Thursday, July 24, 2014

Oca: Early flowering

Last year, I struggled to find oca varieties to cross early in the season.  I had more short-styled varieties than mid-styled.  About a month after flowering began, I had enough different varieties flowering so that I was able to produce a good quantity of seed.

This year, I have a similar problem.  Flowering started a month earlier than last year, but almost all the early flowering varieties were mid-styled.  You can see in the chart below that from June 1st until July 3rd, I had mostly mid-styled varieties.  One long-styled and one short-styled variety did flower during this period, but they produced very few flowers.
This is not a serious problem for me.  Flowering continues all season, so it is just a matter of waiting for more varieties to flower.  As you can see, by the 13th of July, I had plenty of compatible varieties to work with and, ten days later, I'm almost to the point of having more flowers than I have time to work with.

But, looking at the bigger picture, if we're going to make oca a more common crop in the US, we need a larger breeding effort.  Unfortunately, most of the country has a climate that is not conducive to flowering - at least in the summer.  Although oca may survive warm summers, it generally doesn't flower in warm weather.  In most of the US where summers are mild, freezing weather often comes too early for oca.  Four apparent goals arise from these facts:

1. Make oca produce tubers earlier to beat the frost.

2. Make oca more resistant to frost so that it can stay in the field longer.

3. Make oca more resistant to heat so that it will flower in the summer.

4. Make oca flower earlier so that breeding work can be done during milder spring weather.

There are many people trying for goal #1.  I have no idea how much work this will require.  An early oca might pop up among this year's seedlings or it might take years of small, incremental adjustments to oca's day length sensitivity.

Goal #2 could have a lot of payoffs, but increasing cold weather resistance is a notoriously difficult breeding goal.  Despite the fact that oca is grown under conditions that flirt with freezing and that it would undoubtedly have been valuable to be able to grow it at higher elevations in the Andes, most varieties are not particularly frost resistant.  But, if the plant could hold longer in the field, that might give growers in hot summer climates and opportunity to collect seed during fall and to begin breeding better adapted varieties for their conditions.

Goal #3 is difficult to assess.  It doesn't get hot here.  I can't test for differences in heat tolerance and instead rely on what other growers report.  This may require the next goal to be completed first.

Goal #4 seems like it might be the best approach.  Early flowering would provide the opportunity to catch milder weather in the spring without the threat of frost.  Mature seed is produced in three to four weeks, so could be obtained by the beginning of July.  This might be enough of a window to do some oca breeding in climates where oca doesn't normally flower in the summer and early fall due to high temperatures and/or low humidity.

My experience shows that some varieties already flower early enough for this purpose.  That's good news; traits that already exist obviously take less work to breed into new varieties.  Unfortunately, most of the oca that flowered in June was mid-styled.  Mid-styled varieties can cross-pollinate (and self-pollinate) but the odds of success aren't that great.  Ideally, we need several varieties each of at least two of the flower morphs.  They haven't turned up this year, but maybe next year.  A combination of varieties with different flower morphs that begin to flower in early June (or even May if we're feeling ambitious) could provide an opportunity for expanding oca breeding out of ideal climates.

Sunday, July 6, 2014

Oca: How to hand pollinate oca

Several people have asked how I do hand pollinations with oca (Oxalis tuberosa).  It is hard to show in pictures, so we did a brief video demonstration.

The video will make more sense if you read these posts first:
Maximizing seed yield of crosses
Seeds from cuttings

A few points to clarify:

* All mentions of male and female flowers refer simply to the one that I have designated at the female (receiving the pollen) or the male (donating the pollen).  Oca does not have different male and female flowers.

* I do most crosses on cuttings indoors now due to a higher success rate, but you can use the same technique on the plant.

* You can use a brush to transfer the pollen instead of direct contact between the flowers, in which case you don't have to denude the flowers if you have a very small brush and good coordination.  You can also keep both the designated male and female flowers this way, instead of discarding the male.  If you only have a few flowers, that might be advantageous.  If you have hundreds of flowers to work with, this method is more efficient.

