Exploring “Crazy” Watermelon Genetics

The other day there was an interesting discussion about watermelon genetics that started on the Alan Bishop Homegrown Goodness plant breeding forum from a fellow who lives in Australia. Turns out Watermelon genetics are sort-of complicated, but interesting.

watermelon_flesh2
Approx. Watermelon Flesh Color Spectrum, from most dominant to most recessive.

The discussion started by asking about which traits in watermelon were dominant, mostly referring to flesh color but also open to other traits as well. The original poster mentioned that he started his own mass cross of over 30 watermelon varieties together (a grex) in preparation to developing his own landrace adapted watermelon to his Australian climate. He said this past season he planted only the seed for any F1 hybrids from any yellow fleshed watermelons he had but got about 90% red fleshed watermelons and concluded that obviously red-fleshed watermelons were dominant. The interesting thing is they are BOTH dominant AND recessive at the same time! Yes, watermelon genetics is a little complicated to say the least, lol.

Wait… what??!… haha yes, you did read that last sentence correctly. Red-fleshed watermelons are both dominant to yellow-fleshed watermelons AND recessive to yellow-fleshed watermelons. Turns out there are actually TWO different kinds of yellow-fleshed watermelons.

Watermelon Flesh colors range from various forms of red, pink, yellow, orange, and white. So how does one figure out what is recessive and/or dominant over what? Turns out most of these have already been studied and we can interpret that data. I’ve recently resurrected my old website domain and turned it into a plant breeding wiki of sorts. Feel free to check it out @ www.biolumo.com. The main resource i am using is the wonderful watermelon genetics info posted online by the Cucurbit Genetics Cooperative hosted by North Carolina State University and in particular Todd C. Wehner part of the Department of Horticultural Science at North Carolina State University. http://cuke.hort.ncsu.edu/cgc/cgcgenes/wmgenes/gene12wmelon.html

watermelon_flesh2_expanded

From the data available we can come up with a rough basic pictorial based diagram. I like pictures; they help me understand things better. Basically there are at least two types of red-fleshed known as “Scarlet Red” and “Coral Red” in addition to two forms of yellow-fleshed known as “Canary Yellow” and “Salmon Yellow”. Turns out Canary Yellow is dominant to all other forms of color. Scarlet Red is dominant to Coral Red, Orange, and Salmon yellow. Coral Red is dominant to Orange and Salmon Yellow. Orange is dominant to Salmon Yellow. You get the point. And basically seems to work in a cascading effect of “more color” to “less color”.

I personally prefer the taste of the Canary Yellows over most red/pink, though there are still some good red ones out there! What i don’t like are the Salmon Yellows (and maybe orange). To me and in my climate the Salmon-Yellow watermelons have a weird mealy and/or mushy texture and have a muted / poor flavor. By contrast the Canary Yellows seem to be really sweet and might even get sweeter more easily / earlier in a northern and colder climate like mine. That is just my personal preference, your taste buds and soil conditions may differ.

Now this is a general simplified version. There are a few caveats however. Such as the fact that there is a Canary Yellow inhibitor gene that when present will turn a Canary Yellow back into a red that is hiding underneath. Also the fact that there may be a few other minor colors that have not been studied yet such as “dark red“, “rose” , and “pink“. It is possible that these colors are just minor variations of the former reds and function the same way from “more color” to “less color” in terms of dominance. It is also possible that if these are indeed separate shades of color that they may buck this trend and function in completely different ways from different biochemical pathways. Hard to say at this point. But i will leave the possibility open either way in case new studies in the future address these watermelon flesh colors.

Oh, and what about white-flesh?! Yes that’s right, we have completely forgotten to talk about white fleshed watermelons. Oh, you didn’t know there were white-fleshed watermelons? Yeah there are. They are not generally as common but there are white fleshed watermelons out there. Turns out white-fleshed is a little more weird. Let me explain.

White_fleshed_watermelon_f2
F2 Generation of White-Fleshed Watermelon Genetics

 

White-fleshed watermelons are currently being studied more in depth in China and a new paper is due any time in the near future. But until then all we have is the data gathered already from a past study on it. According to that study: white-flesh were found to be dominant over all color. In an F2 (Second Filial) Generation the ratio is: (12 white : 3 canary-yellow : 1 red).

