I was thinking of posting here (over the few next months) some photos and lessons learned from my breeding project from season 2021. I hope this inspires others when planning the season 2022. This is a farm based project as I am mainly breeding these plants for my own use.

So, I am working on several breeding lines, the families, all starting with a cross to the same parent plant, the Mother of all. The Mother is a compact determinate small plant producing pink cherry size tomatoes, non-cracking, with exceptionally good taste for such a small plant. I basically try to develop similar compact tomato varieties with different colors and flavors than the Mother. If someone gets inspired and wants to start a similar project with their favorite tomato cultivar as the Mother, I think the concept is really good as you can learn more about the genetics of the Mother (and the other parents) when you see how they behave in different combinations. I have called my project also as the "Triple Crown project", inspired by three horse racing competitions and winning them all, as I need to fix quite many plant architecture related genes in addition to fruit color and flavor. For example, finding determinates from a det x indet cross is relatively easy but finding a truly compact determinate has proven to be quite challenging. I started this project four years ago and I hopefully have many more years to play with it.

I am growing tomatoes (and other vegs) in raised beds, covered by mypex in poly tunnels. I plant the tomatoes in high density and let them sprawl without support which explains why I am interested in compact plants. I have some no-noes genetic wise, e.g. dwarfs with rugose leaves (the "d" gene), jointless and multiflora, just to mention few. And yes, these types of plans could potentially be grown in containers and small gardens too but I really think there is potential for commercial growers as well.

Below some pictures demonstrating my growing methods, below poly-tunnel is some 50 meters long, 7-8 meters wide. If I planed tomatoes here, I would plant about 500 tomato plants in the three raised beds in the middle. In the sides, I normally plant something else, like beans.
The tomatoes gets planted very close to each other, some 30-40 cm apart.
The plants needs to be quite compact and comfortable growing close to each other for this growing method to work.
There are twelve families that I will be posting here, one posting per family. It may take some time before I get to the end of the list. Stay tuned.

- Vnuchenka family
- Gold dust family
- Lime green salad family
- Flamingo family
- Luciebell family
- Evil olive family
- Aftershock family
- Maglia rosa family
- Justyna family
- Katiebell family
- Green sausage family
- Cream sausage family

And here is the first actual post, on the Vnuchenka family. This is the most successful family what comes to the plant architecture, so it deserves to come first. The rest have been more challenging to tame, some behaving really badly, growing all over the place as they wish.

Some of you may know, Vnuchenka is a russian (commercial?) varienty suitable for growing in hanging baskets. The fruit is a small cherry, intense red, expressing sweetness and deep good flavor, although some people complain that the skin is little tough for their liking. The plant is about 50 cm high, requiring support or extra space to hang out. It is a nice cultivar already but does not really suite to my purposes to be grown freely spreading as it requires little too much space than what I am ready to provide.

Below two pictures demonstrates the plant heights I observed in the F2 generation during 2020. The higher plant is closer to Vnuchenka original heights, the shorter is close to what I am looking for in my project. So my plan here was really to find a plant that is some 40% more compact than Vnuchenka and would contribute its' rich redness and flavor to the Mother. Vnuchenka is a healthy variety too, so some additional resistance to diseases was also in the wish list. I grew 60 plants in F2 generation and (only) one plant really stood out of the bunch as more compact as others (which really surprised me a lot). The descendants of that one plant were also the most impressive in 2021 (F3) grow-outs.
I am not explicitly breeding for earliness, however, Vnuchenka family seems to be early too. In the below picture combo from 2021, you can see 6 week old seedlings (12 of them in tiny pots) at the time of planting (sown 18th of April, planted May 29th), and one of them in three different stages: June 30th, July 18th and July 21st. If I calculated correctly, that would make the last pic 94 days from sowing the seed. Not sure if that counts as ultra early but it is still pretty early.
Some F3 plants were really pretty, having nice posture and long harvest. Some other, not that much worth bragging in tomatojunction. As you probably realized, I don't remove the suckers, just let the plants grow as they wish after planting. The side shoots gives harvest later in the season after the main shoot stops producing. In these plants, the main shoot stops growing very early, typically after 5 true leaves, sometimes even after just 3 leaves. 1-3 flower clusters (each having two inflorescence) will follow. The tomatoes tastes good, they are not high brix or anything like that but still having sweetness and full of good flavor, especially when actually ripe. As this was a cross between red and pink tomatoes, I have now both of these colors to continue to F4 generation.

