How to get from crops in the field…
…to food in the house.

Introduction

When you grow dry legumes and grains on a garden scale, you may run into challenges with finding the appropriate technology to grow and process your harvests efficiently. Since these crops are usually grown on a large scale, the standard equipment used for them in your area is probably not suitable for your garden, and may not be part of how you want to garden anyway. We have experimented with various tools and techniques for various crops, and want to share the combinations we’ve found helpful so far for the harvest end of things. However, we are still learning, and don’t feel that we have reached conclusions on how to best process all of these crops yet; so consider this notes from a work-in-progress, rather than the advice of experts.

For dry legumes (eg. beans, peas, lentils), grains (eg. wheat, millet, corn) and pseudocereals (eg. amaranth, buckwheat), the grower/eater needs to somehow separate the food part (seeds, in all these cases) from the non-food parts of the plant at some point between the garden and the kitchen. For most of these crops, people have divided this process into three stages:

  1. Harvesting involves picking or cutting some part of the plants that includes the seeds which are the actual crop, and securing them from weather and pests.
  2. Threshing involves breaking the seeds loose from the stalks, heads, pods, hulls, etc.
  3. Separating or cleaning involves sorting the seeds out from the other stuff. It usually uses three features of the crop to do this: size, shape, and density. Sorting by size and shape can be done using sieves. Sorting by density is usually done using wind in a process called winnowing.

In this article I will explain these three steps and describe tools we have used for them, and then finish with a few examples of how we get specific crops from the field to storage.

Every process can be thought of as a trade-off or balance between various priorities:

Harvesting a large enough quantity for what you need.

Our target crop size is enough to feed ourselves a complete diet, plus a small surplus for food security and seed sales. However, we are growing a significant diversity of crops, so within larger considerations (e.g. nutrition, crop rotation, soil health, relative risk of crop failure due to predation or disease, etc.), we can decrease one crop in favour of another.

In connection with that, not losing too much of the crop during processing.

I’m pretty careful in handling harvests that are small or rare (so every seed counts). For staple crops, every bit of loss has to be compensated by growing more, or, to look at it the other way, if you can get more food out of a particular area grown, you don’t have to grow as much. But we accept that some loss is fine if it makes the processing sufficiently more efficient in terms of time and effort to offset the loss.

Getting the crop clean enough and in good enough condition (e.g. few enough damaged or spoiled seeds) to use.

We are mostly using these crops to feed ourselves directly, so we clean them with that in mind. When we are going to use the crop for seed, we additionally try to sort out any genetic contamination (such as wheat that got into our rye seed).

Any technological preferences you might have.

We are working on figuring out how to live sustainably, so we are trying to build systems which do not use gas or electricity, can function without industrial supply chains, and use local or low-embodied-energy materials (mostly wood and scrap metal).

How much time and effort you are willing and able to spend processing your crop.

This is our primary work during the growing season, so we have relatively few scheduled things taking our time. But there are always too many things to do in the time we have! We are willing to work hard physically, but are not super-athletic types; like most people, we’ve got a collection of manageable chronic injuries and health problems.

How much time & effort or money you are able/willing to invest in building or buying tools.

We prefer to make our own tools whenever possible, since it is much easier to get what we want from a sustainability perspective, and we are fairly frugal (some have accused me of being parsimonious). This represents a significant investment of time, but frugality helps to make a self-reinforcing cycle where we have more time to do this work since we aren’t trying to earn more money.

Stating our family’s positions on these variables might be helpful, to give some understanding of why we have chosen the tools and techniques we have:

Harvesting

Dry Legumes

With all dry legumes, harvesting at the right time is crucial for good quality. Ideally the seeds are rattly in their pods, and the pods are crisp and brown or yellow. When you open a pod, seeds should barely dent when pressed with a thumbnail. Reality sometimes falls far below the ideal. Legumes at this stage are prone to two problems: if the pods dry too much, they “shatter” (open and drop their seeds in the dirt); and if it rains (especially if it rains a lot or stays damp and cloudy for a few days), the seeds can reabsorb water and shatter, sprout, or mold. Either way you lose. So we try to always harvest before they’ve over-dried, or a bit early if the forecast looks wet.

