Monday, November 10, 2014

Oca: New ocas from seed

Here is a slideshow of new oca varieties harvested in 2014.  I'm pretty happy with the results so far.

Tuesday, November 4, 2014

Yacon: Stem Cuttings

I have something of a love/hate relationship with yacon.  I love to eat the storage tubers, but I hate dealing with storing the crown.  I lose a lot of it over the winter no matter what I do.  I can leave it outside here, but a lot of it rots in the wet soil.  I can bring it inside, but almost always lose a large portion to some combination of rotting, dessication, or early sprouting.  This year, we finally have a walk-in cooler, so I hope that will make a difference, but that is obviously not a solution for the casual yacon grower.

If you live in a mild climate, where you can overwinter potted yacon in a tunnel or a greenhouse, you might consider taking stem cuttings.  They root pretty easily and give you a head start in spring.  This is obviously also a good way to multiply up a variety.  Between stem cuttings and rhizomes, you could pretty easily turn one yacon plant into fifty.

I take two kinds of stem cuttings from yacon: vertical cuttings on the lower part of the stem where it is still solid and horizontal sections from the upper parts of the stem that are hollow.  The reason for the difference is that I had problems with the hollow stems rotting before they would root and the horizontal sections expose the cut interior of the node for rooting, which seems to help.  I noticed that lodged yacon stems sometimes reroot on their own, but only where the stem has cracked and split.

I don't usually take tips for cuttings, although this might be the most tempting approach.  The only reason for this is that I try to preserve yacon flowers as long as possible in the hope of getting some seed.  Even with lodged plants, I can remove the lower parts of the stem and the put the upper part with the flowers in a bucket of water while they finish.  By the time that is done, the stem isn't in very good condition for taking cuttings.

These are certainly not the only ways to root yacon and they may not even be the best ways.  If you have a method that works better, please let us know.

Here's a brief video that demonstrates both techniques:

Friday, September 12, 2014

Ulluco: 500

This has been another exciting year on the ulluco (Ullucus tuberosus) front.  Last year, I was able to collect almost 100 ulluco seeds, which is a rather rare event.  The last published work in English that reported production of ulluco seed was in the early 1990s, although there are some publications in Spanish that at least suggest that more seed has been produced in South America since then.  Anyway, you get the idea: ulluco seed is not all that common.

As of yesterday, I have collected more than 500 ulluco seeds in 2014.  The news is even better than that, because I collect entire inflorescenses with seeds forming and wait for them to mature.  I have eight small jars of ulluco inflorescenses that should conservatively yield another 300 seeds and there are still many seeds forming on the plants.  It seems likely that I will collect more than 1000 seeds and it is not inconceivable that I could end the season with a total in excess of 2000.

500 ulluco seeds
So, this year has shown that we can repeatably produce ulluco seed here and potentially in quantities large enough to support a substantial breeding effort.  I have also collected seeds from nine varieties this year, which is a big increase over the two varieties that set seed last year, although most of those varieties have produced a very small number of seeds.

That is the good news.  The bad news is that the majority of these seeds will never grow an ulluco plant.  A few weeks ago, I dissected 100 ulluco seeds and found the majority of them empty or nearly so.  Only 1 in 7 seeds contained an apparently mature and well-formed embryo.  So, these 500 seeds will probably yield no more than 70 mature embryos and, given the great length of time since ulluco has commonly reproduced sexually, many of those are bound to fail.

Ulluco embryo undergoing a simple paper towel germination test
Knowing what I know now, I plan to try to start the majority of the seeds this year through embryo culture.  If you aren't familiar with the idea, it is pretty simple: You cut open the seed and remove the embryo.  You then put the embryo in a sealed container on a gel medium with some basic vitamins and minerals and a little bit of sugar.  The only tricky part is keeping things sterile so that you don't instead end up growing a fungal or bacterial culture.  If you do everything right, you have basically produced an artificial seed.  Although it is more work up front, it eliminates the waste of trying to grow empty seeds and allows me to see what is going on.  Observing the response of the cultured embryos to conditions may help to determine just what the proper growing conditions are.

