Species Current Tomatoes vs. Hybrids

Started by Joseph Lofthouse, 2024-01-03, 09:11:50 PM

Previous topic - Next topic

Justin Jardinier

Just a quick additional note. In their conclusion they state:
QuoteOur analysis support a two-step domestication as proposed by Blanca et al. [9]; a first domestication in South America and a second step in Mesoamerica,. The distribution of fruit weight and shape alleles also supports these two steps and shows that domestication of SLC occurred in the Andean region of Ecuador and Northern Peru. The definition and clarification of the biological status of SLC is also an important result of this work.

I wonder if this may not actually conflict with the view in TAXONOMY OF WILD TOMATOES AND THEIR RELATIVES (SOLANUM SECT. LYCOPERSICOIDES, SECT. JUGLANDIFOLIA, SECT. LYCOPERSICON; SOLANACEAE) Iris E. Peralta - 2008, if we simply view those 'pure SLC' as pre-Columbian South American SLL. Going back to the beginning of the paper, they state:
QuoteThe location of tomato domestication also remains uncertain. Both Mesoamerica [14] or Ecuador and Northern Peru, near the center of origin of SP [17], have been proposed as the center of domestication. If the former were true, SLC would have had to migrate north to Mesoamerica as a wild or weedy species, where it would have been domesticated into SLL. Instead, a two-step domestication process has been proposed for tomato [9]. The first step would have consisted of a selection from SP or primitive SLC by early farmers resulting in the Ecuadorian and Northern Peruvian SLC. The second step likely occurred in Mesoamerica, and consisted of further selection from these pre-domesticated SLC after their migration from Ecuador and Peru.

Is it perhaps the case, that SLC was claimed as a separate species for that reason - that people thought that it was a wild species that had not been domesticated? Thus they wanted to differentiate it from SLL, which they considered domesticated. And, is it possible that this view is Eurocentric? Tomatoes have been bred continuously for a very long time. Might it be that they are looking at European tomatoes and seeing 2 steps for their history, and thereby somehow denying that those domesticated pre-Columbian South American tomatoes farmed for perhaps countless generations, were 'not really domesticated' because they don't look enough like European tomatoes? Personally, when I see them, they look domesticated to me! And I still find myself struggling to see the continuation of breeding of a small sample of those, in Mesoamerica, which is the common view of historical trajectory which indeed they confirm, as a break to a new species. And even in their conclusion, they state:
QuoteThe distribution of fruit weight and shape alleles also supports these two steps and shows that domestication of SLC occurred in the Andean region of Ecuador and Northern Peru.

So if the creation of the conceptual category 'SLC' was to differentiate it from the 'domesticated tomato species', is that not made redundant if we see SLC as domesticated, and SLL as merely SLC taken to Mesoamerica and further selected over generations?

Justin Jardinier

#16
I had another look and I see this in the paper:
QuoteOne of the other notable exceptions to the correspondence between genetic and passport classification was the subgroup comprised of accessions that were listed as SLL vintage, but instead were genetically classified as a SLC group closely related to SLC Ecuador. This cluster was classified as SLC vintage and consisted of genetically diverse germplasm that included accessions collected mostly at South American markets.

It sounds like those are our pre-Columbian landraces. So it sounds to me as if.. please correct me if I am wrong... they are noticing the genetic difference between those pre-Columbian South American landraces, and the Mesoamerican and Mesoamerican-derived tomatoes. Then classifying them as SLC. And then colouring in their chart to make those blue:
chart.png

But I still do not see anything in the data that would stop us from considering those clearly domesticated tomatoes as SLL, and understanding that if they were domesticated from SP, they are naturally closer to SP as they appear on that chart. And, that some or many of those blue triangles (plotted nearer to SP) may still be hybrids between those pre-Columbian South American SLL, and SP which as we can see on the maps, do appear to occur close together geographically.

And the difference between the colour scheme for marking them as blue triangles (SLC) or grey crosses ('mixture') seems to be on the basis of passport data:
QuoteThe accessions in this region include all ten artificial SLL x wild species hybrids and the accessions BGV007985, BGV012625 and LA1909 are already classified as interspecific hybrids in their passport data, thus we have called this group "mixture".

And since the blue and grey are intermingled, again I do not see how this shows a categorical difference based on the data, rather than a preconceived notion of separation between SLP and SLL-SP hybrids projected onto the data. Furthermore, there is a very strong cluster of 'SLC' at the top right, which flows uniformly into the red SLL cluster, which would seem to me to be consistent with them being the same species and the flow from that blue cluster into the red cluster being a gradual continuation of breeding and environmental adaptation of the species; and then there is a more dispersed group of blue triangles somewhat separate in between that strong long cluster of blue+red and the green SP cluster, which would seem to me to be consistent with them being hybrid populations of the pre-Columbian SLL from the main blue cluster and the SP populations. In particular, most of them seem to cluster around the centre in between the main blue cluster and the SP accessions which are closest to it, which I would think should be expected if the main blue cluster were considered to be SLL domesticated from those SP accessions (which may be in the region of domestication), and those blue in the middle being SLC resulting from the hybridisation of those SLL and their pimp ancestor populations which may still be present in the same region.

