Profoundly Promiscuous and Totally Tasty Tomato Project

I understand that one of the key components of permaculture is to let natural systems do most of the work. Then the steward only has to provide guidance from time to time. Breeding tomatoes is a pain in the neck, because the domesticated genome is so narrow due to a number of "founder events", and because they have been inbreeding for so long that they have lost a tremendous amount of diversity. Heirlooms are often badmouthed because they are so inbred that yields tend towards half of what hybrids produce. So what if there were a strain of tomatoes in which every seed was a new hybrid? Then we could grow hybrid tomatoes and have them be auto-generating-hybrids, and because of 100% out-crossing rates, they could adapt quickly to new growing conditions, pests, etc. We'd be able to take the drudgery out of making hybrids, and out of tomato breeding.

I'm working on developing that variety of tomatoes.

The tomato genome has a self-incompatibility gene. What if I incorporated that into my tomatoes? That would make them mandatory out-crossers. Then every tomato that I grow would end up being a unique F1 hybrid. That would take care of making hundreds of thousands of unique genetic combinations every year instead of the 3 that I was able to make in 2014, and the zero in 2015. It will require a lot of labor and attention for a few years, but once a self-incompatible tomato is available that is a mandatory out-crosser, then that genome can be thrown at all sorts of problems: bugs, viruses, blights, frost, weeds, flavor, color, etc... And it will basically be a self-breeding machine... That excites me. I've been daydreaming about it for months, and applying the science, and wondering about the unknowns, and coming up with schemes so that the project can breed itself, and I can get out of the way. I expect to have plants flowering within a week or two to make the first crosses to start this project.

Details to follow...

Edit to add: Original title of this post
Auto-hybridizing tomatoes

I am currently growing 4 species of self-incompatible wild tomatoes:

S. habrochaites
S. peruvianum
S. corneliomullerii
S. pennellii

Perhaps more... The S. peruvianum complex was split into 4 species about 8 years ago, so no telling so far if the names of the accessions that I'm growing have been updated.

I currently have one wild tomato flowering, LA 1777, S. habrochaites. I have 3 plants from my landrace that will be flowering within a week. So my plan is to use pollen from the wild plant to pollinate the domestic tomatoes. Then, if any seeds are produced, to grow out the plants, and use them to pollinate different wild plants. If I understand the self-incompatibility mechanism correctly, then any seeds that are formed will be self-incompatible. That will produce offspring that are 75% wild, and only 25% domesticated, which sucks. But it seems easier than going the other route of bagging flowers in order to identify and cull self-fertile plants. If the plants can reject pollen that doesn't work, then I don't have to.

I expect one of the biggest problems early on is that the wild accessions may be day-length sensitive, and only prone to flowering during frosty weather. If necessary, I may make the initial crosses indoors during the winter, or under grow lights. So my first selection goals are:

self-incompatibility
day-length neutral
open flowers and exerted stigmas

Then, once the population is stable for those traits, I can use it for crossing with some of my long term landrace projects, using the same scheme as above to maintain self-incompatibility as the defining trait of the project.

We suspect that Sungold has S. habrochaites as one of it's ancestors. If that's true, then it may have some self-incompatible genes running around in it still. So I intend to use some descendants of Sungold to pollinate LA1777 or other wild accessions, just in case some of the pollen has retained the ability to get past the self-incompatibility barriers.

@Joe L: " .....my plan is to use pollen from the wild plant to pollinate the domestic tomatoes."

Based on the scheme downstream from this point as you elaborated, this would probably be okay. By taking your F1 material back across the wild species, you will be regaining the diversity of the incompatibility gene(s) in this way. But as you noted, "That will produce offspring that are 75% wild, and only 25% domesticated, which sucks. But it seems easier than going the other route of bagging flowers in order to identify and cull self-fertile plants."