(Incidentally, the 14 well-formed embryos that I found in the first 100 seeds have been deposited into both sterile cultures and simpler wet paper towel tests and I hope that they will eventually grow.  They have certainly imbibed and increased in size, but no further development yet.)

I'm feeling pretty confident now that I will be able to get some ulluco seedlings.  Hopefully I can get some to grow over the winter and have plenty of material to plant out next year.  Cross your fingers!

(Update: Just 5 days later, September 17th, we passed 1000 seeds.  That means about 140 seeds that at least have a shot a germination.  The odds just keep getting better!)

Sunday, September 7, 2014

Carrot: Ox Cart F3

Back in 2009, I set out to breed a new carrot variety out of the ashes of the Oxheart carrot, for which I could only find poor quality seed.  Oxheart is a good variety for growing in hard soils, as it is short and stout.  Unfortunately, it seems to have been grown by few people in recent decades and the sources of seed that I was able to find had evidence of in-breeding depression in some cases and crossing in others.  The result was a jumble, with less than 20% large, good quality Oxheart type carrots and a lot of smaller, off-type carrots.

So, I grew out and selected the best and then crossed in two other varieties that we grow: St. Valery and Parisian Market, plus a small amount of a variable orange carrot that we started from a mix several years back.  St. Valery is too large, but has the right shape, right maturity, and good flavor.  Parisian Market is a small, stubby carrot.  Between the two, I hoped to get a good mix of different genetics from which to select.

We're calling this carrot Ox Cart, a clear pointer back to its heritage.  Even though the goal is to produce a carrot as much like Oxheart as possible, it will be genetically distinct, so we have to give it a new name.

We've now gotten to year four, the F3 generation, and I had hoped to have a reasonably consistent harvest this year, but it is clear that we are probably still several years from that, although we're making progress.

Here is a sample:

Ox Cart F3: No cracking, but still a lot of variability
The traits that we are selecting for fall in the middle of the picture.  Toward the left we have probably too much St. Valery with longer form (although we've mostly gotten the long carrots out this year) and sloping shoulders.  Toward the right, we have stubbier carrots that favor Parisian Market.  In the Middle, we have stout carrots with broad shoulders.  We want them more triangular and less cylindrical, so that rules out some of the middle carrots as well.

Cracking appeared in the first year and has persisted until now.  Unfortunately, the cracking appears mostly in the carrots that have the form that we want, which means that we're having to eliminate a lot of carrots that otherwise look very good.

A monster carrot, but a bit too monstrous for this project
This is a nice big carrot, but bigger than we want for this project.

Almost perfect
This one is just about perfect.  Broad shoulders, strongly triangular, short, clean, and free of cracking.  Hopefully a couple more years of selection will yield a crop that looks mostly like this.

Thursday, September 4, 2014

Mashua: How to hand pollinate

In the northern hemisphere, outside of the tropics, where day length exceeds about 13 hours, we're just entering the part of the year when mashua (Tropaeolum tuberosum) flowers.  Some varieties are starting now, while others won't begin until after the autumn equinox (September 23).  Most commonly, mashua self pollinates, because it begins to shed viable pollen before the flower opens.  In many ways, that is a good thing.  If you have one variety of mashua and a long enough frost-free autumn, you can obtain seed without doing anything but waiting to collect them.  Because mashua is a tetraploid, you can also count on some variability in the progeny even when seeds result from self-pollination.

But if you want maximum variability, you need cross pollination.  If you do nothing and you have multiple varieties of mashua flowering at the same time, you may get a little crossed seed, because mashua pollen is not at peak viability when the flower opens.  That means that there is a short window, perhaps a day, when natural pollinators may give you some crosses, but you will have a few crossed seeds mixed in with a lot of self-pollinated seeds, which is not ideal.  To get purely cross-pollinated seeds, we must turn to a tried and true plant breeding technique: emasculation.