* If you are very careful, you can get two pollinations from each whorl of stamens on the flower.  For example, with one short-styled flower, you could pollinate two mid-styled and two long-styled.  I've done up to six pollinations from a single whorl of stamens, but the number of pods and the number of seeds per pod drops considerably after the second pollination.

* There are unopened flower buds on this cutting.  As they open, you can pollinate them as well.  There doesn't seem to be any problem with having five or six pods maturing on a single cutting.

Wednesday, July 2, 2014

Some quick updates

I have found very little time for blogging this spring and I don't anticipate much change through the summer.  We're growing more than ever and trying to manage it on top of having a "real" job.  As it happens, my real job is not likely to continue for much longer, so I may become a full time micro farmer just as the growing season comes to an end.  Although there may seem to be more than a little irony in that timing, being free around harvest time in Nov/Dec wouldn't be entirely unwelcome as it is going to be a big job.  Anyway, this may not prove to be much of a blogging year, but there is still a heck of a lot getting done.

I am spending a lot of time updating the Andean Root and Tuber Crops Wiki, which has replaced some of my blogging.  What it lacks in style, it more than makes up for in data value.  I've said a few times that the most important thing that we're growing this year is information and I become more convinced of this every day.  I'm now able to query for data about many varieties of plants and investigate potential correlations that would be very laborious to investigate otherwise.  New answers lead to new questions, but I am hopeful that even the more genetically complex Andean plants will yield some patterns that will be useful in breeding efforts.

Oca bed #1
All 2014 seedlings - every plant is unique

The oca crop is coming along nicely.  We have at least one sixteen foot row for each of 64 established oca varieties.  On top of that, we have another 44 rows in which we are growing this year's seedlings and mini tubers of last year's.  This comes in total to about 800 new oca varieties, despite our early season loss of about 20% of the crop.  We will keep only a small fraction of those.  I'm not sure what the number will be, but probably not more than a quarter of them,  Whether we keep them or not, of course, I hope to collect a full set of data for each plant.

Flowering began early this year, with our first flower from the variety Pale Pink coming in on June 1st.  We've since had another 12 varieties flower.  You can keep up to date on flowering by checking out our Flowering Report.

I've done about two dozen crosses, but have not yet harvested any seed.

I hesitate to mention the most ignominious oca development of the year, but this blog certainly magnifies successes and minimizes the failures, so it is good to add in some of the humbling details on occasion.  I was very excited to have the opportunity earlier this year to swap for some oca seed produced by a researcher in Bolivia.  That was a chance to introduce a lot of new genetics at once.  I received about 200 seeds, started them all and, although the germination was a lot lower than my own seed, I had almost 80 seedlings going.  Then, overnight, some kind of mold swept the trays.  The seedlings all went fuzzy and many of them died before morning.  They looked healthy when I went to bed and were dead in the morning.  In horror, I sprayed everything down with dilute hydrogen peroxide and, either due to or despite this quick action, saved eight seedlings, of which six ultimately survived.

I was rather distraught at the time, but now I have six healthy plants adding to our growing gene pool, which is six more than I had before.  So, not a total loss and, in fact, still a considerable gain in diversity.  The moral of this story is that it is probably a good idea to surface sterilize oca seeds.  I bake my potting soil in the oven, so the mold probably came in either on the seeds or from the air.  The widespread nature of the event suggests that the seeds were the source of contamination.  I can't do much about floating spores, but I can at least try to clean the seeds before sowing.

Survivors of the Great Bolivian Oca Massacre of 2014


I get more inquiries by email about our ulluco seeds than anything else.  I wish that there were more news about my favorite Andean crop.  A few of the ulluco seeds that I harvested last year have burst open, looking promising, but have then failed to develop.  I am now uncertain if this had anything to do with germination or was just some mechanical process.  I console myself that many of the seeds in the Finnish study took more than a year to germinate, so I'll just keep waiting.  Meanwhile, I hope to be able to get a lot more seed this year, which would give me the opportunity to do some experimenting with different germination strategies.