@Pippin good luck in your grow outs to find the perfect plant. This looks like a fun project.

I am really impressed with the cluster size and fruit set on these small plants. Excellent traits.

With regards the height of the plants, I suspect you'll have a long selection process, weeding out taller ones each generation. But perhaps not?
This was the case with all my determinate x indeterminate crosses. Huge variation in growth habit, and quite common to see them throw taller versions of themselves down the line. But I was aiming for a medium sized determinate, and didn't breed with any of the really small ones. They seem to have some dwarfing genes involved (other than regular dwarf-rugose), to keep them so compact.
Anyway that's an interesting strategy for your farm plan, to avoid any kind of supports and the labor involved.

Looking forward to see all the family as you find time to post them!

You are absolutely correct @Bower what comes to weeding out the taller plants. I didn't know this back in season 2020 when screening F2s at the first time, and I was in trouble with the tomato jungle I created. Not all indeterminate plants seem to have this problem though, some seem to be more tricky than others.

The genetics that makes the plants this compact seem to be very complex, I suspect that there are at least three factors that contributes to this phenotype, if not more. Also, some of the genes are probably factors that makes compact determinate plants larger and higher, rather than factors that makes larger determinate plants smaller. The problem is getting rid of some (partially dominant) genes, not getting some recessive genes expressed - if that makes any sense.

@Pippin yes, and besides various genes that have epistasis effects on others, there are certainly some QTL's involved. The length of internodes almost certainly involves multiple QTLs, but somehow in the shuffle if you had a large parent with long internodes, it makes it very hard to get the "sum" of all those gene parts down to the smallest again.
You might try back-crossing to the ideal sized parent, as a shortcut to improve selection for shorter plants.

During the season 2021, I tried a new method of identifying the not-so-compact plants already in F2 generation and weeding them out very early. Of course, this is just a theory and my lucky guess – and may not work in practice – time will tell if it helps having more compact plants in future generations.

I typically sowed 120 or thereabout seeds, planted them as tightly as I could outside, and then inspected the flowering pattern and cut out all unwanted types before they set fruits. On determinate plants, I only save plants that has a very specific flowering pattern - that I call "three crowns". This is the typical flowering pattern that I observe in my shorter plants, and I hope it can be used to identify a stable homozygote plant. I observed very early in the project that when crossing a taller determinate and a sorter determinate, the F1 was typically intermediate between the two. Then I tried to figure out how to identify the short type. This method does not necessarily help with the length of the internodes as sometimes the internodes starts growing longer only after the first flower appears. Below is a picture of the mass selection and weeding out the unwanted plants (which is actually difficult and heard breaking to do).
I draw a picture below what I call “three crowns” (@Bower you call this whole thing a cluster). When the main stem starts blooming, it tends to follow similar pattern when the plant stay compact. The growth is ended by a leaf (1) that has two adjacent inflorescence attached (1a and 1b). The inflorescence may be a normal fish bone and sometimes forked - forked inflorescences makes the cluster very heavy. There is only one leave (2, 4) between the “crowns” (1, 3, 5), if there is more than one leave or inflorescence growing directly from the stem, then the plant tends to become higher (in my opinion). The crown structure repeats ideally three times but sometimes there is only one or two crowns.

It seems like a good selection method, especially because your growing conditions are very uniform for all the plants.
In my small greenhouse I often crowd the plants, and I see a lot of variation in determinate cluster/leaf patterns that is partly attributable to the individual plant's adaptation to the exact position it happened to be in, how crowded or shaded, etc.
OTOH, I believe the most reduced pattern, that is the one you're selecting with a single leaf between clusters, is indicative of the recessive condition when it comes to cluster patterns. So you may discover that you've fixed that trait by this one year's selection.
Staying tuned to see what happens!

Here is the Gold Dust family.

