Turning Wild Grasses into Grain Crops

Here's an idea that's been floating around my head. If (most) grain crops were selected from wild grasses to become one of the largest staple food of humanity, is it possible that modern humans might be able to do the same? If so, how would one go about it? More importantly, is it a good idea, and what are the risks of trying it?

Now that our summer drought has embraced our landscape with unusual enthusiasm, I watch with sorrow that the plants my well cannot support wither and dry. Yet there are weeds that sprout and thrive without a drop of rain. Why can't the plants I want to grow do the same? I've been working with plant breeding and experimenting with different planting times to take advantage of the winter rains, but still I'm taunted by wild green growing things that seem to thrive no matter what the weather throws at it.

Not long ago I was reading Carol Deppe's book Breed Your Own Vegetables, and about the time I got to the chapter that discusses the uses wild plants have for possible improvement of vegetable crops, I fell in love with a grass.



Echinochloa crus-galli , sometimes called Barnyard Grass and a close relative of Japanese Millet is growing in my garden with great gusto. When hemmed in by other plants, it out grows amaranth. Apparently, it has been used as a food crop at different times in history, in many different parts of the world. An Old World grass, there are rumors that it was a staple food of the Norse and Vikings, and has been harvested in times of famine and as peasant food throughout Europe, Asia and Africa. It can be used as a fodder crop for livestock, provided that chemical nitrogen hasn't been applied. I think this grass is amazing.

The trouble with Crusgalli is that it is considered an invasive species and can have a strong negative impact on industrial style agriculture, reducing yields by as much as 20%. It grows in many conditions, from flooded to dry sand, and seems to adapt quickly to herbicides. Any breeding experiments I do with this plant would need to be kept isolated to prevent the plant from becoming more invasive than it already is.

I talked a bit more about my thoughts on this grass in the landrace thread.


My first step is to identify and observe the potential candidate as it grows in the place it choose to live. Then gather seeds from the grasses that have the traits I think would be most useful for my project.

But then... what's next? Just grow and select, grow and select? I suppose that's how our ancestors created grain crops. Although, I wonder if we have any modern knowledge that we could use to make this process a bit more efficient?

I found this passage in Lost Crop of Africa,Vol 1, grains that suggest a method of developing wild grasses into food crops.

-Direct seeding trials using rain as the sole source of moisture;
-Searches for elite specimens (those that, for instance, hold onto the ripe seed, that have bigger seed, and that best survive harsh conditions);
-Trials on various sites (from the most favorable locations to moving sand dunes);
-Analyses of food value (physical, chemical, and nutritional) as well as of the foods prepared from them; and
-Multiplication of seeds or other planting materials for distribution to nomads, farmers, governments, and researchers.

Reading more in this book makes me feel that this project might be promising. It is certainly worth learning more about over the winter.


Anyone else have any interest in this sort of thing?

I am interested R ! I will be watching this thread to see how it goes.

I am always daydreaming about domesticating wild crops. I am working on a few domestication projects... One of my chief impediments is ignorance... I don't know what conditions the plant needs. I don't know it's life-cycle. Researching the data for even simple things like the name of the plant is nearly impossible for me. It took me over a year to find the name of a wild cabbage that I am working with:

Caulanthus crassicaulis


The most powerful plant breeding method that is readily available to me and easy to implement is "sibling group" plantings... That is save the seeds from one mother into the same packet of seed, and plant them together in a short row, in the same field as the other sibling groups. And then save seeds each year from the sibling groups that grew best. It's the next better thing than "recurrent mass selection" which is implemented by planting random bulk seed into a field and then saving bulk seed of whatever grows best. By selecting whole family groups instead of single plants I get a better idea of the true strengths and weaknesses of each family rather than of individuals. For example, one year I was growing popcorn, and every member of one sibling group fell over in a wind storm. Most sibling groups weren't affected at all. It was super easy to decide to cull that family group.

Professional plant breeders do randomized duplicate plantings to compensate for things like differences in soil or climate across a field. That's too complicated for me. I just plant the sibling groups in short rows and then save seed from the families that thrive for me, and cull the families that don't.