To support that hypothesis, it looks rather like there are 3 clusters of blue within that 'in between' cluster:
chart - 1.png

I hypothesise that that could potentially be explained by group 2 being regular hybrids, group 1 being group 2 hybrids that crossed again to SP, and group 3 being group 2 that crossed back to SLL from the main blue cluster.

Like I said there may be things I'm missing but these are my thoughts. I also don't discount the idea of some of the blue dots representing the various degrees of domestication represented by what we could call SLL, in the multicultural landscape of pre-colonial South America, possibly including crossing with SLL landraces traded back from Mesoamerica. Or perhaps even, since they have reclassified "accessions that were listed as SLL vintage" as SLC, perhaps even the strongest cluster of 'SLC' at the top right continuous with 'SLL' might be post-Columbian, perhaps being SLL received from Mesoamerica or perhaps from European tomatoes, crossed with their older pre-Columbian SLL landraces, and cluster 3 and maybe even also 2 may represent their pre-Columbian SLL landraces, while cluster 1 represents hybrids of those with SP; and those 'SLC' below cluster 3 might represent post-Columbian SLL crossed to some extent with SP or their pre-Columbian SLL landraces.

Had I the data and capability, I think it would be very interesting to makr the known South American landraces in a separate colour, and remove all of the European-derived accessions, then run a new principal component analysis. That could be interesting.

William Schlegel

[quote author=Justin Jardinier
Now, if we take their 'SLC vintage' to mean pre-Columbian South American SLL, then their evolution tree makes sense to me, showing Mesoamerican and European SLL are derived from South American SLL. And what I think could be interesting, would be to cross some of the pure pre-Columbian South American SLL still being farmed by native farmers, from as diverse regions as possible, with various SP (and if there were time enough, allow them to self-stabilise to some degree under feral conditions), then plot them on such charts of genetic comparison with both parent populations, along with claimed SLC and modern cherry tomatoes derived from SLL-SP crosses.

There is much I may have missed but I have to work now.

[/quote]

Sounds like it took some heavy lifting but you've realized now correctly that they've redefined most South American heirloom tomatoes regardless of fruit size as SLC based on them forming a discrete and genetically distinct monophylletic group.

Not an a priori assumption. One of their major findings is that SLL didn't exist in South America as they have redefined SLL and SLC based on the genetic data.

This sort of thing can be hard to take as morphological groups make a lot of sense, but then at the genetic level they turn out to be useless. This is the case here. SLC goes from being all cherry tomatoes everywhere as a morphological group to being the larger fruited tomatoes of South America as a distinct genetic group. Also as a distinct genetic group they are immensely valuable and previously undervalued. 
Western Montana garden, glacial lake Missoula sediment lacustrian parent material and shallow 7" silty clay loam mollisoil topsoil sometimes with added sand in places. Zone 6A with 100 to 130 frost free days

William Schlegel

#18
I think that blue triangle group in that figure is more or less the proof that SLC as they have redefined it is a monophylletic group. I think it is a quality figure. The thing that stood out to me is the distance between SP and SLC. Strikes me a little bit from the text that this means SL started evolving on its own from SP as it habitat differentiated. The galapagense accession right in line with SP suggests strongly that SLC is way more differentiated from SP than is galapagense. That means SLC which they've redefined with genetic data to include all South American landraces is a very important gene pool! Which I think proves Justin's earlier idea that we should be using them for breeding. 

https://bmcgenomics.biomedcentral.com/articles/10.1186/s12864-015-1444-1/figures/7

Though Figure 7 is quite interesting! When it comes to number of alleles at a Loci SLC is super important, but when it comes to private alleles its all SP all the time. Suggests to me that both taxa have importance for breeding as is likely true of all tomato taxa. Though no one has been ignoring SP but I think SLC has been undervalued, likely because as morphologically defined it was largely invisible.

"The SP category was the most genetically diverse group (He = 0.21), followed by SLC (He = 0.17) and SLL (He = 0.12) (Table 2)." and "However, when combining the contemporary SLL subgroups (processing and fresh), slightly higher levels of diversity were found when compared to SLL vintage (He = 0.12 vs. 0.09), a situation that is likely due to the effect of introgression during breeding and differentiation into distinct market classes..." These two quotes from the article are also very interesting. SP has the most diversity by far, SLC has the second most, and SLL has significantly increased, but is still depressed compared to SLC and really depressed compared to SP.

This seems to be part of a wider story. Traditional landraces get ignored in favor of wild relatives whenever wild relatives are available.

I suspect for non-commercial tomato breeders that SLC is a potential gold mine allowing significant genetic diversification but with less of a downside than breeding with even say SP which leads to a lot of size reduction. However, it is also clear from these numbers "SP(He = 0.21), followed by SLC (He = 0.17) and SLL (He = 0.12)"  that SP and SLC both currently have more diversity than does SLL- so in theory we could spend our whole tomato breeding careers working with just the relatively safe SP and SLC and not even touch such things as Solanum habrochaites and make good progress towards the diversification of domestic tomatoes.