I'm guessing that your more domesticated landrace collections are more uniform for growth characteristics, while the wild accessions will differ more widely in aspects like germination time after sowing, bolting and flowering traits as you noted, etc. Since any given plant of your wild species is self-incompatible (and for the sake of argument, I will assume 100% self-inc.), I'm wondering if it would be worth caging, with the proper timing, multiple members of your landrace plants (self-compatible) with one wild plant. Then all seeds produced on the wild plant will almost necessarily have to have come from pollination from your landrace. If you do this with multiple cages, then you will have multiple wild plants whose fruit will be F1, but as important, will all have different self-incompatibility alleles. If I'm thinking about this right, since self-incompatibility is a dominant trait, then the plants arising from all of these F1s will be self-incompatible, but if you mass interpollinate all of these F1s, then you will get mixing of your self-incompatibility genes/alleles, and (I think) retain 50% of your landrace genome. Probably downstream from this point in the F2s you may need to check for the percent self-compatible plants, but after a while it may turn out that the inbreds become less vigorous and the outcrossers will dominate the breeding program....don't know about this last part. There may still need to be some bagging done to detect self-compatibility. It seems like inbreeding depression eventually culls self-fertility in a lot of natural populations.

It's possible as well that Sungold may have self-inc. still floating around, but my guess would be that the desirable traits from S. habrochaites were selected for and self-inc was deliberately culled out. But it still could be there at low frequency....

Just one question.....when do you find time to sleep?

Edit: Forgot to note that it may turn out that there is some color or morphological trait that is linked to self-inc. If so, then you can follow the presence of self-inc by noting the presence of the color or other such trait.

@Me: "If I'm thinking about this right, since self-incompatibility is a dominant trait, then the plants arising from all of these F1s will be self-incompatible....."

Except that I wasn't thinking about this right. My mind tapped me on the head at 3 am and sent me into terrible insomnia over the fact that these F1s will in fact, be self-fertile at a rather high rate. Any pollen harboring the original self-compatible landrace gene (allele) will be able to fertilize any ovary on that plant. It would only be the case the the pollen bearing the self-inc. gene falling on the stigma of that same plant would yield an incompatible reaction and no seed. So, yes, I think you would still be battling a lot of selfing in this scheme, and yet still could use the multiple F1s to back-cross to your wild species or populations. Upon writing it up yesterday, I was recalling a breeding scheme of introducing dominant self-fertility into an already self-incompatible population, in which case it's easier to follow the trait of self-fertility. (Either that, or I was thinking too much of green beer to be had later in the day.... ) In your case, the self-fertility is what you want to get rid of which, without some added laboratory tools, is not so easy. So I wanted to get back to you ASAP on this in case it was causing some confusion for which I offer apologies. You may be right that you will have to take the rather arduous road of crossing landrace pollen back to the wild species population and re-select for desired traits along with self-incompatibility.

John: I like the idea... I have evaluated the farm, and nearby gardens where I could plant isolated patches for this project... It looks like I could easily put in about 9 isolated plantings: One wild species plant, surrounded by 9 [F1: Wild X Domestic]. The beauty of doing the back-crossing in this fashion is that I wouldn't have to do manual pollinations. I could let the plants take care of it themselves.

A possible problem with this scheme, is that insect pollinators are not all that keen on visiting most strains of domestic tomato flowers. And the closed flower traits are dominant, so the crosses aren't likely to shed lots of pollen, but on the other hand, it doesn't take much pollen to do the job, and the wild species have super exerted stigmas so a brush-by may be sufficient.

Another possible problem would be if flowering times don't overlap.... In that case, It would still generate F2 seeds that could be selected for traits that would be useful in their own right, such as more open flowers, day neutrality, short-season, determinate growth habit, taste, color, etc...

I attempted the first cross pollinations today for this project:

Pollen from a wild tomato, S. habrochaites, LA1777, was used to attempt to pollinate two different domestic tomatoes. I put a spoon under a wild flower and vibrated it with an adult toy, then dipped the stigma from the domestic tomatoes into the collected pollen after tearing the immature anthers from the domestic tomato flowers. Straight forward manual pollination of tomatoes. Tomorrow I intend to repeat the process on today's stigmas, and on a flower from a different plant that is ready.

I put a spoon under a wild flower and vibrated it with an adult toy

I love this.

I'm also very excited to see how your breeding project goes. Thanks for sharing this with us.