My great-great grandfather developed a variety of wheat that eventually went on to become the most widely planted wheat in southern Idaho and northern Utah. The family story is that one day he noticed one head of wheat in a field that did much better than any other plant in the field. He harvested it separately and grew it in a garden plot for a couple years until the seed was multiplied enough to share. It still grows great in this area. I have a crop ready to be harvested this week. The wheat acquired the family name, and I am still benefiting as a farmer because of it: "Those Lofthouses really know how to grow things."

RR, i think it's admirable that you are being so considerate before experimenting with a species that is considered invasive in some parts of world.
(found this factsheet by the way: http://www.hear.org/pier/species/echinochloa_crus_galli.htm -- the only area that notes it's native is Hong Kong. interesting!)
but since you live on an island in a non-tropical climate, wouldn't this risk be diminished in your case? are there any grain farmers there?

and as you so eloquently said in this thread:
https://permies.com/t/27470/plants/Breed-Vegetable-Varieties-book-Carol
"There is a fine line between vigour and invasiveness. Though I wonder if it's more a management issue than actual plant problem.""

i/o/w, let it be your guide along the way, but don't let the cautiousness get in the way of the experiment.

still think your idea of crossing crus galli with japanese millet is excellent. if i may introduce a corollary to your theorem : "There is a fine line between nutrition & edibility. Though I wonder if it's more a matter of personal preference shaped by cultural conditioning."

perhaps there is a sweet spot between the two species where vigour & edibility are ideal?

R Ranson wrote:Here's an idea that's been floating around my head. If (most) grain crops were selected from wild grasses to become one of the largest staple food of humanity, is it possible that modern humans might be able to do the same? If so, how would one go about it? More importantly, is it a good idea, and what are the risks of trying it?

A risk benefit analysis? Hmm.

A big risk is that you may spend years working on a project only to have a militarized police force raid your garden and declare your seeds to be a public nuisance and destroy the project. We've run into that sort of interference a lot!!! It can be mitigated by spreading the seed around widely. Maintain feral patches. Grow in multiple locations. Collaborate with other farmers in other jurisdictions. Stash seeds in unlikely locations, or with friends or family. Mail a few small lots of seed instead of one large package. We're getting pretty clever about not talking about our plant breeding projects in public places on-line: Facebook and eBay are not private!!!. We're encrypting our eMails to keep our plant breeding projects more private. We exchange paper letters with each other without return-addresses on the outside. I personally don't keep records regarding who sends me seeds, and who I send seeds to. Records that don't exist can't be seized. The good news is that the plant-police are dead-broke, more broke I suppose than every other bankruptcy in the history of the world combined... It costs thousands of dollars to send an agent out to seize a plant or a packet of seeds. It costs tens of thousands of dollars to send a swat team. They simply lack the resources to do much more than make noise (and target an occasional individual as an example.)

There's the risk that you might spend a lifetime working on domesticating a crop and have it end up being about the same as what you started out with. I spent some time with a farmer in Idaho that is domesticating wild plants for landscaping purposes (I was way disappointed he isn't working on food, but whatever). He has made significant progress in selection in 5 to 7 years. Things like the size of the root or foliage may have a lot of leeway in them... But what if I choose to work with a poisonous plant, and can't discover non-poisonous offspring? The more plants you start out with, the more likely that something will be a bit different. That takes space.

It takes a lot of time to really look at a plant and to compare and contrast it with the other plants in the garden. And there is a real talent associated with looking for things that you are not looking for... As an example: What if I'm working on domesticating a spring cabbage to eat it's leaves, and I get so focused on big leaves that I fail to notice one plant that has a root like a turnip? Large edible roots could be really clever. Then I might end up with two projects instead of one. Then I have to consider isolation... Plant selection can take up a lot of garden space. I sometimes feel like I need to plant each plant as a specimen, and not crowded together like I'd do with production crops. Other times I don't mind way over-crowding if I'm only looking for one trait that is easily observed and isolated.

Many of the previously eaten species that I have tried eating are too bitter for my modern sensibilities. Domesticating those would require tasting a lot of too-bitter foods to find those that are less bitter.

With beans, I can weigh seeds, so that I can replant those with the biggest seeds in hope of eventually harvesting larger beans. My inexpensive equipment doesn't work that well with much smaller seeds, so I'd end up counting out a set number of seeds and weighing them. Or measuring with calipers or something. Collecting and replanting by sibling groups can really help with these sorts of selections.