Pretty interesting huh? Yeah, basically if i interpret this information correctly is that for whatever reason white flesh overrides color. In the wild, watermelons were originally thought to be white fleshed and low in sweetness. This is certainly the case in the wild citron melon (Citrullus lanatus var. citroides) which has hard white flesh and bland flavor. The bitter apple melon too (Citrullus colocynthis) but it obviously is very bitter.

The genetics for watermelon at this point captured my interest so i decided to find out what i could about seed coat colors. If you thought watermelon flesh genetics was complicated, you’ll find the genetics for watermelon seeds is a nightmare. Nevertheless i waded knee deep into the confusing data and came up with some generic info that i think can give us a basic trend that we can use.

WatermelonSeedGenetics
Approx. Watermelon Seed Coat Color Genetics

The genetics for watermelon seed colors and patterns is a nightmare. Truly it is. Partly because the studies we have don’t all agree and we don’t have examples of what these old researchers were really studying. One person’s “tan” might be another persons “light brown”, etc. You get the point. Very subjective. But based on the studies we have it basically looks like in general there are three genes working together and we can come up with a basic trend that we can follow.

Basically black seeds are dominant to other colors. Brownish or greyish seeds with a particular black mottling striping with black dots is next in line. Tan or brown seeds are probably next in line. Green seeds (not pictured here and rare) are dominant over red. Red seeds are the most recessive except for white. White seeds are the most recessive and recessive for all three gene combinations. This is a very simplified interpretation and there are probably actually more than three genes. In my population i have grey seeds which is not a color that has been studied. Also i have no idea what “tan” actually is so i lumped it in with brown. Brown too has not been studied, nor has “reddish-brown” among others.

Watermelon_fruit
Watermelon Fruit Shape is Co-dominant. Elongate (OO), Oval (Oo), and Spherical (oo).

Watermelon Fruit shape is relatively simple however. Yay! Simple co-dominance at work. Two long genes (OO) give you long fruit. One long gene and one round gene (Oo) or heterozygous gene pairs give you medium oval shaped fruit. And two copies of the other round gene (oo) gives you round spherical fruit. Easy peasy!

Golden-rind fruit are easy genetics too. Simple recessive (go). This is a trait more common now as it helps people identify when a watermelon is ripe. They turn bright yellow when ripe.

Watermelon_yellow_rind
Yellow-rind fruit are recessive (go). Fruit become golden yellow as they mature.

And the last trait i will mention is the “explosive rind” trait.

Haha, it’s not as scary as it sounds, but it’s not particularly a trait you want in your watermelons. Fortunately it is recessive and hopefully you wont encounter it in many varieties. I’ve seen it in the unusual striped variety but fantastic tasting ‘Osh Kirgizia’ watermelon, but otherwise not that much. Officially explosive rind (e) causes the fruit rind to burst or split when cut. This is true, but i also find that often when this trait is present the fruits themselves have a higher rate of splitting open while ripening on the ground and even when you lightly grab one to harvest. Not a trait that a market grower would want. For a small backyard gardener it’s not a huge deal as you can eat them right away, but still a slight inconvenience, especially if they split in the field and ants get to them. Black ants really do love sweet watermelon flesh.

Watermelon split
The recessive explosive rind trait (e) causes watermelon fruit rind to burst or split

 

Advertisements

New Watermelon Breeding Project 2018 and Beyond…

20170927_175812

Today i’m sharing about a new plant breeding project i am planning on working on. The Watermelon Landrace project I’ve been developing for Northern Colorado has started to progress quite well and i am very pleased in the direction it is heading. This past summer of 2017 i harvested many that were of decent size, grew in my soil, and tasted excellent. I started to eliminate the ones that still develop blossom end rot and other poor traits such as funky shape or poor flavor. Starting to only save the best seeds.

Citron-Red-Seeded-watermelon

I originally added some Colorado Red-seeded Citron melons to my watermelon landrace because i wanted to breed watermelons with red seeds and frost tolerance. Citron Melons are supposed to be pretty damn hardy and supposedly have this desired frost tolerance. The problem? Citron melons aren’t exactly edible. They are not poisonous, just super hard white flesh and bland bland bland. Actually they are a very old heirloom type of watermelon called the Colorado Preserving Melon or the Colorado Red-seeded Citron. Apparently they have lots of natural pectin in them which is useful for making jams and jellies for toast. And did i mention they can breed quite easily with modern watermelons?