Large-seeded legumes (including common beans, peas, runner beans, and favas) are practical to harvest by hand. We tend to hand-pick all of them except peas. This enables harvesting when each pod is at ideal maturity even if the plant has pods at various stages. Our favourite varieties of dry pea (St. Hubert’s and Capucijner) mature evenly enough that we can take the whole plant at once, which is a lot faster. We simply cut the plants at the base with a sickle, untie their netting from the trellis, and roll the net and vines up into a big bundle, which we then put on a tarp to haul in from the garden. Grasspea (Lathyrus sativus) also works well with this method.

Small-seeded legumes are considerably more bother to harvest by picking. Chickpeas and lentils can be cut like peas. Unlike peas, neither is well-suited to our climate; it’s more of a guess when to harvest since there are likely to be some immature seeds and some pods which have begun to shatter.

After harvesting, we spread either pods or whole plants on tarps to dry further, usually storing them indoors for the night/during rain and putting them out in the sun during the day. Only when they are properly dry do we thresh them.

Even then, some seeds are often still too wet for storage, so they go back out on the tarps (watch out for chipmunks) or sit indoors in open containers or on screens near our woodstove in the fall, until they are thoroughly dry.

Small Grains

Wheat, rye, barley, oats, emmer, and einkorn all grow similarly and can be harvested similarly. If you have a very small plot (just preserving germplasm), you can harvest individual heads with scissors, but I tend to find this maddeningly slow. For small quantities I favour a sickle. Many people have used a sickle for somewhat larger food quantities too, but I have found the grain cradle (sometimes incorrectly called a cradle scythe) to be a brilliant solution that balances simple technology and greater efficiency for the worker. Cradles are, sadly, hard to find nowadays. If you are a fairly competent woodworker, I think it should be possible to build one using a scythe blade; however, I am not one (and Andre, who’s a bit better, hasn’t tried it yet), and my efforts to copy our old cradle have taught me one main thing: every part of the design is vital, and has already been revised to make it work as efficiently as can be done.

Theo’s method of cradling

(Incidentally, a note for all scythe-users: as far as we can tell, the motion for using a grain cradle is quite different from using an Austrian-style scythe. With the scythe, one cuts in a sort of horizontal arc, using the shifting of weight and tension in your legs, hips, and back to bring the scythe around on the ground. With our cradle, anyway, using this technique would be incredibly un-ergonomic. We know no one to teach us how to use it properly, but as far as we can figure out from trial and error, with the cradle the proper motion is more of a vertical pendulum motion: start horizontally with the blade to your right, then sort of flip your arms past each other – right under, left over for a right-handed setup. You will end up with a cradle of cut grain. Andre thinks you should then tidily drop the grain on the ground by rolling the cradle over. Theo prefers ending the stroke by catching the back of the upright of the cradle’s “rake” on the left thigh, then scooping the grain out of the cradle and dropping it behind him with the left hand while continuing to hold the cradle’s handle (nib) with the right. This is a lot more ergonomic than it sounds.)

To decide when to harvest small grains, choose a few random seed heads and thresh out a few kernels from each into your hand. Try to dent them with a thumbnail. You should harvest when the kernels are at the “hard dough” stage – dry enough to make it challenging to break them in half, but still dentable. If you wait until they are so dry that you cannot dent them, the heads will be more prone to shattering. As with legumes, watch the weather forecast, and (preferably) cut before rain.

Once you have cut your grain, you need to let it finish drying (until the kernels are too hard to dent) before threshing. Usually drying can be done in the field by making sheaves and stooks. The process is described below:

  • Collect a bundle of grain a bit larger than you can comfortably hold in one hand, and tuck it under your arm. This bundle is a sheaf.
  • If your straw is long enough and not too brittle, use it for binding. Pull out about a dozen of the longest straws from your sheaf and twist them to make a sort of rope; wrap it around the sheaf and tie an overhand knot, squeezing the sheaf tight with your arm; pull the binding as tight as you can; release it from under your arm and tuck one of the ends of your binding through again (to make a double overhand knot).
  • When you have made ten sheaves, stook them: carry them all to a reasonably level place in the grain stubble. Take two sheaves and lean them against each other at about a 50° angle (viewed from the side) with the grain heads at the top, jamming the bottom ends into the stubble to decrease their chances of tipping over. Take a second pair and lean them against the first pair at a 90° angle (viewed from the top) to the first sheaves. Repeat with two more pairs of sheaves on each of the corners. Then lay your last two sheaves on top as a sort of roof, fanning out the ends of the sheaves to protect those below from rain. The goal here is to make a little pile of grain which sheds rain reasonably well, but allows air circulation so that the grain can dry.
  • Leave your stooks for 1-2 weeks, until they are fully dry; if the forecast calls for a lot of rain, though, bring them under a roof sooner – if you have enough space indoors.

