Profoundly Promiscuous and Totally Tasty Tomato Project

It will be really interesting to see this year what the offspring of the best plants from last year are like. Offspring from plants like the second one in your post. Already tasted good last year, will be really neat to see what you get this year.

"Fairy Hollow" looks interesting to those of us who dislike "tomato snot" (the clear goopy stuff).

Rez Zircon wrote:"Fairy Hollow" looks interesting to those of us who dislike "tomato snot" (the clear goopy stuff).

Yup. And it might have utility as a paste tomato. It was very prolific, so I shared the seeds widely. No telling what interesting things might come from the Fairy Hollow family.

Joseph Lofthouse wrote:

Rez Zircon wrote:"Fairy Hollow" looks interesting to those of us who dislike "tomato snot" (the clear goopy stuff).

Yup. And it might have utility as a paste tomato. It was very prolific, so I shared the seeds widely. No telling what interesting things might come from the Fairy Hollow family.

Fairy Hollow is one I included this year in both my transplant and direct seeded patch. It's not interesting for the Auto-Hybridizing project, but its still 50% Habrochaites and its segregating still, seedlings have multiple leaf types, and it tasted good last year. So should be fun this year.

One thing that has startled me about this project, is how susceptible the wild species are to plain old domestication. There is so much diversity within them, that it's simple to select for larger fruits, better flavors, and local adaptation, even without crossing with domestic tomatoes.

Joseph Lofthouse wrote:One thing that has startled me about this project, is how susceptible the wild species are to plain old domestication. There is so much diversity within them, that it's simple to select for larger fruits, better flavors, and local adaptation, even without crossing with domestic tomatoes.

Yep.....I suspect this is true for so many crop species.  In the "too big to fail" department, it's just too costly and to precarious for the larger scale food/seed producers to dabble in the wilds to find new variants of the kinds you are observing.  For sure much of what will be selected from wild material might still end up looking, genomically speaking, much like their domesticated relatives on account of selection at loci that are fundamental to desired domesticated traits like fruit number, size, quality, etc.  But there will likely be new combinations of genes from your efforts that are NOT present in the general 'domesticated tomato', heirloom or otherwise, that nevertheless give simultaneously similar yet different qualities to the new varieties/landraces being generated.....and as noted earlier may additionally bring along novel resistances to pests, pathogens, and adverse climate conditions.  

From your reading, Joseph, is the self-incompatibility locus/loci going to be dominant and pretty straightforward or is it more complex?   Should be some interesting progeny from all of this!.....

John Weiland wrote:From your reading, Joseph, is the self-incompatibility locus/loci going to be dominant and pretty straightforward or is it more complex?   Should be some interesting progeny from all of this!.....

I haven't found much to read about this topic other than what I have written myself, so here goes....

Based on what I am observing in the field. The self-incompatibility trait seems to be dominant. From a plant breeding perspective that is a bit troublesome. It is easy to select for recessive traits, because once you do the selection, it is permanent, until the plants are crossed again. When trying to select for a dominant trait, there is always a little bit of the recessive trait hanging around, waiting to show up in later generations. The most common way around that is to self the plants then grow out enough to determine whether the mother contained the recessive trait. That method isn't available in a self-incompatible population. I'm still fussing about how to deal with the recessive selfing trait. Might end up just doing selection year after year to cull the self-compatible plants as they show up, diminishing the chances of self-compatibility year by year. We put a lot of effort in over the winter into selecting for  promiscuous flowers, and for plants that acted like they were self-incompatible. That work will be ongoing this summer.

There are a lot of different genes controlling flower shape and thus promiscuity. Some are recessive, some are dominant, some seem co-dominant. Again, it just seems like a process of constant observation to select for promiscuous flowers, and to cull flowers that would be more suitable for selfing.

I'm finding plants that combine self-incompatibility with closed up flowers. That's not a good combination for survival. If they don't self-eliminate, I'm culling them.

Tomatoes that are self-compatible with promiscuous flowers would be a wonderful improvement over the ever more inbred varieties that people are currently growing. So I may spin off some of the self-compatible tomatoes into other projects.

