked great. They were fruitful. I think most say upwards of 90% were obligate outcrossers and the same with the elite lines I transplanted.

So ultimately I think from my experience an accidental promiscuous line seems unlikely. Though possible.

However, I'm curious to see what happens when I isolate a habrochaites plant amongst the promiscuous G4 plants in 2021. Have a furry habrochaites line and a G4 packet with good bicolor. Yep it would be possible to have a furry bicolor. If our promiscuous lines can back cross to a full habrochaites that's convincing proof that everything works. I grew the G2? Of such an experiment with the usual habrochaites lines this year from Joseph but all were wild type so far.

William Schlegel wrote:If our promiscuous lines can back cross to a full habrochaites that's convincing proof that everything works. I grew the G2? Of such an experiment with the usual habrochaites lines this year from Joseph but all were wild type so far.

I have intentionally made that cross 5 times. Twice by manual pollination, and three times by open pollination. So far, I have only recovered wild-type (green) fruits from them, but some of the foliage has an obvious domestic influence, and some of the plants have larger diameter fruits (domestic trait), and fruits with 3 locules (wild has 2). Some of the fruits are non-hairy (another domestic trait).

I'm intending to keep doing this sort of back-crossing until yellow/orange fruits show up. An alternative tactic would be to plant out hundreds of plants and find the few that segregate for non-green fruits.