I was at a latino-market yesterday and saw domesticated lambsquarters flower heads. I suppose that there are a number of species that were already semi-domesticated by the Indians that might be useful to explore again. Many of the seedy crops might be very useful as poultry forage, even if the seeds are too small or too husked to want to feed to my family.

Another trait that seems important to me to deal with when domesticating crops is to get rid of the intense germination inhibitors. Some species of plants are renowned for sitting in the ground for decades before germinating. That is great for a wild plant, but it really sucks for a farmer. Part of the domestication contract that we make with the plants is that they agree to give up their germination inhibitors and we agree to plant them at the appropriate time of year so that they can grow well. In my watermelon project I went a bit onto the wild side, so I had about 15% "hard seed". It didn't absorb water and germinate with the rest of the patch, but germinated weeks or months later. The last few years I put a lot of effort into eliminating that trait. I aught to do the same screening with okra and one variety of dry beans.

Hard seed in okra, germinating much later than the rest:

Here's an idea that's been floating around my head. If (most) grain crops were selected from wild grasses to become one of the largest staple food of humanity, is it possible that modern humans might be able to do the same? If so, how would one go about it? More importantly, is it a good idea, and what are the risks of trying it?

hau R.Ranson, There are several things you would need to decide upon before stepping into this project.
1) what are your goals for use of this "new grain"
2) which variants are already available or near enough to your answer for number one?
3) how many years are you willing to devote to this project? it will take around 5-7 years to get somewhere if you are not artificial pollenating.
4) grain development is usually done on acreage, do you have at least 1/4 acre to devote to this experiment so you have the ability to make many different crosses?
Or are you planning on just letting nature do it for you? the former will see faster progress than the latter.

I have several land race grains growing on Buzzard's Roost; Wheat, Oat and a variety of Barnyard Millet. Each of these are being bred for specific ends; the Wheat is going to end up as one of our flour grains, the Oat will end up as a stock feed and the millet is being bred for plump kernels for bird feed.
At the end of year one the grains are already showing a bit of plumpness but more importantly they are showing signs of increased hardiness and earlier maturity. I expect it will take at least seven more crosses to get these grains to the objectives of plump grains that mature earlier in august with lower gluten and higher protein contents than the commercial varieties we currently are using. I am using a land size of 0.5 acres for the test field for these grains and I use wax paper cones to isolate each head as I brush the pollen for each cross. Each of these cones carry the written information of the pollen parent, and a reference number that takes me to the correct data page on the computer for logging all the conditions during that cross pollination event. Record keeping is very important when developing any new variety, no matter if it is animal or vegetable, if you don't have the road map, you can't reproduce it accurately later.

Good luck with your development project.

(What we call barnyard millet has indeed been a staple cereal grain for over two thousand years, it also gave rise to grain sorghums as well as the millets. It is already nutritious and can be cooked whole, ground into flour it becomes the base for many of the Native American breads).

I find myself fully immersed in the mindset that pedigrees and paperwork are not required in order to obtain excellent progress with plant breeding. When I look at the myriads of domesticated crops that are available to me, it seems like the vast majority of them were developed by illiterate peasants. Not only illiterate, but many crops were domesticated before reading was invented.

In any case, labels in fields are notoriously fickle... They fall off, the kids move them, they fade, they get misapplied, animals destroy them. Batches of seed inadvertently get mixed together, or switched. There's just way too much chaos associated with my garden and lifestyle for me to be labeling things thinking that it has any real meaning. Back when I used to keep records regarding plant breeding, I found myself being more of a clerk than a farmer. That stopped immediately once I realized that it was happening.

These days my record keeping is almost exclusively a label on a seed packet which describes the traits of the mother plant: fig-leaved, yellow fruit, 10#, 19/20 kernels popped. etc...

I tend to encourage promiscuous pollination in everything, and I watch very carefully for naturally occurring hybrids in non-promiscuous crops. I'm not much interested in doing the labor necessary to make very many intentional hybrids. I'm especially not interesting in doing the labor to keep records. I don't have to know who's the daddy. If a plant grows well in my garden, it doesn't matter where it came from or how it's related to the rest of the world.