This process is admittedly a bit tricky; sometimes our stooks fall over on windy days. If that happens, we try to put them back up fairly promptly.

A stook of spelt
To keep small quantities of grain safe from small rodents, a small carnivore is just the thing

Corn

Corn ears are so large that it is practical to pick individual ears by hand when they are dry enough. We like to determine dryness in two ways: on most corn, the ear will be ready to harvest when it flops from an upright position to hanging down. (This sheds rain better. Clever, isn’t it?) However, some flour corns don’t seem to do this, and some flints won’t do it reliably (or maybe we’re just too impatient). For those, we harvest when the plant’s stalk turns brown and starts shriveling where the ear attaches.

It is also time to harvest if birds or chipmunks start ripping into the ends of your ears and eating them. If stored where it can get lots of air circulation, even quite wet corn can dry down successfully. If, for some reason, you have to harvest early, consider moving the corn inside and outside, as described for dry beans above.

We have never tried just leaving all the corn out until it is all dry, but this is how field corn is treated. I expect it would make for more time-efficient harvest, but also more loss to mould, sprouting, and animals.

We try to husk our corn within a day of picking, so it can dry better. We usually braid most of our corn and hang it up, both to make life harder for thieving rodents, and because we don’t have space to store it any other way.

To braid corn, leave the innermost husk or two attached to the ear. If you later find the braid is too bulky, which makes the husks run out too fast, leave only one husk attached to each ear. Cut off the shank (the stem which attached the cob to the plant) just above those husks, leaving as little possible since it is often a bit green and can mold.

Take two cobs, cross their husks, and wrap them around each other to make an overhand knot. Lay the husk from a third cob over top of these two, and begin braiding, the same as for hair.

I add another ear to the braid each time I bring the right-side strand back to the centre.

People who are really good at this can make long braids; I don’t usually go over about 30 cobs, because as the braid gets heavier it is more prone to coming apart (also, it is hard to attach a very heavy braid to whatever you are hanging it from if you have to hold it above your head while tying knots or hooking a twine onto something). When I am ready to end a braid I stop adding cobs, braid a few more turns, and tie a strip of husk around the braid to keep it from undoing itself. To hang it, we use a noose of twine; it is also possible to make a loop in the end of the braid and hang it on cornhusks alone, but I lack the skill to make a strong enough loop to reliably hold up all that weight.

We store the braids wherever we notice that there is a space between a ceiling and our heads which isn’t being used. I have hung corn on a frame built over my bed; we also use a frame in the space over the stairs in our house, and the rafters in our barn and workshop:

Cornhusks can be used to make various things, so as we husk we save out all the husks which are clean and free of mildew. We dry them for a few days by spreading them out on the floor, and then store them in sacks until we are ready to use them.

Threshing

Threshing, the task of getting the seeds detached from the pod or head or cob, is usually done in one of two ways: hitting (impact) or rubbing (friction). In this section I will organize by tool, rather than crop, and mention which tools have worked well for which crops for us.

Hand threshing is, arguably, a third or combination type of threshing. We generally only use our hands to shell dry legumes if there is a very small quantity of a variety, or if it is something very pretty which we want to spend time looking at! I also use my hands for rub threshing many vegetable seed crops (arugula, for example), and for catching individual pods or heads which are missed when using a more labour-efficient method. However, this sort of hand work feels too slow to be a mainstay for most staple crops.