First, observe your mashua flowers for a few days.  You want to find flowers at the stage about 24 hours before they open and the easiest way to learn what they look like is just to watch them open for a little while.

When you are ready to make a cross, find yourself a soon-to-open mashua flower that you want to use as the female parent.  (You obviously need to leave the flower attached to the plant.)

The next part is a bit barbarous.  Don't let your seedlings watch.  Remove the sepals and petals.  It is easy enough just to rip them off with tweezers, although you can use a razor knife if you are fastidious.

This exposes the stamens, which should not yet be shedding pollen.  If you touch them with a figertip and see pollen, you're too late.  Use tweezers to pull them out or snip the filaments with small scizzors.  Be careful, as the stigma is hidden in the middle of the stamens and you don't want to remove it.

One you have removed the stamens, the stigma is exposed.  The flower is now ready for pollination, but results seem to be better after 24 hours.  I bag them until the next morning and then transfer pollen to the stigma.

To pollinate, find a flower that has already opened and pluck one of the stamens.  Give it a little rub against the stigma and your job is done.  You can also use a brush in order to leave the stamen on the other flower, but mashua makes so many flowers that I don't lack for stamens to use.  I put the plastic bag back over the flower when done to protect it.

If all goes well, you will spot seed pods forming within a week and in about five weeks you will have seeds.

You might wonder why it is worth going to this effort.  Asking someone who breeds plants this question is kind of like asking a mountain climber why he climbs mountains.  Answer: Because it is there.  You don't really need a reason or even a goal in breeding plants; sometimes it is just to see what happy accidents nature has in store.

That said, mashua is a particularly interesting plant for the breeder because it combines strongly appealing and unappealing traits.  On the positive side, it is vigorous, pest-resistant, and produces a large amount of food in a small amount of space.  On the negative side, like many other Andean root crops, most varieties don't produce tubers until late in the season outside the tropics and, frankly, most varieties don't taste very good.  That said, some mild varieties taste a lot like turnip and at least one variety (Ken Aslet) begins to form tubers earlier in the season.  So, while I'm open to any happy accidents that result from mashua breeding, my main goal is to produce a variety that is both early and mild-tasting.  These traits already exist; how hard could it be to get them into the same package?

Tuesday, September 2, 2014

True oca seed now back in stock

Oca seeds
Oca (Oxalis tuberosa) seed is now back in stock in our shop.  Depending on the rate at which orders come in, we may run out periodically, but I expect to collect enough to satisfy orders through spring, so if you want to wait to buy when tubers are available, that shouldn't be a problem.  (No guarantees, of course; we could get an unexpectedly large number of orders.)

Our shop is currently in a bit of disorder as we get ready for a full update in November.  Most items are out of stock.  Here is a link that will take you straight to the oca seeds.

If you ordered last year, you will notice some things that are different:

Seed Groups

We are not offering controlled crosses this year.  Once I calculated the time that it takes to produce those seeds, I arrived at a price that few would be willing to pay.  So, we don't plan to offer controlled crosses for sale again.  Instead, we are offering open pollinated seed collected from two different groups of plants.

Group 1 is collected from all of our popular varieties.  This is comparatively easy since we grow a lot of each variety, which makes seed collection less laborious.  We think of this as a high quality but lower diversity batch.  You get double the number of seeds (60) that we offer for Group 2.  18 varieties contribute seed to this batch.

Group 2 is collected from our preservation and experimental varieties, which involves a lot more labor, so you only get 30 seeds.  We think of this as a high diversity but probably lower quality batch.  If you want to be on the cutting edge and get maximum genetic diversity in your seed, this is what you should choose.  About 30 heirloom varieties and more than 300 varieties from our breeding project contribute seed to this batch.


Collecting oca seed involves a fair amount of labor.  Last year we took a guess at a reasonable price, but this year we actually tracked the time involved and came to the conclusion that prices had to be increased.  So, Group 1 is $9/60 seeds and Group 2 is $9/30 seeds.

