Monday, April 13, 2015

Mauka: Going to seed

I was lucky to get a few seeds of the mauka (Mirabilis expansa) variety Blanco last year.  They grew very well, although a fairly late start meant that they didn't produce substantial roots.  I had two immediate goals for these plants: to keep them alive over the winter and to get them to produce a seed crop.

The first goal didn't seem like a big problem.  Mauka foliage is killed by frost, but the roots have survived the winter in climates similar to ours.  The second goal was the more challenging proposition.  Most Andean root vegetables require short days for root production.  Mauka appears to grow its storage roots under any day length, but flowers only under short day conditions and possibly only following short days of increasing length.  Short days of increasing length occur in late winter and early spring, which is generally not the best time of year for frost-sensitive plants to flower.

With this in mind, I transplanted four mauka plants into the greenhouse to overwinter with foliage intact.  This went surprisingly smoothly.  Most plants look pretty bad after a winter spent in our chilly, drippy, insect-infested greenhouse, but mauka seemed not the least bit bothered and happily grew into an unruly tangle of five foot long stems.

In late February, we began to see flower buds on all of the plants:

Mauka flower buds
By the second week of March, flower buds had disappeared from all but one of the plants, but it was beginning to open flowers.  Initially, flowers were difficult to spot, because they only opened in the early morning - sometime between 1AM and 6AM.

Mauka flower opening at 1:30AM
But then things started to get weird.  The flowers remained open a little later each day.  I'm not sure if they opened any later, because I had given up on wandering out to the greenhouse at 3AM to look for flowers by this point.  By the end of March, flowers were staying open until noon.  I'm not sure what to make of this.  Maybe this is a totally normal behavior for mauka or perhaps the increasing day length is confusing it.

Mauka flower open at mid-day
Mauka flowers don't remain open for long - a day or two at the most.  They then sink down into a little doughnut shape.  In about a week, you can tell if the flower was successfully pollinated because the calyx (the outer green leaves surrounding the flower) begins to lengthen.  Our results look very good so far, with more than 80% of flowers going on to lengthen into anthocarps (a single-seeded fruit contained by a calyx).

Finished mauka flower
The anthocarp at the top of the following picture is getting close to maturity.  You can see that it is significantly longer than the others.

Mauka anthocarp at full size

A couple of the anthocarps are now starting to yellow a bit, which I understand is a sign that they are nearly mature.  Based on this I expect to begin harvesting seed in a week or two.  There is no sign of this mauka plant slowing down yet, and there are about 300 buds, flowers, or anthocarps at this point, so I am optimistic about getting a good seed crop.  It will be a fussy job though; the inflorescenses are fragile and sticky and located throughout the tangled mass of mauka stems.  Casualties are to be expected.
Flowers on the left, none on the right.  These are two different plants entagled with each other, but only the left one is flowering.

I am puzzled that only one of the plants continued flowering long enough to produce seed.  All four plants were kept under the same conditions.  They are all about the same size.  They all started flowering at about the same time.  Three just stopped.  So, is there something special about this particular plant?  They were all grown from seed, so there is the possibility of some differences in genetics from one to another.  Is this a superior variety for seed production, less affected by increasing day length, or was it just slightly closer to the heater or placed under a convenient drip (or not placed under said drip)?  There's no way to know at this point, but I will be keeping track of cuttings from this plant in order to see if the same thing happens next year.




Thursday, April 9, 2015

Oca: Crunching the numbers on 2 years of oca from seed

Last week, I answered the most frequently asked question that I receive about oca.  This week, I am taking on the second most common question: What are the odds that I will get a variety worth keeping from oca seed?

Oca, as an octaploid, is a fundamentally heterozygous plant.  Every oca seed produces a different result and the variability is considerable, even when all the seeds come from the same cross.

Here is a sampling of varieties obtained from the Hopin x Mexican Red cross:


See what I mean?  A couple of varieties look a lot like Hopin.  None of them look much like Mexican Red.  And most of them look nothing like either parent.  Yield ranges from 99 to 854 grams.  Plant characteristics and flavor vary to a similar degree.

Without even crunching any numbers, you can see that there are a few in this group that are probably keepers and a few that should be discarded.

Yield is one of the more important criteria in the evaluation of new varieties.  Of course, they should also taste good, flower well, and have good tolerance to different environmental conditions, but if the yield isn't good, none of the other characteristics are particularly compelling.  I took the yields for the 41 heirloom varieties that we grow and divided them into percentiles.  The 95th percentile for yield among heirlooms was 911 grams (almost exactly 2 pounds) and contained 3 varieties: Sunset, Hopin, and White.  This seems like a good goal, although I hope that we can eventually produce varieties that produce 2kg or more.