Impact Threshing

Flail

This is a very old method of threshing. A flail is simply two pieces of wood and something to connect them together. Our flail is a wooden swipple (also called a swingle, the part that hits the crop being threshed),about as long as my forearm and 1 ¾” thick, which is attached with some twine and wire to a wooden handle about as tall as I am. To use it, you make a pile of material to be threshed on a clean tarp or an easily sweepable floor, and then hit it with the flail for a long time.

We like to use a circular motion with the swipple (approximately 90° to the handle, viewed from above), and not lift the far end of the handle above shoulder-height. Make sure people and other animals are out of reach and not in line with your swing (in case the swipple breaks loose from the handle). The flail’s great advantage is its simplicity and portability; its great disadvantage is that it only hits a relatively small part of the pile each time, so it takes a while to do a thorough job of threshing. It works fine for small grains and buckwheat, and we are experimenting with using it for red clover.

Bicycle Thresher

The bike thresher has a similar action to the flail, but uses a bicycle drive train to power a spinning shaft with a number of flails attached to it. This makes it much faster. The basic design for this is available online, thanks to Lu Yoder, who designed it and released the plans with a grant from the U.S. Department of Agriculture’s Sustainable Agriculture Research and Education (SARE) program.

Ours is somewhat modified, and we continue working on it to make it suit our needs; I expect the details would bore anyone not actively building their own. But if you happen to be in that situation, we would be glad to discuss what we changed!

We have mainly used our thresher for small grains and soybeans. With small grains, we simply open the sheaves, divide them into handfuls, and stick the head end of each handful through the front hatch of the thresher. Once the grain has been knocked off, we remove the straw and throw it on a pile. You have to be very careful when doing this: occasionally the flails will tangle in the straw, and pull it in. Let go rather than following it!

Other than that, it works great.

We also use it for soybeans. Tall plants tend to wrap, so we cut them into pieces about a foot long with long-handled pruners and then pour them in through the top hatches.

The thresher also works well for bean and pea pods that have been hand-picked. However, pea vines wrap badly.

With any crop, this design tends to accumulate threshed plant material in the threshing drum, so we need to stop relatively often to clean it out. While it does a fairly thorough job, there is probably a higher rate of unthreshed (lost) food in this plant material than with more labour-intensive methods, such as hand-shelling beans. It also requires two people to operate it (one pedaling and one feeding crops into it). Still, it is fairly quick and effective, so we generally use it for our largest-quantity threshing jobs.

Although we have not had accidents, the bike thresher is likely more dangerous than our other methods and should be used by alert people who understand how it works and where to be careful (but the risk is still more on the scale of bruises or broken appendages rather than the mangling and amputations that larger farm machinery has caused). We think that its advantages well outweigh the risk if you know what you’re doing.

Friction Threshing

Feet

The easiest way to thresh many crops is with your feet. Spread a tarp on a hard surface, or work on a surface that can be swept and can keep the seeds contained. I usually put my tarp on our porch; the ground can work, but it doesn’t seem quite as effective. Dump your crop on it, and then scuffle back and forth, scrubbing the seeds loose. This works well for peas on the vine and bean pods.

Threshing Box

The threshing box is a very simple tool which makes foot threshing more efficient. It works for all the same crops. It is just a wooden box, about 24” long by 16” wide and 12” high, with a few thin wooden slats attached to the inside bottom (ours are 1” wide and ¼” high, spaced 3 ½” apart, but sometimes I feel like they should be closer together or taller, so consider what you will be threshing if you build your own box). We have mostly switched from using a tarp to using the box; the slats make your feet more effective by catching the crop and increasing the friction, and the sides of the box nicely contain everything so that peas don’t go rolling away. We also can use it sitting down, or leaning with our hands on a wall, so that we can use both feet simultaneously instead of having to stand on one foot while “chicken-scratching” with the other. The box works well for amaranth, peas, beans, lentils, chickpeas, sorghum, and foxtail millet. It works alright for small grains, but it is a bit harder to clean them than with the bike thresher; more heads break off the straw and are too heavy to easily winnow out.

Thanks to Dan Jason at Salt Spring Seeds for this idea, and to Owen Bridge (Annapolis Seeds) and Rebecca Jehn (Rebecca’s Garden) for their helpful photographs of their boxes! This is a great tool.