I wonder what tomato varieties are available from the true South American SLC group?
Western Montana garden, glacial lake Missoula sediment lacustrian parent material and shallow 7" silty clay loam mollisoil topsoil sometimes with added sand in places. Zone 6A with 100 to 130 frost free days

William Schlegel

#19
https://bmcgenomics.biomedcentral.com/articles/10.1186/s12864-015-1444-1/figures/1

Another thought on this figure 1 from the paper we've been discussing. The group they call "Mixture" is where a decent number of tomatoes in my garden likely already belong. I think it is also the group I aspire for my tomatoes to belong to and not just in regards to the three main taxa here. I don't want my tomatoes to cluster with modern SLL (domestic) or with SLC (South American domestic) or with SP (wild current). I would want ideally, an admixture of all three- plus some genetics from additional wild species. So accessions that are already a mixture- would be an ideal source of genetics for breeding in my estimation. The most genetically robust possibility.

"One striking difference between the two classifications occurred for 102 SLC accessions that were located in the PCA between SLC and SLL and classified as mixture (Figure 1). These accessions included many of the commercial cherry tomatoes. These data imply that most cultivated cherry tomatoes are not true SLC." From (Blanca et al 2015)

While not true SLC the fact that commercial cherry tomatoes went into the "mixture" part of this graph actually elevates them a great deal in my estimation as germplasm. A great historical example is Farthest North available from Victory Seeds. Famed breeder Albert Yeager simply crossed Bison with a Solanum pimpinellifolium creating a mixture and breeders in my region have been working with descended crosses ever since. Later in the Blanca et al 2015 article they point out that the modern breeding introgression process started before 1920 so we have had 100 years of introgression. That is certainly true in my region with well documented pedigrees for the 45 year old Sandpoint and Kootenai varieties containing ancestry from pimp, peruvianum, and habrochaites.
Western Montana garden, glacial lake Missoula sediment lacustrian parent material and shallow 7" silty clay loam mollisoil topsoil sometimes with added sand in places. Zone 6A with 100 to 130 frost free days

William Schlegel

https://bmcgenomics.biomedcentral.com/articles/10.1186/s12864-015-1444-1/figures/6

Figure six in the paper is a whole lot of fun! Apparently Peruvian pimps are more different at the genetic level from Ecuadorian pimps then are Ecuadorian pimps from galapagense and the SLC are also quite distance. 
Western Montana garden, glacial lake Missoula sediment lacustrian parent material and shallow 7" silty clay loam mollisoil topsoil sometimes with added sand in places. Zone 6A with 100 to 130 frost free days

Justin Jardinier

#21
I wanted to do some data analysis on their huge genotype file but unfortunately I'm not having much luck, I think the file has too much data for my applications, my best one can only handle around half a million cells it seems - wow that sounds a lot, not sure if I made a mistake there but anywa, I have given up, for now at least.

Quote from: William Schlegel on 2024-01-05, 11:36:58 AMSounds like it took some heavy lifting but you've realized now correctly that they've redefined most South American heirloom tomatoes regardless of fruit size as SLC based on them forming a discrete and genetically distinct monophylletic group.

Not an a priori assumption. One of their major findings is that SLL didn't exist in South America as they have redefined SLL and SLC based on the genetic data.

Hmm, to me that is not necessarily entirely logical. What I mean by that is, well, let's take a look at that sentence, which actually sounds quite accurate:
"SLL didn't exist in South America as they have redefined SLL"

Do you see what I mean? We could say it did exist, but... well let's think of an example. The bananas we eat now are different from those in traditional ones Papua New Guinea. We could redefine the traditional bananas in Papua New Guinea and say they are not bananas, and then we could say banananas never existed in Papua New Guinea until colonisers brought our modern bananas there. Personally, to me that seems a strange approach. I personally don't want to redefine the word 'banana' like that.

And it concerns me that redefining pre-Columbian SLL as SLC seems... hmm, maybe even rather Eurocentric or coloniser-centric? And potentially a projection of a narrative that the South Americans never domesticated tomatoes. Almost as if to say well our tomatoes, came from Mesoamerica, and when we look at South American tomatoes, they look inferior, and so I won't even acknowledge the centuries of domestication and cultivation undertaken by those 'natives' - they are too different from ours and I will not give them respect.

Now of course this is not exactly what the paper is saying. But to repeat the quote from it I gave earlier:
Quote from: Justin Jardinier on 2024-01-05, 08:38:41 AMEntertaining one of the domestication hypotheses, they state:
QuoteThe first step would have consisted of a selection from SP or primitive SLC by early farmers resulting in the Ecuadorian and Northern Peruvian SLC. The second step likely occurred in Mesoamerica, and consisted of further selection from these pre-domesticated SLC after their migration from Ecuador and Peru. This second step completed the domestication process of tomato. Genetic data confirmed that European SLL accessions originated from Mesoamerica and constitute the genetic base of the SLL vintage varieties [9].