We are growing 18 varieties this year, about half of which I started from tissue cultures over the winter.  The plants are looking good and flowering began about two weeks ago.  I head out with my paintbrush to do pollinations whenever I can find the time.

I am also undertaking an experiment to try to produce octoploid ullucos as described by Viehmannova et al. in Induced polyploidization and its influence on yield, morphological, and qualitative characteristics of microtubers in Ullucus tuberosus.  I have treated ulluco cultures as described and have sixteen survivors.  Now I just have to wait until they grow large enough to check for chromosome doubling.  The odds aren't that great with only sixteen plants, but if it doesn't work, I'll keep trying.  It could be a route to improved fertility.  Or none at all.


Mauka Blanca plant
I am very excited to have a third try at mauka.  I have twice managed to kill cuttings that I was assured "root easily in wet soil."  My experience was more like "collapse readily into piles of corruption in wet soil."  Well, this time things seem to be going better!  I purchased a rooted plant of 'Blanca' from Sacred Succulents and while it hasn't exactly erupted into new growth, it looks healthy and has survived well past the two day period in which I have previously killed mauka.  Yay!

Even more exiting, a friend was able to obtain some mauka seeds and was kind enough to share them with me.  They germinated pretty easily and are now working on their first true leaves.  I believe that these are self-pollinated seeds of the variety Blanca.  They will probably be true to type, because mauka is a diploid (kind of unusual for an Andean crop).  So, I may have taken two routes to get to the same variety of mauka.

By the way, I soaked the seeds for three hours in water and then surface sowed on saturated potting soil on a 70F day / 60F night heat mat.  I had germination within a week and ultimately got 12 out of 15 to germinate, so I can recommend that method if you find yourself in possession of mauka seeds.

Mauka Blanca seedlings
All lopsided, with one cotyledon larger than the other
Those of you uninitiated in the mysteries of obscure Andean crops may be wondering what mauka is.  Nope, it is not the same as mashua nor maca.  Try to keep it straight, will you?  Mauka is probably the rarest of the Andean crops, having been known only from historical descriptions until it was rediscovered in some garden plots in the 1960s.  It was unknown to science, that is; I'm sure the people tending the gardens knew that it was there.

Anyway, I'll skip recapitulating descriptions that are available elsewhere and refer you to The Vegetable Garden for an intro to mauka.

Now I just need to find a plant of the variety Roja and I'll have the makings of a mauka breeding project.  Of course, that takes for granted that these plants survive the rest of the year, which may be tempting fate.


I am only half as good at killing arracacha as I am at killing mauka, but that is probably only because I have had half the opportunity.  My previous arracacha lived its whole life in the greenhouse, an unhappy looking plant the entire time, and slowly faded away over the first half of winter, leaving no trace.  I bought a plant from Sacred Succulents this year, giddily potted it up in a bucket, set it in partial shade to adjust, and watched all of its foliage turn yellow and develop brown spots over a few days.

I was philosophical about it.  While I expect mauka should be successful in this climate, arracacha is probably a stretch.  It is grown a bit lower in elevation in the Andes and expects probably a bit more warmth.  It will be hard to get it through the winter.  My best hope of growing arracacha is to get a plant to set seed and try to grow some better adapted plants (or perhaps to find a Peruvian variety, but those seem to be quite difficult to locate).

Anyway, looking at a probable loss of the plant, I cut it back to nothing but the cormel and shrugged my shoulders.  When a plant looks that sick, you might as well try desperation tactics.  To my surprise, it grew anew and the yellowing affliction has not reappeared.  This is a Puerto Rican variety, whereas the variety that I previously killed was probably Brazilian.  I really have no idea what the climate conditions are like where arracacha is grown in Puerto Rico, but I imagine it is probably fairly warm compared to this cool maritime climate.  I will continue to cross my fingers, but I'm not counting on bringing in an arracacha harvest any time soon.

If you now know mauka from maca but aren't so sure about arracacha, here is a good article that covers the basics.

There is a lot more growing, not the least of which are weeds that must be pulled before the jungle closes in on us.  So, until next time, good growing!

Monday, April 28, 2014

Sea kale: What's that sticking out of your thong?