Threshing Blocks

For small-seeded crops, we use a very scaled-down version of the threshing box, which we call threshing blocks. They are simply two pieces of wood (our arbitrary dimensions are 12”x6”x2” for the bottom and 4”x5”x1” for the top), with shallow saw cuts across them to increase abrasion. When the faces wear down too far, we just redo the saw cuts.

We put the seedheads on the bottom one and then scrub them with the top one. The seeds scatter somewhat. A cookie sheet under the blocks will catch most of them, but a tarp or sheet under that is still a good idea. The threshing blocks work well for smaller quantities of crops like flax and foxtail millet.

If you are concerned about not mixing up different seed lots, it is important to clean out the grooves on the blocks between batches. A wire brush works nicely for this.

Corn Shelling

One of the ways that corn is a special grain is how easy it is to thresh (per volume of grain).

You can simply use your hands, either by grabbing the cob with both hands and twisting opposite directions, or, if that doesn’t work, popping out a lengthwise row or two of kernels with your thumbs and then shelling the rest by pushing them toward the gap a row at a time.

That can get time-consuming and tiring on the hands if you eat corn as a big part of your diet or need to prepare a lot at once. We have found a simple one-piece corn sheller is a sufficient tool for the job, and it is elegantly simple to make: just a carefully shaped block of wood. We have instructions for how to make one, and we also sell them.

Separating/Cleaning

After threshing you will be left with a pile of seeds mixed with other plant parts – bits of stem and seedhead, pods, awns, etc. Separation is all about sorting them out. Sometimes the last bit of cleaning (such as for ergot, small bits of dirt, or split or moldy seeds) needs to be done by hand on a plate or cookie sheet. But this can become tedious, so we like to get it as clean as possible before moving to hand sorting. In most cases, you can get your seeds clean enough by sorting them en masse using two features: size and density. I will first explain the theory of this, and then talk about how we integrate that theory into our actual practice.

In theory…

Size

Sorting by size involves putting your crop through screens or sieves of various sizes. Ideally, you use two screens: one which is just big enough to let your seeds through, which catches anything larger than them (this is called scalping), and one which is just small enough to catch the seeds, which lets anything smaller than them through (this is called sifting). In this way you can end up with a pile of seeds and things about the same size as them.

We have collected lots of different screens for various crops; however, the ones we use most commonly are two made of hardware cloth (½” and ¼”) and an old aluminum colander which happens to have just the right size of holes for many crops, which I sneak out of the kitchen when it’s not being used for cooking. More accurate screens with holes of various shapes are made specifically for seed cleaning; these make it easier to sort seed by size as well as shape (for example, by using something with long narrow slots to clean rye, or something with round holes to clean peas). However, these can be quite expensive, so unless you are really into seed cleaning and know exactly what size of holes you want, I would recommend improvising with things from the thrift store or hardware store. A basket with regularly sized gaps can also work, and could be produced locally.

Density

Sorting by density works because seeds are denser (i.e. heavier for their size) than any other part of the plant – when it is dry. The standard way to sort them is by wind in a process called winnowing. To winnow, find a nice steady wind and pour your seeds needing cleaning from one container to another. The seeds will fall most vertically, while lighter debris (called chaff) will be blown farther away. By catching them carefully, you can collect your seeds and not the other stuff. With a strong wind or light seeds, you may also be able to sort out a third category of things heavier than seeds (rocks or unthreshed heads/pods) that falls upwind of the seeds.

It is possible to do this using the wind outdoors, if you happen to live in a place with very steady wind; however, our wind often is too gusty, which makes for spilled seeds. So we usually winnow indoors. A box fan works well for this.

It is easier to winnow seed if you use rectangular containers, because they allow you to pour a thin, even “curtain” of seeds, and catch accurately.

Back in the early 1900s, farmers used clever gadgets called fanning mills which combined both screens and a fan. If you can find one in working order (or repairable order), it can be a time-efficient tool that runs on muscles instead of electricity. We use an elderly Chatham fanning mill for larger quantities of dry beans, peas, and grain. A great feature of these tools is their serviceability: ours reached us without screens and various other pieces, but we were able to build replacements for most of the missing components out of wood, and used old combine screens obtained from a farm equipment dealership which salvages combines to build new screens.