And to repeat another quote I gave earlier from TAXONOMY OF WILD TOMATOES AND THEIR RELATIVES (SOLANUM SECT. LYCOPERSICOIDES, SECT. JUGLANDIFOLIA, SECT. LYCOPERSICON; SOLANACEAE) Iris E. Peralta - 2008:
QuoteWe do not recognize taxa below the species level, most notably the small-fruited tomatoes known to many as "var. cerasiforme." The name "cerasiforme" has been used to refer to putatively wild forms of S. lycopersicum that have been regarded as progenitors of the cultivated tomato (although see Frary et al., 2000, and Nesbitt & Tanksley, 2002).

So, unless I am misunderstanding, this name's function, SLC, was to define what was considered to be a wild subspecies. And that fits with the 'pre-domestication' narrative given in the paper. And, to quote Peralta et al. from their more recent work, written after this paper (thanks for that tip William!) The Tomato (Solanum lycopersicum L., Solanaceae) and Its
Botanical Relatives - 2016-12-09:

QuoteBlanca et al. (2013) distinguish these pre-breeding Andean populations at the varietal level as var. cerasiforme. This has been traditional in the tomato literature for plants of S. lycopersicum with small fruits, but we consider these plants to be the product of domestication, not of evolution by natural selection, and thus should not be named using the International Code of Nomenclature for algae, fungi, and plants (McNeill et al. 2102). In addition, Blanca et al. (2013) found that the South American accessions they identified as distinct had a wide range of fruit sizes; the accessions were better distinguished using a panel of morphological characteristics (similar to those used to distinguish S. pimpinellifolium and S. lycopersicum by Peralta et al. 2008), thus use of 'cerasiforme' could cause confusion. We suggest this distinct set of accessions be named according to the International Code of Nomenclature for Cultivated Plants (Brickell et al. 2009), as has been done for potatoes (Huáman and Spooner 2002).

This makes sense to me.


Quote from: William Schlegel on 2024-01-05, 11:58:31 AMThese two quotes from the article are also very interesting. SP has the most diversity by far, SLC has the second most, and SLL has significantly increased, but is still depressed compared to SLC and really depressed compared to SP.

I believe that is exactly what should have been expected and so far as I understand, was already known. Of course it's always nice to have more data on it. But I could not imagine it any other way. And if we keep Peralta's naming system, it perfectly fits their narrative - wild tomatoes were domesticated into SLL in South America, there was a lot of diversity, then some went to Mesoamerica resulting in a drastically less genetically diverse population, which underwent further breeding there (and presumably the South Americans also continued their own breeding too, as they had been for so long already), then the second bottle neck to Europe. And finally some modern breeding using crossing with some wild species.

Quote from: William Schlegel on 2024-01-06, 01:59:40 PM"One striking difference between the two classifications occurred for 102 SLC accessions that were located in the PCA between SLC and SLL and classified as mixture (Figure 1). These accessions included many of the commercial cherry tomatoes. These data imply that most cultivated cherry tomatoes are not true SLC." From (Blanca et al 2015)

Again I think this is entirely expected - although I think those 'mixtures' were in between SP and SLL, not SLC and SLL, no? Anyway, suppose we follow Peralta's narrative, we should expect something like these groups as being genetically distinguishable from each other (or potentially anyway, though some might not be possible to distinguish with any clear borders, which itself may be a rather significant point). Instead of saying 'pre-Columbian South American SLL' I am going to say 'SASLL':
1) SP
2) Feral SP-SASLL crosses
3) SASLL
4) Mesoamerican SLL
5) Feral SP-Mesoamerican SLL crosses
6) European SLL (include all modern tomatoes in this and for now let's forget about the complexity of Asian etc.)
7) Modern SLL-SP crosses
8 ) Modern SLL that include some wild introgressions.

The modern SLL-SP crosses should be distinguishable because their SLL half would be European SLL, and also for that reason far less genetically diverse, from their SLL half anyway (with some exceptions if they have also been including other species from modern breeding, so I will add an 8th category above). We might also expect some difference on the SP side, since modern breeders might be choosing specific SP from obscure ranges for their traits, whereas we might expect the feral SP-SASLL crosses to be crossed with SP either within the range of SASLL, or along travel routes, though there might also be some level of transmission via birds. So I might expect something very roughly like this perhaps:
tomato hypothesis.png

It would also be quite hard to distinguish category 2 clearly, Feral SP-SASLL crosses, since I would expect such crossing to have been going on all along the history of domestication, so there might be crosses with very early SASLL as well as later SASLL, and for example an early SASLL crossed once to SP might be hard to distinguish from a later SASLL crossed once to SP. And if a very early SASLL managed to survive feral in a pure form for some reason, then that also might be hard to distinguish from a later SP-SASLL cross. Indeed some of the feral SP-Mesoamerican SLL crosses might also be hard to distinguish from feral SP-SASLL crosses, and the line between those and some feral SP-European SLL crosses might be expected to be blurred also - I did not make a category for that but the remaining grey crosses might be expected to include some of all of these feral crosses, including those. Furthermore feral vs non-feral should be hard to distinguish between if ever even possible at all, perhaps only with the exceptions of the use of specific modern SLL parents and/or specific unusual (in terms of range) SP, and in cases of backcrossing such as late blight breeding etc.