Many of you must know Gold Dust, often marketed as early orange tomato with good flavor. Compact determinate plants that produces good sized (60-100 g), non-cracking fruits, thick meaty flesh. Tomato enthusiasts often report good production but short in flavor. Very mild, almost bland. Probably better for cooking than fresh consumption. It makes very good orange salsa with orange peppers, onions, garlic and chili. It was my first "tangerine" tomato and sure is firm and prolific. It ripens quite slowly, not ready for consumption when the color turns orange but needs further time to get more intense orange inside.

Gold Dust is one of the rare cultivars whose pedigree has been published. The flower is of the classical “tangerine” color and the tangerine gene is even recorded in literature. It has really large and beautiful flowers, very easy to try as the first female tomato cross. That is probably why I used it in way too many crosses in the first year, many of which I have not continued further. Gold Dust behaves little strangely in the crosses, for example F1 is often very large even when crossing with a smaller tomato, like a cherry. The larger size seems to be difficult to stabilize though: I had larger F2 in one of my many crosses but the F3 suddenly was smaller.

Gold Dust is rather compact determinate already but still somehow more spreading than my target. As I like my tomatoes small, close to cherry size, so the original fruit size was way too large as well. From the cross with my compact pink cherry, Gold Dust segregated as expected: red in F1, 1/4 tangerine orange colors in F2, mostly with yellow skin but one or two with clear skin. My F2 grow-out in 2020 was about 100 plants in total, offspring mainly more compact than Gold Dust. Red F2s were very tasty and better tasting than Gold Dust. The flavor of the orange F2s, well, was still little short but maybe an improvement from Gold Dust. Below two pictures from F2 2020 (only the orange tomatoes are from the Gold Dust family).
I was hoping to find a more compact plants with smaller tomatoes in the F3 grow-out but I was not lucky this time. The tomatoes are still more on the saladette category rather than cherry. The plants are also a bit too spreading, not totally out of control, but I would prefer having them more compact. They are heavy producers, maybe more on the latter part of the season. Below is the only F3 that I saved seeds from the season 2021. This plant produced ridiculous amount of tomatoes, not evident from the pictures but it did.
In the below pic, all yellowish-orange tomatoes at the bottom are from the Gold Dust F3 cross family, some with yellow and others with clear skin. The clear skin tangerine tomatoes are really pretty, especially if the skin is matte, not this shiny as in the pic. Not my best tomato family, but I still like them as they are firm, non-cracking fruits that taste good when ripe.
My plan with this family is really to get back to the drawing board as the original cross does not seem to be giving the wanted results. Next phase could be back-grossing, for example. I did cross one F3 in the spring 2021 with a small plant from the Vnuchenka family, and I even managed to grow the F1 generation in a pot as well, so I will be screening a new “Gold Dust Vnuchenka” F2 family line next season.

Also, I have another crossing line on Gold Dust with Lizziebell, a very sweet indeterminate. Both Gold Dust and Lizziebell are orange, so my original plan was to simply enhance the flavor of Gold Dust and make the fruit smaller. Unfortunately, Lizziebell had one of those dominant orange genes, most likely Beta, so the F1 was very strong orange-red tomato. I had very small F2 population, and got only one determinate sp line that was pure red - which in F3 further segregated back to orange (assumingly tangerine now), and was really sweet. The plant architecture is way too sprawling for my project, however, I try to use this line in some of my future crosses to make it more compact. The below pic is the sweet F3 from this Gold Dust x Lizziebell cross. It is an interesting tomato inside, very thick flesh on walls and the middle.

Great tale of breeding workarounds and beautiful tomatoes. Love the clear skin tangerine, very cool! I had a clear skin Beta this year and didn't know why it was such an odd peachy color until I remembered there was pink lineage. I guess there are a half dozen color variations possible from tangerine, if you count gf and the red or yellow background with different skin colors.

You wrote : "Gold Dust behaves little strangely in the crosses, for example F1 is often very large even when crossing with a smaller tomato, like a cherry. The larger size seems to be difficult to stabilize though: I had larger F2 in one of my many crosses but the F3 suddenly was smaller. "
At first I thought you were talking about plant size but now I believe you mean fruit size?
The mysterious trend towards smaller tomatoes, is one I am familiar with for sure.
Frogsleap did some interesting research on this and graphed the stats in his blog here:
http://frogsleapfarm.blogspot.com/2015/07/
Why exactly size distribution is skewed towards smaller fruit, IDK.