So, what are the odds that a new variety from seed will have a top yield?
Out of 601 varieties, 5 reached the 95th percentile of heirloom yield (two of those significantly exceeding it).  That's 0.83%.  That may not sound encouraging, but it really isn't bad at all.  1 out of 120 seedlings performed as well as the best heirlooms.  The 60th percentile was 433 grams, which, for the metrically challenged, is close to a pound.  So, about 1 in 3 seedlings yielded 1 pound or more per plant.  If you also want a tuber of a certain color, excellent flavor, and abundant flowering, you will exclude a lot of these plants, but the odds that a home gardener growing a few dozen plants will turn up several keepers seems very good.

Monday, April 6, 2015

Know Your Andean Root Vegetables: Part 2

Here is the second of our Andean root vegetable fliers.  This one has the lesser known crops, at least from a North American perspective.  There is a link to the full two-page version below.


Thursday, March 26, 2015

Oca: What does it taste like?

Sunset oca tuber interior
Sunset has low acid, medium sweetness,
and medium density, traits that combine to
produce a flavor a bit like carrot.
I get asked a lot of questions about oca, but one is asked more commonly than all of the others combined: how does it taste?  Most people don't want a long answer, so it is easy to fall back on the standard descriptions like "lemony potato" or "potato with sour cream."  These answers satisfy those who aren't planning to grow oca and just want a general idea of how it can be used, but they fall far short of exploring the spectrum of flavors and textures offered by oca.

Bolivian Red oca tuber interior
Bolivian Red has low acid, medium
sweetness, and high density, traits
that combine to produce a flavor very
much like winter squash.
I've tasted a lot of oca, particularly through this winter as we evaluated several hundred new varieties grown last year.  I think that there are three important components of oca flavor: sweetness, acidity, and density.  These combine to produce very different results.  A variety with high sweetness, high acidity, and low density will be sharp and crisp, evoking comparisons to pie apples.  A variety with medium sweetness, low acidity, and low density may taste a lot like carrot.  A variety with low sweetness, low acidity, and medium density can be quite similar to a potato, particularly if it also has some bitterness.  A variety with medium sweetness, low acidity, and high density can be a lot like winter squash, with some developing strongly nutty flavors.

Hopin oca tuber interior
Hopin is medium acid, low sweetness,
medium density, traits that combine to
produce the classic "lemony potato," or
"lemony turnip" flavor.
The matter is complicated further by the change that oca undergoes after harvest.  When exposed to sunlight for a couple of weeks (or when simply left in storage for about a month), oca undergoes biochemical changes that reduce acidity and, to a lesser degree, increase sweetness.  So, a variety that comes out of the ground powerfully sour may be just pleasantly lemony in about a month, while a variety that is slightly sour may lose that character entirely.  As a consequence, some varieties are best when freshly harvested, some after a period of storage, and others have great (but different) flavors before and after.

Redshift oca tuber interior
Redshift has low acid, low sweetness,
 medium density, and a slight bitterness,
 traits that combine to produce a result
very similar to potato.
All of these combinations have their charms, but the range can also be a challenge; different oca varieties aren't necessarily interchangeable.  A really acidic variety of oca (White, for example) can be used in a German style potato salad without adding any vinegar.  It is a pretty unusual substitute for mashed potato and the acidity doesn't pair well with every dish.  The higher acid ocas are a somewhat unfamiliar combination of flavors compared to many common vegetables (sometimes, they are more akin to fruits) and so require a little more planning to incorporate into recipes.  On the other hand, the low acid ocas are relatively versatile and slot right into recipes calling for ingredients as diverse as potato, carrot, sweet potato, and squash.

So, what does oca taste like?  That question is about as easy to answer as "what does wine taste like?"  I have made some generalizations, but the spectrum of flavors in oca is wide and I feel like I have only just begun to explore it.  New seedlings routinely turn up new combinations of flavor that I haven't encountered before.

Saturday, March 21, 2015

Know Your Andean Root Vegetables: Part 1

We have been very lucky to find an artist who can translate my crude scribbling into a beautiful finished product.  Here is a handy introduction to the major Andean roots.  There is also a two page flier with the same information attached below.


Click here for the two-page PDF.

Friday, March 20, 2015

Mashua and Ulluco: Rainy day seedlings

Mashua (Tropaeolum tuberosum) is one of the few Andean root crops that is reportedly easy to grow from seed.  It is happy self- or cross-pollinating and produces lots of flowers which grow into clusters of 3 to 4 seeds.  The seeds are regarded as easy to germinate.  They come up sometimes as volunteers in mild climates.  They are said to have low or no dormancy.  You won't find much more information than that on the subject.  It's "easy," but apparently not many people haven't really tried it or written much about it if they did.