Of course in order to group them properly, I would need to look into the data, but I present that rough example as an hypothesis. And it looks to me as if the paper in question has potentially chosen to take these groups:
2) Feral SP-SASLL crosses
3) SASLL

And perhaps part of this group:
5) Feral SP-Mesoamerican SLL crosses

And named them all 'SLC', and it does seem to make the claim that that is a wild species:
QuoteSLC grows either as a true wild species, in home gardens, along roads, sympatrically with tomato landraces, or as a cultivated crop [9].

And only acknowledging that it might have been semi-domesticated in South America:
QuoteSLL arose in Mesoamerica as there is no evidence of the existence of ancestral SLL in South America. All SLL accessions sampled from South America were found to carry introgressions from wild relatives suggesting that they were derived from breeding efforts taking place in the last 100 years. Therefore, to complete the domestication of SLL, SLC would have had to migrate to Mesoamerica possibly as a semi-domesticated type.

But remember, in order to make this claim they redefine the very term SLL! I quoted this before but let's take another look:
QuoteOne of the other notable exceptions to the correspondence between genetic and passport classification was the subgroup comprised of accessions that were listed as SLL vintage, but instead were genetically classified as a SLC group closely related to SLC Ecuador. This cluster was classified as SLC vintage and consisted of genetically diverse germplasm that included accessions collected mostly at South American markets.

Now their actual Principal Component Analyses was done "without considering the a priori classification based on passport data (i.e., species, location and use)." But their colouring (and circle, triangle, cross) appears to have been done by them, on 1) the basis of passport data, and 2) their own categorisation, which included changing the passport data. So of course there is "no evidence of the existence of ancestral SLL in South America" - they defined it away! And yet, so far as I can see, the evidence is that South Americans were farming tomatoes which they had domesticated there, and still continue to grow them today, or at least those which survived colonisation. And by the paper's own admittance, they are still farming them and selling them at markets etc.

So this is why I am left with my continued suspicion that they have projected the old narrative of the primitive nature of South American landraces not even being domesticated tomatoes, onto the data. But from the data itself, I actually see potentially no disagreement with the understanding already presented by Peralta. In fact I see the view from this paper being potentially more disorganised, since from their colouring of the data, it would seem that they potentially even deny the existence of Feral SP-SASLL crosses, or at least what I would expect to be the majority of them within the natural range of SASLL within the region of their origin, which should be found in the data as lying in between SASLL and those SP most closely related to them, hence the position of my 'group 2' circle on the chart. I mean, even if we deliberately took an SASLL from group 3 on the chart, and crossed it with an SP from the top of group 1, we would presumably all entirely expect the cross to plot within group 2, right? But if they went by data with no passport info, then they would surely colour it blue and call it 'SLC', right? Which directly reaffirms Peralta's point:
QuoteWe do not recognize taxa below the species level, most notably the small-fruited tomatoes known to many as "var. cerasiforme." The name "cerasiforme" has been used to refer to putatively wild forms of S. lycopersicum that have been regarded as progenitors of the cultivated tomato (although see Frary et al., 2000, and Nesbitt & Tanksley, 2002). It is impossible to distinguish wild from cultivated forms using herbarium specimens, and we regard many specimens labeled as "var. cerasiforme" to be possible revertants from cultivation (i.e., feral plants) or possible hybrids of wild and weedy taxa.

So it seems whilst Peralta has done away with the SLC category entirely (and the reasoning seems to me sound), this paper seems to have taken the opposite approach and instead expanded it, thereby subsuming what I think must be at least some (I would guess most) feral (and perhaps even non-feral) SP-SASLL crosses, possibly some feral (and perhaps even non-feral) SP-Mesoamerican SLL crosses, and, denied that the South Americans ever even domesticated tomatoes. As it stands, I just don't see sufficient reasoning to agree with this position. I would even go further - to me it actually seems wrong.

Again I wish I could work directly with the data, I think it could be quite a treasure trove for going more into depth with this topic. Alas. Well, anyway those are my thoughts.