Hmmmm....I have to admit that, like probably many gardeners, I just save the seeds from some fruits from the previous generation without having made observations on the flower morphology to see if we are inadvertently selecting for reduced promiscuity.  As you noted, if we are accidentally selecting for those with closed flower types then we would be driving the batch towards homozygosity (highly inbred).......which is not terrible in an unchanging world, but not so good for the diversity needed to confront an ever-changing climate and pests/disease scenario.  Without using those slick-a-roo molecular markers that could ID which plants are homozygous for your self-inc loci, I can't see another way off the top of my head to increase that trait in your population except as you noted.....to 'bag' a branch of each of the plants growing in a large population and see if that bagged branch produces fruit....which would suggest a self-compatible plant, possible to be culled.  Time consuming, but over many generations should reduce self-compatibility to a low level.  The variation in your tomato patch must be like walking through the desert during peak post-monsoon bloom!  'Fairy Hollow' sounds quite interesting as well.....

Culling the self-compatible plants has been straight forward. They set an abundance of fruit starting with the first flower cluster.

The self-incompatible fruits tend to not set fruits on the first few flower clusters, until the bumblebees really start working the flowers.

The domestic tomatoes had 3 bottlenecks during domestication. Combine that with the ongoing selection for purity (inbreeding). The end result is that domestic tomatoes have shed something like 95% of their genetic intelligence about how to deal with ever changing climate, soils, insects, blights, etc.  One study I read indicated that they found more genetic diversity within one accession of wild tomatoes, than within all accessions of domestic tomatoes in the study.

Joseph Lofthouse wrote:
Culling the self-compatible plants has been straight forward. They set an abundance of fruit starting with the first flower cluster.

The self-incompatible fruits tend to not set fruits on the first few flower clusters, until the bumblebees really start working the flowers.

The domestic tomatoes had 3 bottlenecks during domestication. Combine that with the ongoing selection for purity (inbreeding). The end result is that domestic tomatoes have shed something like 95% of their genetic intelligence about how to deal with ever changing climate, soils, insects, blights, etc.  One study I read indicated that they found more genetic diversity within one accession of wild tomatoes, than within all accessions of domestic tomatoes in the study.

Wonder if there might be a grant source that would let you use marker assisted selection to clear the unwanted trait from the population. Or if that would even work. Maybe we should ask Carol what she thinks about the problem.

William Schlegel wrote:Wonder if there might be a grant source that would let you use marker assisted selection to clear the unwanted trait from the population. Or if that would even work. Maybe we should ask Carol what she thinks about the problem.

I'm having conversations with a genomics fellow on this very topic. Seems like a very possible way to select against self-compatibility. The responsible genes are known.

Here is a promising looking berry on a Nymph.

William Schlegel wrote:Here is a promising looking berry on a Nymph.

My most promising looking berry on a [Big Hill X Nymph], and the flower (huge/bright) from the same plant compared to a domestic flower (small/pale).

Also of note, is that the first few flowers to open didn't set fruits. Classic self-sterile behavior.

Five of the Big Hill x Nymph plants I grew have fruit set, but still small fruits, early in developent. Mine all have the small type flowers with barely exserted to equal to the pollen tube stigmas.

Most nymph plants have fruit set. There is one that must be determinate. It dropped all its flowers and stopped blooming. Only a few have good exsertion. The rest are about equal to the pollen tube and set correspondingly fewer fruits and drop more flowers.

Most of my auto-hybridizing style plants seem to be dropping a few flowers before getting pollinated. Including my home grown patch of hab from last year's one plant that set.

Penelope plants mostly are not setting fruit so far. Had similar luck last year with the F2 of that same cross. Except for the plant I think of as "Three Way" it's setting tons and it has the Penellii smell Andrew talks about. Lemony. Otherwise the Penellii F1 Andrew sent one seed of is producing and the Penellii ? X you sent seed for is setting fruit. They may be getting pollen from the plants not setting fruit. They are also all near the 3/4 hab patch.

Of all the auto hybridizing plants the best pollen producers are the 3/4 hab.

That row of  tomatoes makes me happy, which  have blossoms large enough to see from across the field.

I'm growing one sibling group that I think looks like a three-species cross.