@Joseph L: "I put a spoon under a wild flower and vibrated it with an adult toy......"

Gosh, Joseph....Um....(still coughing).....I'm still recovering after reverse-snorting my morning coffee all over the keyboard! Where did you learn this technique?? LOL..... I couldn't find it in any plant breeding manual
Yet I can't help but wonder if, like farmers of old copulating in their fields as a fertility rite towards a good harvest, the device and practice might not increase your pollination rates in the same way! If this has not been published yet, I can see the title in a forthcoming issue of Crop Science: "Increased pollen recovery and fertilization of emasculated Lycopersicon via marital aid vibrational stimulatory dispersal".....

Do you recall if the two different plant parental types had different colored hypocotyls upon seedling emergence? This may serve as a tool for tracking crosses in the future....not always, but can be useful in some situations. Sounds like a fun project and I'll be curious to hear what kind of tomato types result from these and other crosses.

John: I tend towards being creative... Tomatoes need some type of vibration for the pollen to be released so it can fall on the stigma. Wind can be OK in an open field, but in greenhouses they often need to be manually pollinated. "Buzz pollination" is what I call the type of pollination done by bumblebees. A few years ago I noticed that I get earlier fruit set and higher production on plants that are visited most frequently by bumblebees. So I started buzz pollinating the earliest tomato flowers. I think it helps a lot with getting earlier fruit. People commonly use electric toothbrushes to increase fruit set in tomatoes. I don't like electric toothbrushes because they are expensive, and the shape is inconvenient to carry into the field with me. So I use adult toys. My favorite brand was "Pocket Rocket". It was very discrete. I don't like my current model as well, because it is larger, and BRIGHT PINK.

Buzz pollination of indoor tomato flower.


I didn't pay attention to the hypocotyls.

I don't like my current model as well, because it is larger, and BRIGHT PINK.

This sounds like a great upgrade!

Maybe one of the really heavy duty hitachi-style wands would let you vibrate the entire plant all at once?

Here's a photo of the wild parent I am currently using as a pollen donor: LA1777, Solanum habrochaites. I am pollinating domestic varieties: Brad, Black Prince, and a NoID from my landrace. I love the orange anther cones, and that the stigma sticks out so far. Brad is often the earliest ripening tomato in my garden.

This project has inspired me to try a similar experiment.

Three years ago a natural and unexpected cross between 'Mexico Midget' and/or both 'Brandywine (Sudduth)', and 'Italian Heirloom' occurred in my back vegetable plot. The next year (2014) hundreds of little vigorous tomato seedlings appeared and we let a good portion of them grow to maturity, and we had this small-to-medium sized red tomato that was pretty darn tasty and enjoyed by everyone we shared with.

In 2014, we planted other varieties in the back garden with our "wild" tomato variety. I expected and hoped for some crosses to enhance genetic biodiversity, but in 2015 when the wild tomatoes came up they were the same as the year before, and as far as my eyes and tongue could tell, were no different than the year before. I found this strange, because as far as I know, F1 hybrids, don't usually grow true. Perhaps some other genetics ended up getting in there after all.

After reading your posts and work, I see now that most cultivated tomatoes have poor flower structure for pollination and are natural inbreeders. I searched a rather massive database of wild tomato accessions and have selected quite a few to evaluate and incorporate into my tomato population.

I selected accessions that had the following characteristics:
Pollinators noted visiting flowers
Large Flowers, or high amounts of flowering
Disease or Pest Resistance
Cold, Heat or Drought Tolerance
Good Fruit Sets
Flowering and Fruiting Time of Year

I have made selections from:

S. cheesmaniae
S. chmielewski
S. habrochaites
S. neorickii
S. pennelli

These 5 are listed as easily hybridizing with S. lycopersicum (common cultivated tomato), and according to my research can either be inbreeders or outcrossers, depending on the form.

They have a host of other helpful characteristics I'd like to incorporate like disease resistance, pest resistance, drought, heat and cold resistance, depending on the species and collection in question.