William Schlegel

https://www.nature.com/articles/s41438-020-0291-7

"In order to help to exploit the variability detected in our collection and facilitate its use to the research community, F1 and F2 generations were constructed for the 163 accessions by crossing each accession with one accession representative of each species (SP, SLC, and SLL). The accessions, BGV007109 of SP, LA2278 of SLC and Money Maker of SLL were selected as parents." - above article quote

The above article is about the development of a remarkable set of germplasm resources based on the findings in that older Blanca et al 2015 paper. The accessions created by the above crosses are available to researchers from TGRC and a gene bank in Spain.
Western Montana garden, glacial lake Missoula sediment lacustrian parent material and shallow 7" silty clay loam mollisoil topsoil sometimes with added sand in places. Zone 6A with 100 to 130 frost free days

Justin Jardinier

#23
William I think you would really enjoy this article, Variation Revealed by SNP Genotyping and Morphology Provides Insight into the Origin of the Tomato  -Jose Blanca et al.- 2012:
https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0048198

It has some great graphics. This one might be of special interest for breeding in terms of being able to see the climate for different accessions:
climate.png

Then also, Sucúa region sounds pretty interesting:
QuoteDespite the abundance of admixture in S. l. cerasiforme, not all accessions belonging to this variety appear to be the result of an extensive hybridization process. For instance, the accessions from the Sucúa region in Ecuador and some from San Martín in northern Peru do not seem to be admixtures according to the Structure analysis, although in several a small contribution of the northern Ecuadorian S. pimpinellifolium genome could be detected. If we ignore the clear southern Peruvian admixtures, the Sucúa accessions are also the closest S. l. cerasiforme accessions to S. pimpinellifolium in the PCA. Both their lack of admixture and their closeness to the wild species might indicate that the S. l. cerasiforme accessions of the Sucúa region might be similar to the S. l. cerasiforme that inhabited the region before foreign tomatoes arrived in the Andean region. It is remarkable that the S. l. cerasiforme from this region also has high polymorphism and high heterozygosity that are equivalent to those found for S. pimpinellifolium in Piura, its region of maximum variability and allogamy [23,24].

Here's one from pretty nearby:
https://tgrc-mvc.plantsciences.ucdavis.edu/Accession/Detail/LA4699

"Very long trusses, multiloculed and flattened fruits, exserted stigma, 3cm fruit diameter."  17L4021: h/h, increased branching."


LA4696 and LA4702 seem similar and are nearby, both also exserted, the latter described as "Multiloculed, flattened and ribbed fruits, hairless stems, exserted stigmas,  2.5-4cm fruit diameter." - perhaps the most interesting yet!:


And also nearby is LA4700, which they have no picture for at TGRC but relates to that original question of colour:
Quote"Long trusses, peanut shaped fruits, exserted stigmas, 2cm fruit diameter, fruits drop when they are orange color."  17L4022: erect, dense bush, increased branching, h/h.

Back to this paper, here's another interesting part that could have consequences for what we might like to breed with if we are in wet conditions:

QuoteThe data obtained in the present study agree with S. pimpinellifolium as the origin of S. l. cerasiforme. In the PCA, the northern Ecuadorian accessions of S. l. cerasiforme, including the non-admixtured ones, were close to the wild northern Ecuadorian S. pimpinellifolium. Despite the PCA result, due to the lower diversity and heterozygosity of the northern Ecuadorian S. pimpinellifolium, we would not hypothesize that the coastal Ecuadorian S. pimpinellifolium was the only origin of the highly diverse eastern S. l. cerasiforme. The highly diverse Ecuadorian S. l. cerasiforme may have originated with the participation of the Montane S. pimpinellifolium from Machala, Ecuador, with the possible participation of other regions. Rick and Holle [17] also proposed that the eastern Andean S. l. cerasiforme might have originated from the coastal S. pimpinellifolium. This process could have been favored by the high rainfall typical of both slopes of the Ecuadorian Andes. The wild plants, adapted to high humidity environments on the western Andean slopes, could have easily migrated to the wet eastern slopes once the landscape was modified by man. This high rainfall characteristic of eastern Ecuador could also have prevented the invasion of foreign cultivated tomatoes and could have favored the isolation of some ancestral Ecuadorian S. l. cerasiforme populations. Rick [42] and Nuez (personal observation) reported that it is not possible to cultivate modern tomato varieties in these regions due to their susceptibility to fungal attacks. In fact, the tomatoes cultivated by locals are local S. l. cerasiforme. It is worthy of note that Ecuadorian S. l. cerasiforme is a valuable breeding resource that has been neglected despite its high variability and the unusual habitat that it occupies.

Justin Jardinier

#24
Quote from: William Schlegel on 2024-01-07, 06:40:23 AMThe above article is about the development of a remarkable set of germplasm resources based on the findings in that older Blanca et al 2015 paper. The accessions created by the above crosses are available to researchers from TGRC and a gene bank in Spain.

Ha ha ha, our google searches must be very similar! I was just reading that too. And many of the authors are the same for the article I just shared. Also I was looking into the one I shared's source of Sucúa region 'LLC' and it seems that was their main source - or they might have even collected them for that study, they're students at /members of that university which has that gene bank.