I devoured this article earlier tonight:

The Plant Journal(2014)77,727–736 Restoring pistil-side self-incompatibility factors recapitulates an interspecific reproductive barrier between tomato species Alejandro Tovar, Aruna Kumar, Katsuhiko Kondo, Amy Ashford, You S. Baek, Lillian Welch,Patricia A. Bedinger and Bruce A. McClure

It basically says that there are two genes that affect self incompatibility in tomatoes. One is expressed in the pistil, and one in the pollen. And that self-incompatibility can be restored by re-introducing both genes. Whew!!! Good thing, since that's been the whole basis of my Beautifully Promiscuous and Tasty Tomato project.

Joseph Lofthouse wrote:
...
Based on what I am observing in the field. The self-incompatibility trait seems to be dominant. From a plant breeding perspective that is a bit troublesome. It is easy to select for recessive traits, because once you do the selection, it is permanent, until the plants are crossed again. When trying to select for a dominant trait, there is always a little bit of the recessive trait hanging around, waiting to show up in later generations. The most common way around that is to self the plants then grow out enough to determine whether the mother contained the recessive trait. That method isn't available in a self-incompatible population. I'm still fussing about how to deal with the recessive selfing trait. Might end up just doing selection year after year to cull the self-compatible plants as they show up, diminishing the chances of self-compatibility year by year. We put a lot of effort in over the winter into selecting for  promiscuous flowers, and for plants that acted like they were self-incompatible. That work will be ongoing this summer.
...

There would be a way to do it, but it requires a lot of hand pollination and a lot of record keeping. I know, not what you like the most.
YEAR 1
You have Y plants which you don't know if they contain the recessive, let's call them 1, 2, 3, 4...
You hand cross 1x2, 1x3, 1x4, 2x3, 2x4, 3x4...
You cross all Y plants with a self-compatible one, let's call it S: 1xS, 2xS, 3xS, 4xS...
YEAR 2
You grow 1xS, 2xS, 3xS, 4xS, ... and try to self them.
If 2xS can be selfed, 1x3, 1x4, and 3x4 are self incompatible and you can mix them, along with other self-incompatible ones.

Maybe if you can get a few people from OSSI breeding forum together it would be doable?

Hans, we tend to talk around in different theads and Joseph and I just had a similar conversation centered around the idea of pollen rejection.

Pollen is haploid, if Sbroken is rejected, than strains where only working S alleles are included are probably swiftly self selected for. Therefore planting mother to row should work fairly well as a methodology after all. Identify an entire row that is non-selfing, and there is a fair chance it's got two S alleles. This would also work with back crossing to a parental species.

In my garden this year it really does seem hard to find a plant that has reverted to selfing except for Fairy Hollow. That suggests to me that reversion is difficult somehow and might require selfing which I think the self incompatible plants may do to a small degree.

I wonder also if perfection is necessary here? If self incompatible tomatoes really are self stabilizing because of sbroken pollen rejection, we might be able to get to "pretty good" really really soon. I would say in my garden it seems pretty good now in the descendants of Josephs intense selection last year.

Though I must admit I haven't looked over that article yet.

Update: looked over the article some. Interesting figure in there. Even the sc strains are unidirectional on the green fruited species.

Also reread Hans's breeding scheme. Understood it more clearly the second time around. It's doable and may be necessary if this pollen sorting thing doesn't quite hold true for some reason. Personally I have trouble making deliberate crosses during my busy season. Seems to work better when a bit cooler out temperature wise or in my greenhouse and to dip the stigma a few days in a row. Another way to go might be to try to keep a clone going of each half wild until after the children of the sc cross are blooming and setting fruit. That eliminates all the unnecessary crosses with plants that have one allele of Sbroken and one working S allele. Hard to keep cloned going though, but I do have one plant from last year still going. It's been an incredibly worthless plant: I think of it as being the worst of both worlds. So inserted it never gets crossed and self incompatible. So far; all my attempts at deliberately crossing it have failed as well.

In principle though, if I could clone three plants from my 2019 garden. I have two particularly promising exserted Nymph plants and a nicely exserted Blue Ambrosia. If I could clone those three and make test crosses with them. It would be interesting to me and if one of the two Nymph plants proved to have two working S alleles it could go into the 2020 garden. The resulting hybrids would be interesting too. Maybe I should take a few cuttings of some of my favorites. Some of the 3/4 wilds also are really good pollen producers.