So these days, I tend to let each plant or each sibling group tell their own history. I can sometimes tell by looking at a squash plant that it is a descendant of XYZ variety that I grew 7 years ago. Other times, they are just squash that grow better or worse in my garden.

These are great thoughts. Thank you all for joining in the discussion.

I have some pretty specific desires for a grain crop - be it one bred from wild plants, or a domesticated grain adapted to my conditions.

In order of priority:
-it must be delicious!
-easy to process the grain in the home setting.
-can grow in our summer weather. Rain stops at the end of April, and starts in Mid Oct.
-plant early april or later, and harvest before the end of Sep.
-can still produce a crop in unusual weather years
-can be used to make beer or other alcohol
-is also edible to livestock
-has other uses - like straw from traditional grains can be used for weaving, baskets, sandals, and housing.

There are a few other things that would be nice to have which include:
-can grow and thrive Fukuoka style - No cultivation of the soil, no chemical fertilizer or prepared compost, no weeding by tillage or herbicides, and no dependence on chemicals.
-produces a staple crop that is easy to digest and low in fibre
-Is beautiful growing, in storage, and on the plate. Or at least conforms to my personal standards of beauty.
-is tall enough to compete with weeds.
-can be trampled in the first month of life without destroying the plant (to fit with my Fukuoka style fields).

I have no idea if Crus Galli is going to be the grass for me. At the moment, I'm just enjoying watching it grow and having thought experiments on how it could be useful. One of the things that got me interested in it, other than it's beauty, is that it is already a food crop in some parts of the world. Already food crop and already grows in my environment without irrigation or other human inputs! Looks like this grass is already halfway to fulfilling my desires for a summer grain crop.


I'm glad what Joseph wrote above about one of the biggest risks of plant breeding without the correct funding from the correct sources to protect you. It hasn't been a huge issue in Canada yet, but I think it will be soon as we've recently updated some federal laws to make it easier for large scale plant breeders to protect their copyright, and for the government to protect consumers by limiting what plant varieties are approved for sale. I'm trying to say this nicely - but basically it's legislation that scares the pants off me. More about Bill C-18 here.

Even before all that, the government had the ability to dictate what was a 'weed' and what was a 'food crop'. So, if I do move forward with this idea, I would need some solid support that the wild grass I'm starting with has a historical precedent as a food crop.


As for time, I have the rest of my life, but for space, I'll probably give it a couple hundred square feet for the first years. That's about the size of the area I can easily isolate and cleanse if it becomes necessary to move mid-project. If there is definite promise and deliciousness, then I'll give it a quarter acre. If things get that far.


I'm of two minds on record keeping. I do keep notes on overall observations and the more important traits, like dates. When I save seeds, I tend to do them in batches, saving seeds from the top 10% in one batch, then the next 20% in another batch, &c. I do label why I thought these were the best, and what promise they showed, &c. This year, I'm thinking of trying the family grouping idea to see what it gives me.

Labeling everything in the field, although a behaviour I admire, is not one that works well for me. I spend my attention looking at the labels rather than observing the plants. I like the old ways of doing things that focus more on observation than record keeping - if that makes any sense to you. The only time I am completely obsessive compulsive about labeling in situ, is when I'm working with very specific heritage or endangered crops - then I write every detail, if a moth sneezed on a leaf, I make note.

I really like your priority list, mine is very similar. I also agree with your and Joseph's assessment of record keeping. I could happily do my experiments with out any records but since I am part of the scientific world, old habits die hard. I do find that I have fewer setbacks by using the methodology I am using.
I can say that I have never had any labeling records go away and that is a 40 year track record, but that is probably because I use a three sided methodology, I label in the field as I cross, then transfer that to a map (piece of paper with lots of circles which represent groups of plants) of each field and also keep spreadsheets for each field. If a field label goes by the wayside, it doesn't matter. Yes it is a lot of trouble but I got into this habit while working both in the commercial seed industry and then in the USDA. One of my goals at the homestead is to release myself from this habit, but I know it probably will pop up time to time, since we are getting into breeding Dominique Chickens and American Blue Rabbits.

Mostly I now do the record keeping incase I come up with something that could be published, or used in the stories I write.