I agree that it is easy to get mashua to start making seeds.  My plants have done a lot of that.  The problem is getting them to mature.  Mashua starts making seeds relatively late in the year and they take quite a while to mature.  Around here, they need to make it to about mid-December.  The plants are frost sensitive, and several times now I have almost made it to mature seed, only to be surprised by a frost that wiped them all out at the last minute.

Last year, I spent quite a bit of time making crosses, planning to skin a hoop house over the plants before cold weather arrived.  Unfortunately, we were surprised in November by a hard freeze on a night with a predicted low of 40 degrees and that was the end of most of my mashua seed crop.  I had picked a few dozen seeds early as a bit of insurance.  I wasn't sure whether they were sufficiently mature, but I sowed them and a few dozen more received in trade back in December.  And waited.

By the time March came around, I started to despair that my seeds were not going to germinate.  I put the tray on the lowest shelf of the greenhouse, where seedling trays go to die.  A couple weeks ago, I saw a mashua seed pushed up from the surface of the soil and pulled it out to find a radicle emerging.    More careful examination revealed 5 more, so it looks like I am going to have some new mashuas this year.


It still took quite some time to see any growth.  Two weeks later, the first bit of green appeared.  Mashua cotyledons remain below ground, so it takes the plant a while to form a true stem and leaves. By the time a mashua seeding breaks the surface, it has probably been growing for 2 or 3 weeks.


This mashua, a Ken Aslet x Orange cross, is the farthest along.  It looks happy enough.  I have high hopes for mashua breeding.  At minimum, I would like to see a mashua with the size and color of Orange, but the early flowering and tuber formation of Ken Aslet.  Even that would be just an intermediate step, because my mashua breeding project has one simple, overarching goal: improved flavor.

In other news, ulluco seedling #5 arrived today:


This makes two with red cotyledons and two with green (plus a volunteer, whose cotyledons I never saw).  It will be very interesting to see if seedling color is reflected in tuber color to any degree.

I've been keeping the ulluco seed flats very wet and heated to 70 degrees during the daytime and 55 at night.  I don't know if that is the best approach, but it seems to work.  Because wet soil is better for germinating hard coated seeds than it is for keeping seedlings alive, I remove the seedlings to individual pots as soon as I spot them.

Monday, March 16, 2015

Ulluco: If you're not bored yet, I'll keep working on it

Well, it appears that we have cracked the ulluco code.  Since our first volunteer seedling last year, we have now sprouted three more seeds under controlled conditions.  I couldn't be more pleased.  In addition to being one of my favorite plants, I've spent more time in the pursuit of ulluco seed than anything else that I've ever grown.  A quick and dirty calculation revealed that each seed that has sprouted so far is the result of nearly 110 hours of work.  Hopefully more are still to come, which would make that number look a bit less obsessive.  ;)

Here is the state of ulluco in the last days of winter, 2015:


My new ullucos tray is filling up.  I'm multiplying plants in preparation for planting out sometime in April.  There's no particular rush to get the ulluco in the ground, which gives me some time to make more plants and hopefully increase the yield.


These pots contain OC-14-CB4x01, our volunteer seedling from last year.  It produced only one tuber, which has been reluctant to sprout.  I was able to coax a single sprout and have since split that plant four ways.  I'm trying to get two more sprouts from the two halves of the tuber at the right of the photo.  I'd like to have a dozen plants of this variety by the end of April.


OC-14-CB4x01 looks very similar to its parent variety, BK10425.2, but can be easily distinguished by the very red undersides of its leaves.


Some ulluco varieties exhibit very strong apical dominance in the tubers, meaning that they tend to form a single sprout and will not form others without stress or increased physiological age.  This variety appears to be a good example of that.  I eventually cut the tuber and treated with gibberellic acid in order to encourage sprouting from lateral eyes.  Now we're starting to see a little sprouting.


These pots contain OC-15-CB2x02, our first variety that germinated under controlled conditions.  I took a cutting last week, which established easily and has already doubled in size.  I hope to have four plants by the time that we plant them out in April.


In comparison to OC-14-CB4x01, this variety has very pale green leaves, top and bottom.  Interestingly, the cotyledons are spotted with red underneath, but the true leaves are not.  I have a guess that the tubers of this variety will also be blotched with red.


Here is a first look at OC-15-GF6x01, which appeared yesterday. It has red cotyledons, which is quite a contrast to the completely green cotyledons of OC-15-CB2x02.  It has been freshly transplanted to a new pot from its algae and insect-larvae infested flat.


And here is OC-15-GF6x02, which also appeared yesterday.  It hasn't yet lost its shell.  Also freshly transplanted and taken inside to be coddled with the rest.

There is no guarantee that any of these plants will make to to fall, or that any of them will produce tubers even if they do.  There is really not much that we know about the probabilities with ulluco.  Of course, that is what makes this so much fun.  Nature doesn't give up her secrets all at once.