Sea kale has a rather remarkable ability to re-sprout from its roots.  One of the popular ways of propagating sea kale relies on this tendency.  Sea kale root cuttings, or "thongs" as they are commonly described, provide a fast-growing clone of the parent plant without all the fuss of starting from seed.

Until recently, I wasn't sure of exactly what goes on with those thongs under cover of soil.  I knew that planting them a few inches deep would eventually result in a new plant, but what happened between planting and plant was a mystery.  To satisfy my curiosity, I dug a few up this year and all looked much like this:

Sea kale thong dug in late April
In every case, the segment of root has swelled at the thicker end and sprouted.  You can see that this one is just getting started.

In the past, I have planted thongs about two inches deep without much regard for orientation.  Now that I have seen that sprouting happens very near the cut at the thicker end, I think that it makes sense to try to plant the thongs more vertically, with the thicker or cut end near the surface.  Really, this shouldn't be much of a surprise, since that is how the root was positioned on the original plant.

At any rate, this should serve to bolster the confidence of those of you who have expressed doubts about your rather dull-looking packet of sea kale thongs.

Wednesday, April 23, 2014

Oca: Seedling transplant perils

We have over 1000 oca seedlings this year and have been planting them out for several weeks.  Most are about six inches tall with a single stem, but there is quite a considerable range from spindly plants that are still less than two inches tall to 18 inch tall multiple stemmed plants.  We can't manage this many plants indoors, so the plan has been to grow them in the field.

So far, we have hardened off plants for two weeks, then planted them out in the field slightly deeper than they were in the pots.  This has generally worked well for oca from tubers and cuttings.  It appeared to work well for the seedlings at first.  Then the wind arrived.

We live on the coast, where our dominant spring/summer weather pattern is created by inland warming.  As inland air rises, cool winds flow inland off the Pacific.  That means that any time the weather is clear inland, we have winds that start in mid-morning and last until evening.  They are not particularly strong, but may blow at 15-20MPH all day long.

Oca is generally pretty resistant to wind, although it sometimes causes us problems with flowers and seed pods dropping.  So, I wasn't too worried when we got a week of such winds after planting out the first bed of oca seedlings.  That was a mistake.

Oca seedlings have much less tolerance to wind than tuber-grown plants or even minimally rooted cuttings.  Two days of steady wind badly damaged about 400 seedlings, to the point that I am not sure that more than a small percentage of them will survive.  That is certainly a disappointment, although the sting is relieved somewhat by remaining seedlings that had not been planted out and the promise of more seed this year.

The question is: what should I try next?  I could harden the seedlings off for a longer period, although  moving a thousand plants in trays in and out of shelter every day for weeks may not be possible.  I'm considering two possibilities and may try both.

The first option is to wait and start seeds outdoors.  We get volunteer seedlings here, so I know it is possible.  They are very slow growing compared to plants started indoors, but they also are able to withstand the rigors of the environment.  So, perhaps starting batches of seed outdoors in April may provide hardy, transplant-ready plants by July or August.  There is definitely an appeal to the labor savings.

The second option is to grow the seedling generation entirely indoors.  Even a plant only four inches tall is capable of making mini-tubers.  I could start plants indoors and put them in total darkness once they reach a certain size.  The mini-tubers could then be planted out to the field with the greater hardiness of tuber-grown plants.  The additional advantage is that multiple tubers will often be produced, which provides some backup.  I think four months should be sufficient to produce mini-tubers.

Of course, neither option solves the immediate problem, which is the remaining seedlings that I have to plant out this year.  As I am limited by time (although our success with the store has raised the possibility of doing this full time eventually, it remains a hobby for now) I'm just going to have to harden them off a little more and keep planting them out.  Some will survive, but more will not.  I think that I will try planting them deeper and at an angle in order to put more of the stem underground.  This will reduce wind exposure and provide more opportunities for the plants to resprout if the topgrowth is damaged.

On the up side, this year will definitely select for hardier varieties.  Even if only a tenth of the plants survive, that is a lot of new oca to evaluate and the education that I've received in raising so many plants to this point will certainly pay off in future years.