Citron-Zimska

When it comes to the Colorado Red-seeded citron i absolutely love their red seeds. I really want that trait in my watermelon landrace. I guess there are a few red seeded watermelon varieties out there already, but they are few and far between.

So, what happens when you breed a modern red or yellow fleshed watermelon with a citron melon? Well, i don’t exactly know. Yet. This year i planted a few of the red seeds i harvested from the citrons from last year that were mixed in with the landrace. The seeds i got were all still red so i figured they probably self pollinated. Regardless i added them to the landrace watermelon seed i planted this year in hopes that they would grow (not die), cross, and produce viable seed.

V_Watermelon_JamMelon

I’m happy to report that so far that part of the project was a success. The top photo above shows what i think are confirmed F1 Citron x Watermelon hybrids. I suppose they could be F2, but i’m just going to assume F1. The seed was harvested from fruits that showed the characteristic “white cloverleaf striped mottling” that Citron Melons have, and from fruits that had hard white bland flesh when all the other watermelons had ripe yellow or red flesh. How do i know these seeds are hybrids? Well because the seeds were not red this time! In fact they were all different kinds of patterns and colors. Some red-black, some pure black, some greyish, some grey-black-mottled, etc.

Looking forward to growing this line of seeds out and reselecting for the traits i want. Red seeds would be awesome, but not necessary. Frost tolerance would be even more awesome, but not necessary. Even without those traits what impresses me most about Colorado Citron Melons is the fact that they grow so darn well in my climate, with my poor soil, and still grow full size melons even when over crowded with other watermelons that don’t do well, and even thrive with relatively low amounts of water. These traits alone are so very desirable to be folded into my watermelon landrace that this project is so exciting even now when i’m just beginning.

20170927_175805

I’ve heard a rumor that way back the Soviet Union (USSR) did lots of plant breeding experiments (maybe because of the breeding genius known as Ivan Vladimirovich Michurin), and part of these experiments involved Wide Hybridization or Distant Hybridization, which means crazy breeding like interspecific, intergeneric, intrageneric, and intraspecific breeding and attempted crosses that most people would never try or attempt. Some of these crosses were successful. What i’m interested in is the Soviets work on Citron-Watermelon hybrids. Apparently they experimented with these long before i have and rumor has it that they were able to recover some nice tasting watermelons that were able to be stored for several months into the winter. Awesome. I will update this blog post when i have more information about this. There are already supposed “Winter Watermelons” that supposedly keep for several months, but i’m sure those can be improved, or i can just breed my own winter watermelon variety. Exciting stuff!

Edit Nov-15-2017:

Okay, so i finally received a copy of the rare book titled “Wide hybridization of plants (Otdalennaya gibridizatsiya rastenii) Proceedings of the Conference on Wide Hybridization of Plants and Animals; collection of reports” from inter-library loan. A mouthful, i know. Thanks to WorldCat to helping me track it down. Not many copies of it left around.

Originally written and published in Russian in the U.S.S.R. in 1958, and Translated into English in Jerusalem Israel in 1962. The Soviet Union was known in those times for great scientific advances including launching the space race, the first cosmonaut in space, Sputnik, and other crazy medical advances like the Skenar and Bacteriophage medicine, to strange sci-fi spy weapons in the Cold War. Apparently they also were advancing in novel plant breeding techniques and programs. Michurin was one of these guy’s. If you’ve never heard of him or his plant breeding techniques and success go look him up. I honestly don’t know much about him myself, but i do know he was an accomplished plant breeder, most notably with wide genetic crosses that noone else thought would work.

Anyway, back to the Interspecific Hybridization of Watermelon work done by the Russians with Citron x Watermelon crosses in 1958. Turns out they did have success with it. The F1 generation was mostly like the wild Citron with bland hard tasteless flesh. F1 and F2 Hybrids with Citrullus colocynthoides are similar to their wild Citron parent genetics. Late-ripening, coarse compact unsweet fruit pulp and a thick rind. Quite unremarkable. But that’s what i was already expecting. One cool note though is that some of these in future segregating generations or backcrosses to domestic watermelons can produce some sweet watermelons that have some storage ability. Meaning they ship well and can store for many months. In fact some of them get sweeter over time whereas domestic watermelons do not. So all in all some cool potential in the project after all! I’m even more excited now!