I agree @Bower that tangerine has a lot of potential. Maybe there are even more color variations than six if counting gs, U and some of these high pigment genes. And ry bicolor.

And yes, I was referring to the fruit size, not plant size, when writing about the unexpected results with Gold Dust crosses. I have seen this square root formula before, maybe not from Frogsleap (thanks for the link! ) but maybe from some other source. This is spot on what I actually meant: Gold Dust didn't behave in this way in my crosses as the F1 was much larger than expected. For example, the F1 from crosses like Gold Dust x Bundaberg Rumball or Gold Dust x Lime Green Salad were very large, the fruit size was closer to the larger parent if not even larger. That shouldn't happen.

Lime Green Salad family

LGS is another well-known determinate cultivar, a dwarf, multi-flora green-when-ripe tomato. The taste is probably a typical for GWR, on the tangy and sharp side of the spectrum. Some report good disease resistance. I have not had the best experience with the original LGW as the fruit tends to crack and split for me and the pollination is imperfect but it anyway seemed as a fast track to get a GWR variety for my collection.

LGS has quite many recessive genes that I don’t want to incorporate into my breeding project, at least rugose leaf dwarf (d), multi-flora and jointless. My compact pink cherry has already some other dwarf gene that I think is related to early flowering, so I want to mainly focus on that one dwarf gene. But LGS clearly has some interesting genetics that gives novel looking plants, e.g. the double dwarfs were on the Red Robin plant size categhory, while dwarf multi-floras can be really funny looking plants. But I am not saving these types of plants in my breeding project, I have just sometimes grown them out of curiosity to see what they look like.
I had around 100 plants in F2 generation, and surprisingly, I had only one good GWR to continue with, would have hoped to have more but what can you do. This plant still needs some further work, e.g. crossing to smaller plant with smaller non-cracking fruits. Also the taste and flavor needs more work. Below some pictures from different GWR plants in F3 generation, some of them having very orange look when ripe but they should be simple YY+rr+gfgf, no orange genes involved in this cross. IDK what results the orange look, is it the skin, flesh or their combination. One wild guess is that there might be some traces of red lycopene forming in the skin (in addition to yellow flavonoids) even tough there should not be any lycopene in rr flesh.
My favorite F3 plant from this family is the below brown/black tomato. Perfectly sized tomatoes, good flavor and small enough plant. Not all flowers produce fruit though (at least in this very hot season) which is pity. Not sure if there is a genetic factor involved as some of my other families seem to be more capable of producing fruit in hot weather.

Any chance there is tangerine in the LGS background? That could explain orange, but it looks too dark for orange in a yellow flesh background.. ah but you have the red background as well from your cross. So it could be possible?

I suppose anything is possible when it comes to tomato genetics.

But if there was tangerine gene involved, I would have expected to see pure tangerine plants out of those 100 F2 - and I didn't. Only yellows, some black/brown/purple and green. Mostly plain red and some pink. Not tangerine flowers.

Below one more picture on the cross-section of an orangey-greenish tomatoes. These fruits are actually the same than I ate in my mozzarella salad with those green peppers in earlier pic. These tomatoes are really really ripe - which I hope explains why there is so little green on them.
In below picture the orange seems to be very clearly in the skin - not in the flesh. Glad I found this.

Aha, so this is a function of the Y allele for yellow skin, which continues to accumulate pigment to the point of being orange.
TBH there is a huge range of yellow skin colors, maybe poorly defined in the genetics. Everything from a light clear yellow to so intensely pigmented it is actually becomes a shade of orange, if you let them continue to ripen as long as possible. OTOH if they actually get to red, you know it is lycopene still being made to a small extent and building up over time.
The same thing can happen with black/brown/purple tomatoes that they continue ripening until they look like a pink or a brick-red. If the skin has a load of gf chlorophyll available, and no switch to turn off the process, it continues to convert all of that to the underlying color until it no longer looks like a gf/gf.
I have no personal experience with gwr, as I only grew once and decided it was just too difficult to decide when it was ripe. But in principle, why not the same issue, that the amount of gf chlorophyll available to convert to color is enough to pile up yellow until it looks orange? That is if you don't pick and eat them before they have time to do it.