…In Practice

When you actually try it, you will find that neither winnowing nor sieving works perfectly the first time. I usually go back and forth several times before I get my seeds clean enough to satisfy me, and occasionally have to use other tricks. Sieving is slower than winnowing, and the finer the screen, the slower it goes. I begin by considering how large the pieces I have to sort out are compared to the seed. If they are much larger (for example, pieces of millet seedhead mixed in with the seeds) I will run it through a large screen, which just catches those largest bits to quickly reduce the volume I have to work with. If they aren’t that big I might start by winnowing, since this can reduce the volume significantly. I then scalp with something only slightly larger than the seeds, and winnow again (often pouring the seed in front of the fan two or three times, or sieving it several times). Sometimes I sift, but often it is not necessary; most things smaller than the seed (except soil) blow away.

Two little tricks:

– Once you are done threshing, dump all your seeds and debris into a tarp or sheet. Gather up the corners and shake it. This makes the seeds settle to the bottom, and you can then usually skim off some of the plant material which you would otherwise have to remove during cleaning. If you are using a threshing box, you can also just do this in the box.

– For round seeds, it is sometimes possible to sort out non-round debris (bits of plant and broken seeds) by pouring the threshed mixture down an angled surface. However, you have to do this very carefully, or the seeds will carry debris along with them as they roll.

Some Examples

Chickpeas

Some chickpeas are indeterminate; that is, the seed matures unevenly. This is, in my opinion, very undesirable, so I am currently selecting for chickpeas which mature evenly. Those which do well can be easily pulled up when the seeds are rattly. I make them into bundles and pile them on a little tarp to carry in from the garden.

Chickpea pods are small; each holds one or two seeds. They are also oddly slippery, and cling to the seeds more than common beans or peas. All of this makes them a bit of a nuisance to thresh. To make them shatter more easily when I am threshing, I dry the plants for several days on a tarp in the sun. Then, late on the afternoon of a sunny day (so they are maximally dry), I trample them in the threshing box. I add a small bundle to the box, trample it, and add more until there are enough seeds and broken fragments of stem and pods and dust in the box to make the slats less effective. Then I shake everything into one corner of the box, skim off as much of the light debris on top of the pile as I can, and dump the rest into another container. I repeat this process until everything is threshed.

Most of the bits of chaff are about the same size as the chickpeas, so scalping isn’t really possible. However the seeds are so heavy, compared to the very dry plants, that they are easily winnowed with a fan. Because chickpeas don’t really like how much rain we have, there are usually some seeds which have molded or sprouted; I remove them by pouring them onto a plate and picking them out by hand. When I am done, I test a number of seeds to see if they are dry enough; usually they need bit more drying, so they go back on a tarp in the sun for a few days before I store them in a jar.

Rye

Rye is a fall-planted grain, so it matures in late July or early August. As it nears maturity, the stem turns from green to pale yellow, and the heads go from standing upright to tilted downwards. The last part of the straw to remain green is the final node below the head; usually small grains are ready to harvest about the time this changes colour. As the grain nears maturity, I check it nearly daily. When it is ready, we cradle, bind and stook it; we have found that one binder can about keep pace with one cradler in good conditions, if they both know what they’re doing but aren’t highly experienced (tip for binders: give the cradler lots of room and make sure you don’t catch up).

We let the stooks stand for one or two weeks, then bring them in and make a big pile next to the bike thresher. Two people then work together: one pedals while the other opens sheaves, divides them into smaller handfuls, and sticks the heads in the thresher. Every 5-10 minutes the thresher manages to tangle enough straw onto the shaft to drag it down; so we stop, open it and clean it out, and continue. The thresher has a built-in scalping screen. By the time a bucketful of chaff and grain has gone into the bucket underneath we are usually ready for a break. So we switch to the fanning mill, where one person cranks the handle to power it while the other adjusts various settings on the mill and watches what is happening to make sure it is working optimally. A few runs through the fanning mill cleans the seed quite well, but there is usually some contamination of spelt or barley that we want to pick out by hand before eating (since they’re hulled) or planting (since that would propagate contamination). Rye is especially prone to ergot contamination as described in the growing instructions, so we pick that out by hand too.

Amaranth

We enjoy amaranth so much that we wrote an article especially about working with it.