Bulk Seed

Last year we offered larger quantity packets as well.  These were very popular and the result was that we ran out of seed early.  In order to make sure that more customers can get seed, we're not offering larger packets this year, although we might do so in late spring if we still have a lot of inventory.

We're still planning to begin tuber sales sometime in November and we will have a lot of exciting new offerings this year if mother nature takes it easy on us.

Happy growing!

Thursday, August 28, 2014

Ulluco: Ulluco Seed 2: This Time It's Boring!

Sequels are almost always disappointing and the followup to last year's ulluco seed discovery is no exception.  I had hoped by now to be able to report germination of some of the seeds that I collected last year, but I can't do so conclusively.  I think that a couple of seeds have germinated and then failed shortly thereafter, as the seeds have split open to reveal what looks like an embryo.  These seeds have not succeeded even in unfurling cotyledons though, so there may be something else going on.

I did find something interesting in the bed where I collected seeds last year.  This ulluco plant appears to be a seedling:
Possible ulluco volunteer from seed
It is possible that it became disconnected from a small tuber, but the roots don't look that way to me.  I've moved it to a pot for further evaluation at the end of the season.

I'm having fairly good luck with ulluco seed again this year, although I am somewhat embarassed to detail the amount of time that I have spent on it.  I spend about eight hours each week doing hand pollination and about four hours searching out and collecting seeds.  For that investment of roughly 72 hours so far this year, I have collected 79 seeds.  So, roughly one seed per hour.  Ouch.  I have listened to some good audio books along the way though.

As with last year, I am getting much better results with plants on which I have done hand pollinations.  This year, I reserved some plants as controls to try to determine whether all that fussy work with a tiny paint brush is really necessary.  So far, I have collected 78 seeds from plants on which I've done hand pollinations and 1 seed from plants on which I haven't.  It's a beauty though:

One nearly-mature seed hanging at the bottom of the inflorescense
This does inspire the hope that, if I grow an even larger number of plants, I might be able to dispense with hand pollination.  Of course, if I have to search a much larger number of plants for less frequent seeds, it's probably a wash.

This year, I have collected seeds from the following varieties, ordered from most to least:

1. BK10425.2
2. Chugua Roja
3. Pica de Pulga
4. BK08607.1
5. Cusco Market
5. White/Pink Spots

BK10425.2 and Chugua Roja each set many more seeds than the other three varieties combined.

I collect seeds by removing the entire inflorescense whenever I spot a seed that within roughly a week of maturing.  The seeds are so difficult to find in the dense ulluco foliage that I'm afraid I'll never find the same one twice.  In the picture below, I've circled the seeds.  If you stare at the picture for a while, it may help you to recognize seeds on your own plants.  I wouldn't be surprised if other people have seeds on their plants but just havn't learned to notice them yet.

This tangle of inforescenses is the product of a four hour ulluco seed hunt
over 150 row feet of ulluco
To spot ulluco seeds, look for:

* Anything that is still attached below a gap on the inforescence where the higher flowers dropped.

* Thickened or different colored pedicels (very apparent on BK10425.2 and Pica de Pulga, but not on Chugua Roja).
* Flowers that look slightly larger than others when closed (probably really a seed still covered by sepals).

When uncertainty prevails, squeeze. Seeds harden early and don't yield to fingertip pressure.

Of course, you might be wondering why you should go to the effort of searching out ulluco seeds if the damned things don't germinate.  That is certainly a problem that needs solving.  I plan to try to force the issue this year by doing embryo culture.  This involves dissecting the seed and removing the embryo from the testa.  This eliminates the tissues that are typically responsible for enforcing dormancy.  The embryo can then be grown in sterile culture, from which point it develops in the usual fashion.

That all sounds wonderful and it is fairly trivial with large seeds.  Unfortunately, ulluco seeds are pretty small and the dissection is not going to be easy.  I hope to get 300 seeds this year, which might be enough to allow for mangling them as I ascend the learning curve.  If I get enough, I will also set some aside for regular germination experiments at different temperature and light levels and perhaps supplemental hormones as well.