I like your explanation @Bower, I didn’t know there could be different variants of yellow skin. Not all GWR with yellow skin seem to behave this way, but now that I think about it, LGS did have really strongly coloured skin.

Gwr with clear skin are really difficult to tell when they are ripe, I also tend to skip them.

The Flamingo family

Flamingo is an indeterminate F1, and has sweet, firm and crispy pink cherry tomatoes. After growing the original hybrid, I started saving seeds and did not observe that much changes: the Fn plants were all pink cherries as well, very good taste.

I made a cross between Flaming F3 or F4 and my compact pink cherry. Making a cross between two pink cherries may not seem excited, however, I have actually enjoyed the project. Obviously, there is not much pressure to find new color combinations pink x pink equals always pink. But not all pinks are the same as they may still have other factors making them look different, such as green shoulders or various uniform types. There is also this opaque vs translucent factor of the clear skin that is not getting much attention in tomato breeding discussions.

The start of this cross was promising: the F1 (obviously pink) tasted really sweet. I had some 60-70 plants in the F2 generation, and saved the seeds from maybe eight different determinate plants. I did observe some strange disease symptoms in F2 generation which was little strange as I would have expected some more resistance inherited from the original commercial hybrid. Below two pictures from 2020 season grow-out, one of the F2s. As you can see, the plant is quite compact, the main shoot terminating the growth immediate when the first inflorescence cluster appears.
I planted 60 more plants in F3 generation from 6 different F2 plants. Naturally, I tried to select the most compact plants for this grow-out. Unfortunately, none of the F3 plants were very satisfactory what comes to the size of the plants. The growth of the main stem did end very early, however, all of the side shoots had very long inter nodes, making the plants look like those spiders with very thin and long legs. I suppose, these could be good for hanging bassets but obviously not what I am looking for. I find the long inter node phenotype difficult to recognize early in the season as the stretching and elongation seems to start after the main shoot stops growing.

I had one plant growing in a hanging basket in 2021 season, here the plant starts flowering and the inter nodes of the side shoots are obviously getting longer.
And later it seems that also the main shoot inter nodes had become elongated.
The Flaming family is not really behaving very well, as they grow here and there without any control. But it is one of the better tasting families I have, so I will definitely continue growing it. But I don’t have high hopes of finding compact plants from this cross anymore. Some further crossing is needed. Also, this is not the healthies family I have.

Below you can see some F3 fruits with some green shoulders, and more translucent skin (which reminds me of a gooseberry as there are these green veins or “chlorophyll vessels” showing through the skin). These tomatoes are sweet already when not fully ripe and they do not soften easily but are crunchy and crispy very long time. In the second picture, the tomatoes have clear skin but are opaque.
Luckily, I made a backcross already in 2020. I think the cross was successful, but of course, I cannot be 100% sure as both parents were already pink and determinate. I would have expected to see many compact F1s, however, the cross didn't deliver what I was hoping for. I sowed all F1 seeds (maybe 24) but only one seemed little more compact. Getting rid of the long inter nodes is not easy as @Bower has hinted many time in this forum already.

I also have another cross with the Flamingo family in F2 seed phase for next season. It is between the Flaming family and the Green Sausage family. The F1 seemed a bit more compact, so let's keep fingers crossed.

@Pippin it sounds like the 'compact plant' genetics involved in your chosen pheno are recessive, and perhaps already lost in the F3, so they are not visible in the F1, but they should still turn up in the F2 at least.
Maybe at least two genes involved? Or more. Complex stuff, the growth habit.
I know that the number of internodes before the first flower cluster is under separate genetic control, for example.
When you add up all the different genes it is quite a few, so it's lucky for you that you have lots of space to grow out and find the combination you wanted.
In a smaller space, and smaller numbers, I usually hope to get one recessive in the F2 and look for others in the F3 if not already showing.
I really like your 'hanging basket' tomato! With a nice fruit load on board.
It's always fun to find things you didn't intend, which are just cool to see anyway.

I saved some interesting off-type tomatoes last season that I figured might also be good to grow with caging or hanging baskets, a larger plum, and a larger pear. I plan to grow them both next season, with a different growing method though now accepting the “true nature” of these plants , not trying to force them to anything they are not. Hopefully they will be able to proof the value of themselves in some other way than what I was originally looking for.

