Breeding a purple russet potato from seeds

I've finally got my garden to a point where 2018 can be the year of "play."

Something I really, really want is a dark purple fleshed potato that is actually a russet potato that would be perfect for french fries, purple ones. A potato that has a decent size and would be a good keeper.

So - in my research and dabblings with genetics of potatoes and seeds and whatnot, I'm kind of thinking I might just start with two and not grow others. would it make sense to...

choose a potato that has the dark purple flesh and a proper russet keeper/french fry potato and cross them and just keep regrowing the results until I get what I want and narrow it down? Am I correct in thinking the genes, if those are the ONLY potatoes that I grow, would continue just to recombine?

Assuming, of course, I choose two varieties that would be compatible (in the sense that I'd have at least one potato with a fertile stigma and one that has fertile pollen). And I'd most likely landrace it, so I'd grow a bunch, select the closest few to regrow to cross pollinate, and so forth.

All my life I've been growing tomatoes, a lot of varieties. Heirlooms mostly and an occasional hybrid like Early Girl or Big Boy. I've been saving seeds for years and growing them the following year. I've never seen a cross where a variety changed, except that hybrids usually return as a cherry tomato.

A potato is a similar plant, from the same family, I think. I do know you can graft a tomato onto a potato plant and get a franken plant; no I mean a tomato on top with potato tubers under the tomatoes. What I'm thinking is why would the plants cross. If I grow two different potatoes in my garden and save some tubers for next year or if I save seeds I'll get a cross? But for that matter can I just let some go to seed can I even get the same variety of potato next year.

What I'm asking is will they cross?

John Duda wrote:All my life I've been growing tomatoes, a lot of varieties. Heirlooms mostly and an occasional hybrid like Early Girl or Big Boy. I've been saving seeds for years and growing them the following year. I've never seen a cross where a variety changed, except that hybrids usually return as a cherry tomato.

A potato is a similar plant, from the same family, I think. I do know you can graft a tomato onto a potato plant and get a franken plant; no I mean a tomato on top with potato tubers under the tomatoes. What I'm thinking is why would the plants cross. If I grow two different potatoes in my garden and save some tubers for next year or if I save seeds I'll get a cross? But for that matter can I just let some go to seed can I even get the same variety of potato next year.

What I'm asking is will they cross?

Edited because I re-read your post :) So the plants would cross because I'd cross pollinate them manually. I'd then take the resulting seeds, grow them into tubers, and allow the best ones to flower and set more fruit, etc. until I got what I'm looking for.

My first step would be to try all known purple fleshed varieties to see if any of them are good russets. Who knows someone like Tom Wagner may already be selling such a plant!

If that failed I would do as you suggest and cross a purple fleshed potato with a russet. If my trials found something closer to what I desired I would start there.

The first filial generation or F1 is a hybrid. It could potentially be a good potato in its own right.

The second filial generation or F2 will segregate. Because potatoes can be cloned by saving tubers you can stop at an F2 if you find a good one. So your goal could be as simple as finding a really good F2.

Further generations F3, F4, etc might not be useful- once segregation has occurred some genes are lost. So instead of generating good new combinations they may simply be dead ends.

So here is what I would do instead: try to find a potato in the F2 with really good blue or purple coloration. Then backcross it to the russet parent. Eventually if you do this process recursively you should get a really good purple russet. Alternatively if the coloring is the problem you can backcross to a good color parent.

Though your perfect potato could show up in any generation.

Bethany Dutch said:

So the plants would cross because I'd cross pollinate them manually. I'd then take the resulting seeds, grow them into tubers, and allow the best ones to flower and set more fruit, etc. until I got what I'm looking for.

What I'm thinking is that in all my years of growing heirloom tomatoes, why haven't I ever seen a tomato that was cross pollinated by a bee. If that'd happened then the seeds that I grew the following year would produce something dissimilar to what I had the year before. I'm not questioning critically, I'm just asking what seems to me a logical question.

John Duda wrote:in all my years of growing heirloom tomatoes, why haven't I ever seen a tomato that was cross pollinated by a bee. If that'd happened then the seeds that I grew the following year would produce something dissimilar to what I had the year before.