If the back cross to habrochaites is the only way to go, it might be nice to identify SI 1/4's and then plant their offspring next to the 3/4's. Some of the 3/4s should take up their pollen and produce 1/2 offspring with two copies of working S alleles. Repeat a generation or so and you have 5/8 plants. You could take this same scheme further. Produce 1/8th SI plants, back cross to 1/2 or 5/8 plants and so forth until you essentially got the SI system into domestic plants. This scheme though is dependent on pollen sorting working. Also we want lots of the diverse wild DNA, not just the breeding system.

Why do I have doubts about pollen sorting working? Well, plants are kind of rule breakers. The question to me is not if the rule will be broken, rather it's how often.

Joseph Lofthouse wrote:

William Schlegel wrote:Here is a promising looking berry on a Nymph.

My most promising looking berry on a [Big Hill X Nymph], and the flower (huge/bright) from the same plant compared to a domestic flower (small/pale).

Also of note, is that the first few flowers to open didn't set fruits. Classic self-sterile behavior.

That berry looks beef steak like and so does the flower. Potentially promising because beefsteak sized fruits would be a huge step forward for the autohybridizing project. Judging by the pure Big hill fruit I scooped seed from last night it should also be really good for our seed production. Lots of seed in just one of those!

William Schlegel wrote:That berry looks beef steak like and so does the flower. Potentially promising because beefsteak sized fruits would be a huge step forward for the autohybridizing project.  

Yes. It's looking more like beefsteak every day. Out of 250 [Big Hill X Nymph] plants, this is the only one that set fruit early, and has beefsteak-type fruits/flowers. So I asked the question if it was really a cross, or perhaps a bungled pollination. The leaves are archetypal for an F1 hybrid between domestic and habrochaites (around 20 leaflets).  The flower is larger than Big Hill, and less beefsteak-ish, looking mid-way between Big Hill and habrochaites. So it looks like an inter-species hybrid. A full sibling of this plant is also producing lots of precocious fruit, and one of them is already ripening. This sibling group has a lot of anther bruising, indicating that it's very attractive to pollinators.  

We put a lot of resources into making an  F1 beefsteak hybrid last winter. It's looking like the investment is paying off.

I'm doing a lot of work this summer to increase the number of S alleles in the population, and to make back crosses to S habrochaites, and to make 3 species hybrids. Along with the work to get beefsteak-sized fruits.

One of the plants in the population that I've been calling Backcross-1 has early/larger fruits. Yay! looks like the backcross to S habrochaites was successful. It's finally grown enough generations for larger fruits to segregate out of it.

Picked 15 berries off the half wilds this morning. Still greenish mostly except for the first I found which is peach colored but they have well formed abcission layers so I took them. Exciting.

Picked a few more part wilds yesterday.

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.

Might be nice to hand cross elites with one another. Would help to have same plants two years. Clone them. Or have a long enough season to hand cross after first ripe fruits. Also might be a good idea to plant plants far enough apart so can tell them apart (note to my future self).

William Schlegel wrote:Might be nice to hand cross elites with one another. Would help to have same plants two years. Clone them. Or have a long enough season to hand cross after first ripe fruits. Also might be a good idea to plant plants far enough apart so can tell them apart (note to my future self).

Ha! I planted 1/4 wildlings in a 7' X 9' grid, and they have totally overgrown each other!!!

Thanks for the reminder. I'm meaning to try cloning about 4 of my favorites. I can aspire to manually crossing elites if they are growing in a south-facing window in the house.

Last growing season, I grew plants in pots until after they were flowering. Then took the earliest promiscuous plants, and planted them together into an isolated crossing block.

Joseph here is something new to me I encountered today. A blue skinned hab x domestic in my garden.

William Schlegel wrote:Joseph here is something new to me I encountered today. A blue skinned hab x domestic in my garden.

I figure that the blue skin came from Black Prince, which was one of the original varieties pollinated by solanum habrochaites. It didn't do very well, or provide a very high percentage of the seed, so it's good to see that at least some of  it's genetics are still hanging around.