I strive to only work on items that make sense for homesteading with easy peasy as the main objective.
One should not create more work but strive instead to reduce the amount of work required for producing good tasting, nutritious foods.
I want to end up as non store dependent as possible.
Totally self sufficient would be the end goal, especially since it looks like the world is headed for economic collapse and rushing towards it.

Like you, time is the biggest investment I have on our homestead, and the more I can reduce what must be spent on any one project, the better I like it.

Millets make great beer, they can be malted too, but they don't have quite enough amylase on their own so you either have to have a barley malt or purchase some malt extract at the brewers store.
The landrace wheat I have growing wild on our property holds some promise but it will be this years harvest that tells that story.

while I don't consider myself Fukuoka Style, I am entrenched in the No Till methods.

Let me know how you get on with the grains, I am always interested in this kind of stuff.

An update:

I harvested a nice bucket full, but got frustrated because I couldn't get the hulls off. I ended up feeding the lot to the chickens. I still think this grass is gorgeous, it's also very popular with the livestock as grass and grain. The plan is, this year, to keep an eye out for any variation in the grass that might lend itself to become a human crop. I'll be keeping a close eye on it this year, and probably save a few seeds from the best plants for experimenting with next year.

Echinochloa srus-galli R Ranson, you might find some of this information helpful, if you haven't already read it.

It is possible to hull the seed but you would most likely need a hulling machine for this.

R Ranson wrote:I harvested a nice bucket full, but got frustrated because I couldn't get the hulls off.

My strategy for that would be to agitate the seeds until a few started losing their hulls, and then sort them and replant only seeds that have lost their hulls. As a breeding project, you might only need 1 in 1000 seeds to come out of their hulls easily. If repeated year after year, only replanting the seeds that slip out of their hulls easily, you may develop an easily hulled, or even a hull-less variety.

I do something similar with corn. If a cob is hard to shell, then I feed it to the chickens rather than replanting it.

I do something similar with wheat, rye, and beans. I thresh by hand. Any plant that is hard to thresh is culled.

Hello, R. Ranson,
    Sorry to resurrect an old thread. I don't know how your project has gone in the past three years, but I also thought about trying to domesticate a wild crop relative of a fully domesticated cereal grain. Setaria faberi (Bristly foxtail grass), a wild relative of foxtail millet (Setaria italica), grows widely where I live. I have already been able to collect 0.1oz of seed from wild plants. It has even been classified as a noxious weed in some states.
    The plant seems to be an ideal candidate for domestication. It produces large quantities of seeds; it grows in disturbed soil, and it can be easily winnowed when the husk has been freed from the seed by pouring it in a light breeze.
    The plant seems to have three main problems though. The bristles on the ears may splinter bare hands when rubbing the seed free; the husk must be rubbed free from the seed by hand or mechanically, and the seeds do not ripen all at the same time.
     The first priority in domestication would be selecting for more uniform seed ripening. Then, I would select for reduced or delayed dehiscence (shattering). Later priority traits would be increased yield, looser seed husks, and smaller bristles.
Below is a map from BONAP for the reported distribution of Setaria faberi in the US in 2014 and some images of the plant from Wikimedia commons.

An additional method that we have been using in recent years, is making interspecies hybrids between domesticated species and wild species. Then re-selecting for domestic-type traits such as hull-less, larger seeds or fruits, lack of germination inhibitors, etc...

I tried various different grasses and most can be used and eaten as grains as long as they have relatively big seeds that stay on the plant for a little while after maturity. In the images attached, you can see a grass I tried (not sure what grass it is, but I found it in my lawn here in WV). Annual grasses seem to work a lot better.

Here are my thoughts on some grasses I tried:
-Grass in images below - tastes like rice except worse in every aspect; texture is crunchy but also unpleasantly squishy at the same time
-Foxtail grass #1 - very hard and bad texture, tastes more like milk than milk does (I guess that's where the milk flavor comes from)
-Foxtail grass #2 (Setaria faberi from earlier in the thread) - seeds are small and hard to work with, but flavor is nice and the hull is edible
-Forest sedge - very good flavor, sizable seeds, but shatters immediately after maturity so I can't collect mature seeds; edible leaves
-Other forest grass - good flavor, shatters in about 8 hours and seeds mature at different times on the same plant so it can only be eaten young; flavor is similar to sweet corn; can't collect mature seeds
There were about 10 other species of grass that I tried, but I could not figure out how to make them into food.