Modern tomato varieties tend not to be promiscuous enough due to the anatomy of their flowers.  

You might find this thread interesting - Promiscuously Pollinated Tomatoes And The Bees That Make Them Possible  

John Duda wrote:Bethany Dutch said:
What I'm thinking is that in all my years of growing heirloom tomatoes, why haven't I ever seen a tomato that was cross pollinated by a bee. If that'd happened then the seeds that I grew the following year would produce something dissimilar to what I had the year before. I'm not questioning critically, I'm just asking what seems to me a logical question.

Tomatoes usually pollinate themselves before the blossom opens, so bees are always too late. There are some bumblebees (at least here in Europe) that bite through the blossom to get to the pollen, so there is still a small chance of cross pollination. There are also some tomatoe varieties that are open pollinating because of their unusual blossom shape.
Potatoes do not breed true. I can not tell you the reason or the difference to tomatoes, but i tried it myself and the results are wildly different from the parents.

edit: Burra beat me to it, typed too slow...

As potatoes are heterozygous tetraploids you should not use the same strategies you use with homozygous tomato heirlooms.  For example, the F2 generation will produce lots of variation with a potato while it will produce virtually none between siblings with tomato heirlooms.  So grow lots of seedlings out and you may very well find what you're looking for in the first generation.  

Your project sounds great.  Please repost updates with pictures!  :)

Greg Martin wrote:As potatoes are heterozygous tetraploids you should not use the same strategies you use with homozygous tomato heirlooms.  For example, the F2 generation will produce lots of variation with a potato while it will produce virtually none between siblings with tomato heirlooms.  So grow lots of seedlings out and you may very well find what you're looking for in the first generation.  

Your project sounds great.  Please repost updates with pictures!  :)

You mean the F1, and you are right. The F2 will have lots of variation in either case but with clonally propagated plants like potatoes or apples which are not inbred the first generation will have lots of variation. Unless you can find inbred lines.

Varation is great but the principles of descent are still important. If you want to combine the two traits russet and purple you need those two traits present in the parents. It may not be perfectly straightforward to get them. It may be that one of them could be a quantitative trait and not the sort of simple traits Mendel worked with in his peas.

Mendel found that simple traits often are recessive in the F1 and that could be true even if things are complicated by the heterozygosity of the parent potatoes.


So where do you start? With the best parents possible. I would define the best parents as those that are already closest to your goal. Tom Wagner is one of the most famous tomato and potato breeders living. I did a quick search of tater mater seeds and found his "Blue Blood Russett" available as true potato seed. That could be a really good place to start. Looks like there are only 3 units left in stock.

http://tatermaterseeds.com/shop/index.php?main_page=product_info&cPath=2&products_id=119

Some of Toms true potato seed lines are more true breeding than others. This one sounds from the description like it regularly produces small blue fleshed russetts.

So I would try "blue blood russet" if I liked it but felt it lacked something I would try to cross it to a parent that had that trait. Either a Russett or a purple.

Another thing is some potatoes set true seed easier than others. By starting with true seed lines it seems more likely that they will be true seeding. Though potato breeders do have work arounds to encourage seeding in lines that don't often have berries. One of these is to raise the plants in a shallow soil layer on top of some bricks. Which sounds simple enough!

"Blue Blood Russet" might not even be the only or best candidate on Tom's website. Its just one i found pretty quickly. To start the project I would make sure I had true seeds or tubers from as many blue and purple potato varieties that sounded promising to me. What all other traits are really important? If you want a large deep purple Russett that makes great french fries your goals might include large size and good frying texture. If you additionally want good productivity your goals might really be a plant that produces lots of large deep purple russetts with good frying ability. All those things exist. So the trick is to make sure that all those traits are present in the parents or as many as possible. If Blue Blood Russett produces only small tubers and they are blue and not deep purple, but the flesh has good frying ability- then your ideal second parent might be deep purple with large tubers. I think you would still be much closer to your goal then if you started with a purple and a white russett- because you would instead be standing on Tom's figurative shoulders to get you part way to your goal.

William Schlegel wrote:

Greg Martin wrote:As potatoes are heterozygous tetraploids you should not use the same strategies you use with homozygous tomato heirlooms.  For example, the F2 generation will produce lots of variation with a potato while it will produce virtually none between siblings with tomato heirlooms.  So grow lots of seedlings out and you may very well find what you're looking for in the first generation.  