This internode elongation is for sure a complex phenomenon. I have little bit given up of guessing how many and how recessive genes there might be involved. To be really successful, I would need to figure out a method to identify this phenotype preferably at the seeding stage. Growing large populations to adulthood and selecting the smallest plants is of course one way to go, however, it seems little too expensive resource and space wise even for me. Back-crossing also has it’s limitations as in each such cross also the desirable genes would be lost.

I have been thinking of experimenting with “simulated shade” as a selection method. I could try shading the seedlings early on, or expose them to far-red light artificially. The hope here is that those longer internode phenotype seedlings would react differently to the shade then my “compact genes”. I already know that the compact plants tends to get leafy in shade, however, the internodes remains still shorter. If these longer internodes were e.g. because of the so-called “shade avoidance response”, I might be able to identify the phenotype early on e.g. by hypocotyl elongation, shorter cotyledon or something similar.

There is a lot of research being done on shade avoidance, on plants in general but also especially on tomatoes. If interested more, a good place to start with the basic terminology is wikipedia https://en.wikipedia.org/wiki/Shade_avoidance and then continue with Google.

Anyway, the theory is that many domesticated food crop plants have been selected to tolerate shade and the close proximity of other plants, and they tend to stay small even when crowded. Wild plants more often try to escape from shade and the shadow of other plants. The shadow of other plants can be simulated by altering the ratio between red and far-red light. When the R:FR ratio is low (i.e. there is a lot of far red light present in relation to red light), then the plants “think” they are shadowed by other plants. And some react to this type of shade in a very special way. I actually already purchased a far-red light bulb, and experimented it a little in early spring 2021 with few seedlings but I was not able to try it later in the season because of some timing and space problems.

My rationality here is that when using the wild tomato species to enhance the flavor, sugar content and disease resistance in commercial hybrids, the plant breeder may have also brought the strong shade avoidance response back into these F1 hybrids. And that could be the reason why my Flamingo family is behaving so "wildly". This behavior could be related to something else, just guessing here. But I already know that not all of my determinate x indeterminate cross families are behaving in similar way than the Flamingo family does.

The Luciebell family

Luciebell is an indeterminate F1 hybrid, red mini-plum tomatoes, very good sweet taste. This is also one of the F1 that I started saving seeds first, and then continued crossing with some of the later Fn generations. Luciebell segregated determinate growth habbit first, and I saw some variation in the fruit form. I was little bit loosing some of the sweet taste even before I started crossing with it, and I did observed some thinner skin and cracking. The Luciebell Fn parent used in this family cross was some 100-120 cm high plant, flowering like crazy and very heavy producer, so it still seemed like a good parent for the cross.

In the F2 generation, I did observe similar inter-node elongation in Luciebell family as in the Flamingo family. But what made this cross easier was the fact that both parents were already determinate and I was able to screen more determinate plants in F2, around 60. Surprisingly, I only observed three more compact plants out of these 60, all of them having quite erect and vertical growing habit: thin stems growing upwards to a bushy and little messy plant structure.

The F3 grow-out was 12 plants from each of those three most promising F2s (i.e. 36 plants altogether). All F3s were very good tasting, having good amount sweetness as well. Some were little prone to splitting, unfortunately. Below some beautiful fruits which unfortunately were not selected because of the spreading growth habit.
I selected only three plants to go forwards to the F4 phase, all of them in the below picture early in the season (1, 2, 3). They were more compact plants, hopefully not because of the environment (as they were growing next to each other) but because of the genetics. They are from the same F2 plant. It has been little surprising to me how difficult it is to find compact individuals from these crosses.
(1) Below is my favorite F3 plant with very nice, sweet, pear-ish tomatoes, relatively compact plants.
(2) This next plant was very prolific, however, the erect growth habit is making it lean to one side and eventually fall down to one side. I have seen completely prostrate growing habits along the way, and I am not really targeting there either, however, this tendency of growing completely vertically upwards seems to be somewhat undesirable trait too.
(3) The third selection was also in the more compact plant size but the tomato itself was little larger if compared to other members of the Luciebell family. Also sweet, and so far good crack resistance.