I did not find anything worth saving the seeds of.

Unfortunately, I've not been able to continue growing out my wild foxtail grass (Setaria faberi) I originally collected due to having to move my garden and limited growing space. Nevertheless, I have continued my interest in domesticating wild grains. I have been especially interested in redomesticating maygrass (Phalaris caroliniana) and little barley (Hordeum pusillum).

Some information that might help in domesticating new crops from wild grains would be the processes that have been used to domesticate intermediate wheatgrass (Thinopyrum imtermedium) over the past forty years. So far, I have found the following two journal articles, but if anyone finds other sources that might help in domesticating wild grains, please feel free to share it here.
https://www.sciencedirect.com/science/article/pii/S1360138520300534

https://www.researchgate.net/publication/325062446_Development_and_Evolution_of_an_Intermediate_Wheatgrass_Domestication_Program

Hello, all I've been working on domesticating Native wild Rye species. Elymus Canadensis is one I've used for several years. Mostly testing its edibility. Elymus virginicus is another. I like it more than Canadensis. As it's easier to work with and has much larger seeds. The taste is decent. I think If I gather enough grain a sourdough could be possible. As the dough is a bit elastic when worked with.

Alex Delashmit wrote:Hello, all I've been working on domesticating Native wild Rye species. Elymus Canadensis is one I've used for several years. Mostly testing its edibility. Elymus virginicus is another. I like it more than Canadensis. As it's easier to work with and has much larger seeds. The taste is decent. I think If I gather enough grain a sourdough could be possible. As the dough is a bit elastic when worked with.

Earlier this year, I gathered some seeds from a wild rye species (Elymus sp.). I think it's Canadian wild rye (Elymus canadensis), but it could also be virginia wild rye (Elymus virginicus) or Elymus villosus. The problem with this species is that the chaff clings firmly to the seeds so it cannot be easily rubbed off with bare hands. I have yet to try dehulling the grains with a rubber disk huller or an impact huller due to the high cost required to buy or build one.

Ryan M Miller wrote:

Alex Delashmit wrote:Hello, all I've been working on domesticating Native wild Rye species. Elymus Canadensis is one I've used for several years. Mostly testing its edibility. Elymus virginicus is another. I like it more than Canadensis. As it's easier to work with and has much larger seeds. The taste is decent. I think If I gather enough grain a sourdough could be possible. As the dough is a bit elastic when worked with.

Earlier this year, I gathered some seeds from a wild rye species (Elymus sp.). I think it's Canadian wild rye (Elymus canadensis), but it could also be virginia wild rye (Elymus virginicus) or Elymus villosus. The problem with this species is that the chaff clings firmly to the seeds so it cannot be easily rubbed off with bare hands. I have yet to try dehulling the grains with a rubber disk huller or an impact huller due to the high cost required to buy or build one.

I have to peel the chaff off with my fingernails. Sometimes I find other smaller strains of Virginia wild rye where the chaff comes off easily.

Alex Delashmit wrote:
I have to peel the chaff off with my fingernails. Sometimes I find other smaller strains of Virginia wild rye where the chaff comes off easily.

If you're able to isolate strains of Elymus virginicus with loose chaff, you may eventually be able to domesticate this plant faster. I handled some einkorn grains earlier today and even these hulled grains are easier to clean than most strains of wild rye (Elymus ssp.) I've found. Let me know how your progress goes over the next few years.

I encountered a similar problem with sticky chaff when I attempted to clean some little barley (Hordeum pusillum) seeds before winter sowing them last week. The chaff for this species of barley seems to stick to the grains just as firmly as with most wild rye that I find. If anyone encounters a population of little barley with easily threshed grains, please let me know. The seeds ripen from April to May and the plant tends to grow less than one foot tall.

I noticed some strains growing near rivers or lakes have really nice shorter, fatter seeds and the chaff is easy to remove. Others I've found not near rivers have skinner bigger seeds with sticker chaff that latches on to the seeds. Interesting variations.