Your project sounds great.  Please repost updates with pictures!  :)

You mean the F1, and you are right. The F2 will have lots of variation in either case but with clonally propagated plants like potatoes or apples which are not inbred the first generation will have lots of variation. Unless you can find inbred lines.

Varation is great but the principles of descent are still important. If you want to combine the two traits russet and purple you need those two traits present in the parents. It may not be perfectly straightforward to get them. It may be that one of them could be a quantitative trait and not the sort of simple traits Mendel worked with in his peas.

Mendel found that simple traits often are recessive in the F1 and that could be true even if things are complicated by the heterozygosity of the parent potatoes.


So where do you start? With the best parents possible. I would define the best parents as those that are already closest to your goal. Tom Wagner is one of the most famous tomato and potato breeders living. I did a quick search of tater mater seeds and found his "Blue Blood Russett" available as true potato seed. That could be a really good place to start. Looks like there are only 3 units left in stock.

http://tatermaterseeds.com/shop/index.php?main_page=product_info&cPath=2&products_id=119

Some of Toms true potato seed lines are more true breeding than others. This one sounds from the description like it regularly produces small blue fleshed russetts.

So I would try "blue blood russet" if I liked it but felt it lacked something I would try to cross it to a parent that had that trait. Either a Russett or a purple.

Another thing is some potatoes set true seed easier than others. By starting with true seed lines it seems more likely that they will be true seeding. Though potato breeders do have work arounds to encourage seeding in lines that don't often have berries. One of these is to raise the plants in a shallow soil layer on top of some bricks. Which sounds simple enough!

"Blue Blood Russet" might not even be the only or best candidate on Tom's website. Its just one i found pretty quickly. To start the project I would make sure I had true seeds or tubers from as many blue and purple potato varieties that sounded promising to me. What all other traits are really important? If you want a large deep purple Russett that makes great french fries your goals might include large size and good frying texture. If you additionally want good productivity your goals might really be a plant that produces lots of large deep purple russetts with good frying ability. All those things exist. So the trick is to make sure that all those traits are present in the parents or as many as possible. If Blue Blood Russett produces only small tubers and they are blue and not deep purple, but the flesh has good frying ability- then your ideal second parent might be deep purple with large tubers. I think you would still be much closer to your goal then if you started with a purple and a white russett- because you would instead be standing on Tom's figurative shoulders to get you part way to your goal.

Someone give this man an apple!!

Thank you so much, you've been tremendously helpful. I was on Tom's website yesterday but I can't figure out how to order (there is no "add to cart" button) so I had emailed them but haven't gotten a response yet. I agree it would make the most sense to order seeds from him and get started that way and see what I got. Otherwise I thought I may order Purple Majesty potatoes which seem to already be fairly close to what I want (but I do want something bigger in size) and cross it with an Idaho or Kennebec.

Yes, sounds like a fun project. You should start with varieties that are known to be TPS varieties. Tom's varieties would be a good place to start, but i don't think tom is selling seeds anymore from his website. (i could be wrong).

The difference is that you will most likely be working with tetraploid potaotes (despite diploid potatoes existing).

That means that it is a little more complicated than diploid mendel type genetics and may take you longer to isolate recessive genes and may take more generations due to double the heterozygosity in the F2 generation. So in that sense, no, traditional breeding terminology and results should not be used or expected.

Here is my all time favorite potato breeding book! Even Tom Wagner left an outstanding review for it!

The Lost Art of Potato Breeding (Garden Alchemy) by Rebsie Fairholm

https://www.amazon.com/Lost-Potato-Breeding-Garden-Alchemy/dp/190801119X



edit: in addition here are some potatoes being bred by Colorado State University that you might be able to request and grow.

https://source.colostate.edu/csu-researchers-develop-nutrient-rich-potato/

http://potatoes.colostate.edu/wp-content/uploads/pdf/Potato_Breeding_Photo_Essay.pdf

http://potatoes.colostate.edu/programs/potato-breeding/cultivars/

http://potatoes.colostate.edu/potato-breeding/