I'd be really curious to know if many of those Ecuadorian 'SLC' have been tested in late blight trials, or maybe also in the ToBRFV trials. In the paper Screening of Solanum (sections Lycopersicon and Juglandifolia) germplasm for reactions to the tomato brown rugose fruit virus (ToBRFV) - Ahmad Jewehan - 2021, they only tested SLC from the TGRC (though I don't know what classification they are using, maybe some of their SLC and SLL include some of the pre-Columbian South American domesticated ones). Anyway, in that paper the 4 'SLC" that showed 0~20% resistance were from:
LA1385 - Southern Peru
LA1456 - Mexico
LA2675 - Southern Peru
LA2688 - Southern Peru

I don't know if any of the others they tested were from Ecuador, too many for me to check. Of their resistant pimps, this was from Ecuador:
- LA1547 (Northern, not coastal, and also "Resistance to leaf mold (Cladosporium fulvum)")

The others all seem to be from Peru aside from 2 which the TGRC gives no geographic data for. If I had access, that Ecuadorian pimp might be the most appealing of the pimps then. But more to my point, this may also suggest that SLC in that region may also have resistance. And that might even be found in that Spanish genebank. Unfortunately I have never been able to find any database of their tomatoes - the most I could find was a list of their 'tree tomato' accessions!


Justin Jardinier

Back to the debate - here's a paper which appears to support my case, Genomic inference of complex domestication histories in three Solanaceae species - Stéphanie Arnoux et al. - 2020:
https://onlinelibrary.wiley.com/doi/full/10.1111/jeb.13723

QuoteThe cultivated tomato, Solanum lycopersicum L., was domesticated from the wild progenitor S. pimpinellifolium in Peru before experiencing two bottlenecks: first moving to Mesoamerica (Blanca et al., 2012) and then with few cultivars introduced to Europe from Mexico (Atherton & Harris, 1986; Blanca et al., 2015). These domestication events and further genetic improvement led to specific genomic footprints in tomato (Lin et al., 2014).

In section 4.3 | Species-specific timings of the domestication stages:
QuoteThe tomato domestication centre is located in Peru (Blanca et al., 2012), even though it was imported into Mexico, further increasing the bottleneck effect due to cultivation. We estimated this bottleneck to date 4,410–8,480 ya, which accurately fits the first area cultivation records for maize in Peru ~ 6,600 ya (Grobman et al., 2012) and in Mexico ~ 7,300 ya (Pohl et al., 2007). Surprisingly, our estimate of split time (9,900–19,040 ya) is largely in disagreement with the ∂a∂i estimate of ~ 78,000 ya in Razifard et al. (2020), which resemble more to a speciation event, as it predates the human presence in South America. Further sampling in Peru would be needed to evaluate if the geographic origin of the first domestication event occurred in Peru or in Mexico. Additionally, the best fitting model corroborates the two-stage domestication scenario for tomato, already proposed in Blanca et al. (2015). In the African rice, this type of genomic inference approach has been successfully applied to shade light on the geographical origin of the crop, thus unravelling its domestication centre (Cubry et al., 2018).

From reading other papers I wonder if it was really Peru, or perhaps Peru and/or Ecuador. But this date seems interesting.

Diane Whitehead

#26
Wow! Fascinating!  My head is spinning. 

I'd like to get back to the simpler beginning of this topic, though.

Every year I list all vegetable seeds sold in Canada, for Seeds of Diversity:  https://seeds.ca/diversity/seed-catalogue-index/

and we keep two tomato lists: Tomato and Tomato, Currant.  (not 'current') 

This is our current list of currants which seems a bit of an odd mixture. I'd like to know the history of some of these:

I've looked up some and added information given by the various seed companies.  The ones labelled "currant" are described as being S. pimpinellifolium.

Candyland  F1   AAS winner   1/2 inch fruits
Cerise Orange
Chiapas  resistant to both Late and Early Blight as well as Septoria Leaf Spot.
Coyote    resistant to both Late and Early Blight as well as Septoria Leaf Spot.
Cuatomate  Unique Zapotec Indian heirloom. Orange currant tomato with lots of flavour and juice.
                    5-6 feet sprawling plants with  extremely pointed leaves. Very strong disease and blight resistance.
Gold Currant
Gold Rush
Hawaiian
Humboldtii
Matt's Wild Cherry
Petit Moineau
Promise
Red Currant
Sweet Pea Currant
Ted's Pink Currant
White Currant
Wild Galapagos  S cheesmaninae   resistant to blight, fusarium, and blossom end rot. Under ideal growing conditions, these                              plants can yield hundreds of fruits.
Yellow Currant

Tatiana lists even more:  https://tatianastomatobase.com/wiki/Category:Currant_Tomatoes

Victoria, British Columbia, Canada
cool mediterranean climate  warm dry summers, mild wet winters,  70 cm rain,   sandy soil

Justin Jardinier

Quote from: Diane Whitehead on 2024-01-08, 04:14:20 PMWild Galapagos
There might be more than one product being sold by that name but the ones I've seen are hybrids of Solanum lycopersicum and Solanum cheesmaniae. They tend to be sold falsely as Solanum cheesmaniae.

Quote from: Diane Whitehead on 2024-01-08, 04:14:20 PMHumboldtii
I believe that's from an outdated classification system from the 1940's (or maybe earlier?) from Luckwill, who "recognized four subspecies within L. esculentum: subsp. galeni, subsp. humboldtii, subsp. intermedium, and subsp. typicum, and described five varieties within the last subspecies: var. commune, var. grandiflorum, var. pyriforme, var. typicum, and var. validum."