I also included the 4 species that make up the S. peruvianum complex:

S. arcanum
S. corneliomuelleri
S. huaylasense
S. peruvianum

Also useful for disease resistance and all natural outcrossers, as far as I can tell. From the information I could find this complex does not easily hybridize with S. lycoperiscum, and that:

Crossing L. peruvianum to L. esculentum (S. lycoperiscum) is rarely sucessful. Attempts frequently result in embryo or flower abortion. As more lines have been evaluated, a few have produced at least one seed. Fortunately, these hybrids are capable of backcrossing to a L. esculentum (S. lycoperiscum) parent.

Another method which has been successful at overcoming the incompatibilty between the cultivated tomato and the L. peruvianum is the use of L. chilense as a bridge species (L. peruvianum is crossed to L. chilense and that progeny is crossed to L. esculentum). Although this might seem encouraging, most of the time this methods fails, but does yield better results than a direct cross. Most crosses between the cultivated tomato and the members of the "peruvianum-complex" fail due to some sort of incompatibility.

Apparently S. peruvianum complex and S. chilense can cross, but I forgot to acquire S. chilense seed, so that won't happen this year.

I'm not doing any embryo rescue techniques, but hopefully one of the four species I'm evaluating will produce at least a little viable seed for further backcrossing.

I have concerns with flowering time as you have mentioned, but also edibility of the wild species I am crossing. I know I will end up crossing these wild tomatoes many time, therefore diluting any possible toxic compounds, but alas I could find very little information on edibility or toxicity of these wild tomato varieties. There are a lot of toxic members of the nightshade family and I'd hate to create something that got me or someone else sick. Some of the accessions I requested were noted as being edible by local peoples though.

I selected accessions with a diverse flowering time from Summer and Winter, taking into consideration that the Southern Hemisphere has different seasons. I have grow lights if necessary.

I also will be growing 8 conventional varieties, some hybrids, some open-pollinated, with some noted for producing lots of flowers, or lots of pollen, for an addition of genetics into my local landrace. I may have to hand pollinate, but I'm lazy and wish the bees would do this for me.

Thomas Ziminski: Good work! Welcome to the world of wild tomato breeding. Thanks for the analysis of crossing compatibilities.

Nightshade poisons tend to be highly emetic! So, if you eat a wild tomato, and feel like throwing up, you might as well stop eating it, and purge what you have already ingested... I often notice the emetic feeling while the fruit juice is on my tongue. No swallowing necessary.

Out of about 300 domestic tomato plants (mostly segregating hybrids) that I grew last year. One had fruits that tasted "off", so it got culled immediately.

Embryo rescue of tomatoes can be as easy as opening the fruit at 31 days after pollination, taking any seeds out, and planting them into a sterile potting mix: Perlite watered with miracle grow seems like as good a medium as any.

Today, I potted up 66 Sungold F2 plants. I'm intending to screen them for exerted stigmas, and to use those plants as pollen donors to S. habrochaites to see if enough self-incompatible genes are left in it to produce a successful back-cross. (We suspect that S. habrochaites is an ancestor of Sungold). And if not, then I'm still interested in using the exerted stigma trait in some of my other tomato breeding projects. I wonder how compatible descendants of Sungold would be with some of the more difficult-to-cross species? Or in other-words, if Sungold or it's descendants could act as a bridge?

This is a photo of the flower that got me started down the path to self-incompatible tomatoes. Notice how the anther cones are barely connected to each other, and how the stigma is exerted... This is a self-compatible tomato, but with the wide open flower structure, it is inviting cross-pollination. This is from DX52-12, the closest thing we have to a local heirloom. When I was a child, it was the most widely planted tomato in this area.



A typical tomato has a flower that looks like the following. I call these industrialized types: Because they are all closed up. The petals are even arranged like a spike phalanx to prevent bees from approaching. Therefore cross pollination is very unlikely. It is a centralizer's dream. Using every trick in the book to prevent diversity. Even though this might be called an open pollinated tomato, there is nothing promiscuous about it.

So far, so good. Here's a photo of a fruit that I hope contains seeds for an interspecies hybrid between S.habrochaites (LA1777), along the left edge of the photo, and a currently NoID tomato from my landrace.