Hopefully some of you who are growing from seed this year will read this before you repeat my mistake.  I would be interested to hear what you try and how it works for you.

Tuesday, March 25, 2014

Oca: Mini tubers from 2013 seedlings

A side note to start with: Northwest Edible Life invited me to do an introduction to Andean root vegetables, which has gone up today:

I'm really pleased with how it turned out.  Northwest Edible Life is a great blog and definitely worth checking out if you're reading here.


This will be just a quick post to give you a peek at some tubers from 2013 August-sown oca seedlings.  These plants, for whatever reason (ahem... neglect), died over the winter and I just got around to digging them.  The tubers will be replanted shortly (very shortly in the cases of those that are already sprouting).

It is hard to draw any conclusions from mini tubers about the ultimate size and yield of an adult plant, but they do give a preview of the tuber colors at least.  These were all out of the Hopin x Sunset cross and their parentage is obvious in some cases; the top right and top left look a lot like Sunset, while the second from top right looks a lot like Hopin.  White tubers seem common and I imagine this is probably the color of the tubers of oca's ancestors; a lot of tuber-forming Oxalis appear to have white tubers.

Some mini tubers from 2013 oca seedlings

Wednesday, March 5, 2014

Oca: Lifespan of cuttings

We overwintered quite a few seedling oca plants this year.  These plants were started from seed in August and only a few produced tubers before the end of the season.  I wasn't sure what to expect - if the plants would form tubers indoors, if they would grow through the winter, or if they would just suddently hit their expiration date and die off.  Our results indicate a combination of the above.

Fall cutting from an oca plant that formed tubers
Notice that the main stem has died and that it
has formed no roots.
Cuttings that we took from plants that had not yet set tubers grew through the winter.  They have rooted and appear to be growing like any other oca plant.  I've been planting them out this week and I anticipate that they will grow through the season like any plant started at this time of year.

The cuttings that we took from plants that had formed tubers are the more interesting case.  Until recently, they looked as healthy as the other group of cuttings and they continued to grow indoors through the winter.  About a month ago, I noticed that many of them were forming aerial tubers.  The new growth branches began to thicken.  Over the past couple of weeks, it has become obvious that the old stems are dying.  On pulling them from the soil, the cause is clear: none of them formed roots.

The division between the two groups is very clear.  Every cutting from the plants that had not formed tubers has well developed roots.  Every cutting from the plants that had formed tubers has no roots.  In addition, the stumps of the original plants, which obviously had roots, died back as well.  So, it is not just a failure to form new roots; the roots on the original plants appear to die off after tuber formation.

The new growth thickens into aerial tubers
These can be plucked from the dying stem or
will eventually fall off on their own.
Each can be planted and forms a new plant.
So, it appears to me that oca is only perennial through the formation of new plants from tubers.  Even in a mild climate where the plant is not killed by frost, the mother plant eventually dies off to make room for its tubers to start the next generation.  The sample size is pretty small here and this doesn't remotely resemble a controlled experiement, so there may be other factors at work, but the limited evidence seems to point in a pretty clear direction.

This is an important consideration for the oca propagator.  Plants can probably be continued from cuttings indefinitely as long as the plant is not allowed to form tubers.  Cuttings taken from a plant that has formed tubers will be an intermediate step toward the production of sprouted aerial tubers which can be separated and replanted.

This may also provide interesting fodder for future experiments.  It appears that oca produces a chemical signal that probably both starts tuber formation and triggers senescense in the existing growth.  Either the signal persists for such a short time that the new growth (tubers) is not affected by it or the new growth produces a chemical that counters the signal.  The safe assumption is that this signal is produced when daylegth falls below a certain threshold, although it appears that it also requires plants of a certain size before this can happen.  That may be an indication that the new growth produces a countering signal and that the tuberization signal may actually be a long-lived event.  Is it possible to force a plant to produce this chemical signal by moving it into dark conditions and then transfer the chemical, perhaps by sap inoculation, into another plant to force tuberization?  If I have time to play around this year, I will do a few experiments.