L. esculantum is the outdated name for Solanum lycopersicum. Here'sanother quote:
QuoteBrezhnev's (1958, 1964) classification of S. lycopersicum [as Lycopersicon esculentum] is replete with subspecific taxa. Brezhnev attempted to describe all details of the cultivars known to him. He classified escaped plants with smallish fruits as "subsp. subspontaneum" (a nomen nudum, see Appendix 1), reflecting his belief that these plants represented feral or less developed forms of the cultivated tomato. He named six varieties of this subspecies, varieties "cerasiforme," "pyriforme," "pruniforme," "elongatum," humboldtii, and "succenturiatum," most of which are nomina nuda. These "varieties" were then seen to "rejoin" genetically to give rise to two geographical groups (Mexican and Peruvian). The large-fruited, highly morphologically aberrant forms were classified as "subsp. cultum" (again see Appendix 1), within which he recognized several units, each as "grex concultorum" and defined geographically ("australioeuropeum," "occidentalieuropeum," "australirossicum," "mediirossicum," borealirossicum" and "borealiamericanum"). Each "grex concultorum" contained 3–7 cultivars with Russian common names (such as 'Comet' and 'Korol Gumbert'). As with "subsp. spontaneum," each "grex concultorum" was depicted in his diagram as derived from an amalgamation of all three varieties of "subsp. cultum." His classification was an attempt to bring order to the huge number of tomato cultivars then grown in the Soviet Union, linking them to their supposed places of origin. Although Brezhnev's work has been largely neglected by tomato workers in the U.S.A. and western Europe, it is essential for anyone interested in examining the identities of cultivars and heritage varieties from that time. Khrapalova (2001) gave a complicated series of varietal and subvarietal names to many of these variants, but none of her names was validly published (see Appendix 1).

So it sounds like 'Humboldtii' could be one of various tomatoes put into that old category. I can't help with the other aside from having seen some of the 'currant' ones listed as Solanum pimpinellifolium.

William Schlegel

I have a thought about sub-optimal but still somewhat useful germplasm. I think it plays right into Diane's post because some of that material might be really useful it is just hard to know! Like we know Coyote was given to Craig LeHoullier and was from Mexico and we still think those galapagos tomatoes likely have some genetics from Ecuador.

Tomato accessions from Peru and Ecuador are hard to come by except in official gene banks.

I recently ordered a few months ago the four traditional tomato collections currently available from Native Seeds Search in Arizona which is a gene bank seed company org but focused on the southwest US and Northern Mexico. I have my eye on the one tomafo available from Hondurus on EFN. I also have my eye on more of the red/yellow wild tomatoes from HRseeds. Also intrigued by those pleated Oaxacan tomatoes. I'd like to find a paper on how much diversity is in the material in central America and Mexico. It is much more readily available than the South American material. I would also like to find some data again on how much genetic diversity is in the green fruited unpalatable species on the same scale as the paper from 2015 or at least all on the same measure. They had domestic at 0.12 improved from 0.09, the south American traditional tomatoes at 0.17, and the current tomatoes at 0.21. We could reasonably assume that since those South American tomatoes have survived since tomatoes were first domesticated that they are resilient enough. So that (He = 0.17) could be a benchmark for what it means for our tomatoes to be truly sustainable.
Western Montana garden, glacial lake Missoula sediment lacustrian parent material and shallow 7" silty clay loam mollisoil topsoil sometimes with added sand in places. Zone 6A with 100 to 130 frost free days

Justin Jardinier

Quote from: William Schlegel on 2024-01-08, 06:04:59 PMTomato accessions from Peru and Ecuador are hard to come by except in official gene banks.

Without wanting to bore anyone with my previous point about high percentage Galapagos genetics - this might be another reason for those half cheesmaniae making it into gardens whilst real cheesmaniae has not. Well, the size increase and the general fruit quality increase is likely a factor. But being day neutral might be another important factor. Were there relatively pure Galapagos species sufficiently adapted to customers' gardens. they might catch on better, and that might lead to more diverse amateur breeding projects perhaps. Although having said that, I wonder if many people have even bothered breeding with the cheesmaniae hybrids? Although, perhaps they have and have just been disappointed, since they'd anyway only get 25% cheesmaniae in the cross and the 25% lyc they get might be relatively unsuitable - those were my exact reasons for not wanting to breed with those hybrids.

Quote from: William Schlegel on 2024-01-08, 06:04:59 PMI would also like to find some data again on how much genetic diversity is in the green fruited unpalatable species on the same scale as the paper from 2015 or at least all on the same measure. They had domestic at 0.12 improved from 0.09

I hope it's ok to ask - what does that mean? Is it like, the percentage of genes that are not uniform through the population, or something like that?

And, if it's something like that, then I guess.... if you have a single tomato in your garden that has 25% pennellii or even 15%, plus one heirloom, then your garden might blow he entire world of heirloom tomatoes plus all the pre-Columbian lycs out of the water in terms of diversity? Since pennellii is so distant from lyc and pimp whilst those two are so close?