@Thomas Z.: " There are a lot of toxic members of the nightshade family and I'd hate to create something that got me or someone else sick. Some of the accessions I requested were noted as being edible by local peoples though. "

I admit, it's been a bit of a mystery to me as to how, along the winding road of human<->edible co-evolution, the decision was made to jettison certain foodstuffs as toxic. As noted about tomato being in the nightshade family, some toxins work acutely and manifest their effects quickly and efficiently. There are other things that may not be so toxic but are unpalatable....the latter quality often being a human evolved reaction to that potential foodstuff for a good reason. But what about the 'slow' toxics.....how would one know that they may have inadvertently introduced a more slowly acting toxin. Would it arise over time that enough people had a bad reaction to some variety/ecotype that eventually it worked its way into the lore of what to avoid? Part of asking these questions relates to a lot of food toxic and allergy issues in recent decades (including gluten, various legume proteins, etc.) as well as to the question Thomas' comment raises. And it gets complicated when both the eater and the eaten continue to evolve. And then there are medicinals, both natural and synthetic, that may come close to killing you as a part of their healing action. As for "....accessions I requested were noted as being edible by local peoples though. ", I think I've heard that some Peruvians will eat varieties of potato with alkaloid contents that most others would consider intolerably bitter. Mostly just musings here.... Really impressive projects from both you and Joseph L!

The fruits of some of the inter-species hybrid crosses are getting close to maturity. Seems like there will be time to get one more generation grown out this summer, on at least some of them.

Mother = Domestic tomatoes. Pollen donor = Solanum habrochaites, LA1777.

I have been finding some traits that I hope to incorporate into this project. I'm making a few hybrids per week, as pollen and maternal flowers become available at the same time.

Exerted Stigmas:


Bold floral displays:


Huge flowers:

The first fruit from this project is ripening nicely. There is plenty of time left in the growing season to grow out the seeds.

Will be curious to know how the flavor and other traits progress, Joseph. Any concern about some of the less palatable alkaloids showing up or are you pretty confident that this won't happen with the parents that were chosen?

John:

I haven't screened the wild parent species for flavor... A few fruits formed on S. habrochaites, so I cut a green fruit open and tasted it. It tasted like a green tomato. It didn't taste emetic which is hopeful.

My typical strategy when localizing a new species to my garden, is to first get a population that reliably reproduces. Then I start selecting for flavor. I expect that lots of variations in flavor will show up in this project.

Last year, the two tomatoes that tasted best to me were descended from crosses between wild species and domestic tomato.

I just tasted a fruit from LA1777, Solanum habrochaites, one of the wild tomato species that I am using for this project. The fruit had fallen off the plant, so I figured that it was ripe. It's color was white with a light greenish/yellowish stripes. It was not sweet. There were no harsh tastes. Not bitter. Not tart. Not emetic. Slight overtones of subtle flavor. Just a bland fruit. Not much going on to please a primates palate, nor to displease it.

Joseph Lofthouse wrote:
I just tasted a fruit from LA1777, Solanum habrochaites, one of the wild tomato species that I am using for this project. The fruit had fallen off the plant, so I figured that it was ripe. It's color was white with a light greenish/yellowish stripes. It was not sweet. There were no harsh tastes. Not bitter. Not tart. Not emetic. Slight overtones of subtle flavor. Just a bland fruit. Not much going on to please a primates palate, nor to displease it.

That's exciting. Thanks for the update.

My human brain automatically wants to know what it tastes like cooked. Would it make a difference? Would it bring out toxins, or reduce them?

@Joseph L. :"Not much going on to please a primates palate, ...."

What do you think?....Do you suspect that your own palate, even given the re-introduction to un-processed food, is still relatively inclined towards the main culture's benchmark of 'sweet' and flavorful? I'm always curious about what the natives who began selection were tasting and favoring along the way of their breeding efforts. Given that the Euro immigrants generally found them tasty and palatable as well, it sounds right to use "primate palatability". As you are crossing these wilds with a domestic, you are bringing flavor back into the lines. I wonder how many generations and with what population sizes you would need to begin to get size and flavor just out of an LA1777 selection program alone.