If you want to read the Soviet’s 1958 watermelon research yourself, i have taken the effort to scan some of the book into a PDF for you. It’s not the whole book, but it has the relevant chapter on Watermelon and Citron crosses.

Here’s the PDF: Wide_hybridization_in_plants_TSITSIN

Success with Watermelon Landrace in Colorado

EDIT 1-29-16: This post is probably more accurately referred to as a “proto-landrace” or a pre-landrace. This post was from 2011, but even though it was my first attempt at starting my own watermelon landrace i still feel it was a moderate success. I hope to update you with how my watermelon landrace has progressed in the 2016 season. It is starting to do really well and selection pressures have occurred that i feel it can truly be called a landrace at this point and even perhaps the beginning of a variety if selected even further. Better than any commercial variety of watermelon by far!

I’m pleased to announce that my watermelon proto-landrace experiment seems to have been a moderate success so far, and here in Northern Colorado no less!

Northern Colorado isn’t exactly known for having a climate that is hospitable to growing watermelons. In fact, I’ve tried planting and growing watermelons at least three times before, each time ended in a miserable failure. This year was different though in part by my determination to succeed and the landrace method i recently discovered. A landrace is basically a mix of a bunch of seeds with lots of genetic variability. That variability makes it easier to locate a variety that is already at least partially adapted to my climate. While some (or many) varieties in a landrace die off immediately, some usually survive. And the theory is that by selecting only the best ones year after year, we eventually get something that is really good. Much of my inspiration for this project has come from Joseph Lofthouse who has had success with a cantaloupe landrace in utah, the breeding work being done by Rebsie Fairholm in the U.K., and indirectly Carol Depp who i believe inspired the two people i just mentioned. When used modernly and deliberately some people refer to it as “evolutionary breeding”.

This is where my experiment with this idea comes into play. I’ve always liked eating watermelon, but the common belief here in Colorado is that they are hard to grow (if not impossible). There is at least some merit in that belief. I myself have failed previously on three separate occasions to cultivate Mark Twain’s angelic food. The biggest problem was that i never used the right varieties. In a sense i guess you could say that the landrace method is basically a shotgun approach. When in doubt, shoot everything!

And while this is only the first year i have attempted this project, i feel the results so far have been extremely encouraging. This is the first year i have ever gotten any ripe watermelons, which in itself is a milestone. I got dozens of them. Many of them were tiny though, like baseball tiny. Some of those tiny ones actually ripened though, and a few tasted really good. I also got some that were of decent size too. I tried my best to select varieties that were mainly Northern, Native American, and short season varieties. My hope was that some might already be adapted to the temperature requirements, some adapted to the soil, pests, etc, and they would be able to cross pollinate with each other, so i could eventually select for one that has all the best traits.

Even with this criteria probably a third of them did not ripen properly, and went directly into the compost pile. But, i was able to harvest a few really good ones, and few OK ones as well. Something that was interesting though was that many of the early ones to ripen were yellow-fleshed watermelons. I had included some seeds from a few that were supposed to be yellow, and i had heard myths of yellow watermelons, but i didn’t really think i would actually be able to grow my own. The first yellow i ate was probably the best watermelon i’ve ever tasted. I didn’t know the yellow ones tasted better than the red ones! I’ve found out that there might actually be three different shades of yellow watermelons, and i think i saw all three this year. What’s remarkable though is that one of the yellow types is a dominant trait, while one is a recessive trait.  How crazy is that! I suspect that the yellow ones might be able to ripen better here in my climate than many of the red ones, but maybe it was just a coincidence. only time will tell.

Anyway, yeah. This whole landrace breeding is pretty darn awesome! I hope to continue selecting the best of the best, and maybe eventually i will have my own special variety that grows vigorously here in Northern Colorado. 🙂 If you’d like to see a few more pictures and details about this breeding project, then please visit my website at biolumo.com.