My palate is not at all pleased by the standard American diet. I can't eat fruits offered at a restaurant. The tomatoes, strawberries, raspberries, and melons seem like effigies of real food. They seem as bland and lifeless to me as it's possible to get on something that claims to be a food. I used to grow "cantaloupes" but over the years, I selected for fruits that are soft, and super-sweet, and incredibly fragrant and tasty. They are an entirely different product than what is sold in the stores or served at restaurants, so today I call them "muskmelons".

Somewhere around here I have a refractometer to measure sugar levels. I'd estimate that the wild tomato was around 2% sugar... A regular tomato might be around 6%. And a really sweet tomato around 12%.

The subtle flavors in the wild tomato had hints of floral about them, so they could be a nice addition to a fruit that contained more sugar.

I grew about 10 plants last year which were about F3 descendants of crosses between wild and domestic tomatoes. Two of them got awarded "best tasting tomatoes of 2015". They were cherry tomato sized, so there is still work to be done to increase fruit size. This spring I planted around 3 dozen plants descended from "best tasting". So we'll see if anything interesting shows up. Some of the offspring have exerted stigmas, so I'm pleased about that. I made crosses between "best tasting" and LA1777.

A few years ago I made a cross between plants with 1.25" diameter and 3.5" diameter fruits. 17% of the grandchildren of the cross had 3.5" diameter fruits. So in real life that was 3 plants out of 18. Those sorts of odds are very favorable for selecting for good fruit size.

I'm growing an isolated patch of Solanum habrochaites this summer. Thanks John, I'll pay attention to fruit size and odor/taste within the population. See if anything jumps out as being more pleasing to my palate.

Solanum habrochaites fruits:


So would you grow hairy tomatoes?

What if it meant that hairy tomatoes were less likely to be damaged by bugs, sunscald, etc?

Here's a photo of a flower on  Solanum pennellii, LYC 1831. It is part of my project to develop promiscuously pollinating tomatoes. It has an anther cone that can readily shed pollen because of the loose/open end. The anthers aren't joined fully together. The stigma is very exerted. The flowers are carried high above the foliage, so they make a great floral display. What a wonderful addition to this project! I intend to use it to make some crosses to domestic tomatoes.


As a progress report on this project, so far I have harvested and planted seeds from 4 crosses between domestic tomatoes (some of my favorite varieties) and Solanum habrochaites, LA 1777. Two of them have already germinated. I made other crosses. Ripe fruit from those is expected within a week. It's late in the season to be planting tomatoes, but I intend to grow them in pots so that I can bring them into the greenhouse, or indoors if necessary to get mature seeds.

I have a few patches of wild tomatoes growing, and intend to use them as pollen donors to domestic tomatoes, as well as making a wild tomato population founded as a mixture of inter-species hybrids. It will be a few years before I get the self-incompatibility trait established, but in the meantime, I am selecting for promiscuous pollination.

A few more fruits of S. habrochaites have ripened since my last post. I didn't notice them immediately after they fell off, so they sat for some time. When I finally got around to tasting them, they were sweeter, and more floral tasting than those I tasted previously. They would be worth eating as a fruit in their own right. It's hard to get excited about picking marble sized fruits, but since they fall off the plant when ripe, perhaps they could be picked by growing them in a hanging basket, and shaking the fruits onto a tarp.

Here's a photo of another plant that has traits that I hope to incorporate into my auto-hybridizing tomato project. The flower display is very bold.  It is carried high above the foliage like a beacon to any pollinators in the area. The petals are huge. Just the sort of thing that I'd love to encourage in my tomatoes. The plant is a noid accession of Solanum habrochaites. The stigma is inside the anther cone, so boo hoo about that...

You keep this up and I'm going to be ordering tomato seeds from you to grow in my flower garden.

Casie: You got that right! Here's a photo of another of the wild tomato species, Solanum peruvianum. The floral display is so bold that they stand out from across the garden. The photography doesn't do them justice...

Flowers from another species that I'm using for the auto-hybridizing tomato project: Solanum corneliomulleri. Stigma is very exerted.



And a more general picture of the whole plant.