May 1994

Home Index Museums Blog Authors Site Map About


Three Sisters in Jerusalem

My Three Sisters Garden


Stephen Lewandowski

In April of 1987, I moved to a trailer on property I own in the Town of Jerusalem, Yates County, in New York State. No one had lived in the trailer for years, not since the previous owners used it as a summer cabin in the late 1970s. From the village of Naples where I had been living, I brought my cat, a bunch of books and clothes, and a long gardening history.

I had been working for several months on a paper, published later by Agriculture and Human Values, entitled "Three Sisters in Seneca Life" and had immersed myself in the historical, anthropological and archaeological literature of native North American agriculture, but I wanted to give my research and curiosity some practical application. I had studied native gardening techniques while raising gardens of my own in the Seneca territory of upstate New York—Victor, Honeoye Falls, Canandaigua, Mendon and Naples.

For a long time, I've thought that the people who came before us in this place have something important to teach us. Our culture's relationship with this land spans barely two hundred years. Theirs stretched at least five hundred years back of ours, time enough to develop institutions and lifestyles uniquely adapted to the place. Perhaps we can learn to better attune our lives to this place by seeking to understand others' experience. Actually, my conviction in the matter is less equivocal than that "perhaps" indicates; I think our survival requires acknowledgement and adaptation to the special qualities of place. Giving free rein to our technical abilities to subvert and homogenize places could destroy us in the end.

My own experiment in learning what the place requires would involve reconstructing A Three Sisters garden in the summer of 1987 in Jerusalem, New York. Regardless of Thoreau's testimony about the need to simplify our lives in a rural setting, I found that holding to my decision greatly complicated my life. Some of the complication was good and instructive.

The concept of a Three Sisters garden is deceptively simple. Corn, beans, and squash (and pumpkins) are planted together in raised hills. Field preparation and harvest processing involved men, but the planting and daily garden work was carried out by groups of women. Children were enlisted as workers and sentries.

On May 19, about two weeks after moving in, I paced off a fifty by one hundred foot garden plot, marking its outline with sticks and branches stuck in the ground. The next day its soil plowed up nicely behind my neighbor's Ford 9N, rolling off the share in a continuous sod carpet, now green and tan on the bottom and grey-brown on top. My neighbor Lewis Wild was surprised at the ease of plowing, said we had "caught it about right" for tilth. The days were warm, and the soil dried well. He returned two days later to fit the garden with a harrow, dragging it this way and that.

On May 28, with the help of a friend and a Datsun to pull the wire tight, I fenced the garden with a couple hundred feet of four foot tall American wire. We drove ten metal posts for the wire and braced it in the long stretches with short oak and ash branches.

Though plowing is not an authentic beginning for a prehistoric garden, it was necessary to prepare the ground. Under authentic conditions, the ground would have been covered with huge trees, which we would have cleared by girdling them with stone axes and fire. For several seasons, the crops would have been planted among snags of dead trees. In my case, the ground had been cleared of trees many years before and kept open. A few cellar and wall remnants near the trailer are reminders of a vanished farmstead. The open fields had grown hay but by the time of my arrival had reverted to a mix of grasses and golden-rod.

The garden soil is classified as a Fremont silt loam and located at about 1585 feet above mean sea level, probably somewhat higher in elevation than most Seneca gardens. The garden plot slopes moderately to the north and east. Morning sun will strike the garden early. The slope allows for drainage of cold air toward Sherman Hollow and the Guyanoga Valley to the east. The garden is in the upper reaches of the Nettle Creek watershed, which joins Flint Creek several miles to the north. Flint Creek combines with others in the Oswego River system whose waters are bound for Lake Ontario.

Fremont silt loam is a deep but somewhat poorly drained soil formed in glacial till. The topsoil is a dark grayish brown loam and the subsoil is mottled with patches of olive brown and grey. The soil is acid, not very fertile, and was used by farmers mostly for hay and pasture.

A fifty by fifty foot plot, about one-sixteenth of an acre, was devoted to the Three Sisters garden. I used the other fifty by fifty plot for production of standard garden fare: potatoes, tomatoes, cucumbers, lettuce, string beans and flowers.

On May 30, I rototilled the garden plot, which was dry and friable, and constructed one hundred and ten hills about one foot high and four feet between centers with a rake, hoe and shovel. It sounds easy, like that, in a single sentence. In actuality, the work was hard and took hours to complete. At three p.m., the temperature hit 90 degrees F. Main hill rows were arranged north-south and spaced four and a half feet apart. No sort of fertilizing matter was added to the hills. In the cool of evening I planted a row of Jerusalem artichokes and six flowering tobacco (Nicotiana rustica) plants along the garden's eastern margin.

That night, I put Tuscarora seed corn in to soak. The light creamy yellow corn was saved from gardens I raised in the Town of Mendon, Monroe County, in the summers of 1985 and 1986. Originally the seed came from an archeologist associated with the Incinerator Site, a 12th century Ohio village and reconstructed garden.

The next day, I used a short, stout stick to make a circle of holes in the flat top of the hills. In a few hours, I had planted six or seven kernels of the swollen corn in each hill.

Though the literature claims a higher population of hills and plants were present in Seneca fields, my experience of inter-planting multiple crops in hills is that access to the site becomes severely limited during the period of maximum growth, so I spaced the hills more widely, and in retrospect that decision seems to have helped not only with access to the plot but with allowing maximum vegetative growth. One hundred and ten hills on a sixteenth of an acre would have projected to about one thousand, eight hundred and forty hills per acre.

I had decided to give the corn a head start. Warm and moist conditions were nearly perfect for germination. About an inch of rain fell during late May. On June 4, the first corn shoots began to appear.

Weather conditions were variable in early June. On June 9, early morning temperatures fell to 45 degrees. On the 10th of June, with corn shoots up about two to three inches, I planted crookneck and pattypan squashes in the hills. Both varieties of Cucurbita pepo were among those planted throughout the prehistoric Northeast.

On June 13, I planted more members of the Cucurbita pepo family: acorn squashes, sugar pumpkins and field pumpkins. I also put in seeds of mixed varieties of gourds. All of them were planted three to a hill in every third hill. After the squashes, pumpkins and gourds were in, I planted Genuine Cornfield and Wren's Egg beans (Phaseolus vulgaris) at the rate of three per hill.

By June 16, the first seed leaves of the squash had appeared, on the 17th bean seedlings appeared, and by the 24th, true leaves had appeared on both kinds of plants. In late June, I lightly hoed some of the rows and picked a few weeds off the hills. The warm, wet weather was nearly ideal for the growth of food plants and weeds alike. During the month of June, about four and a half inches of rain in five storms fell on Jerusalem.

On July 5, I began the first major weeding of the garden, hand-picking the hills and hoeing the rows. Some of the hills had settled a bit, so the next day I scraped up material from the rows and added it to the hills around the plants. The corn plants were about twenty inches high and reached above my knee. Immediately after the hilling up, both squash and bean plants began to grow luxuriantly up the corn and run through the rows underneath.

The primary weeds of the garden were typical camp-followers of human activity, "agrestals" which have coexisted with cultivated crops for thousands of years. Quackgrass, goldenrod, ragweed, mustard, lamb's quarters, redroot, curly dock, wild carrot, dandelion, spurge and sow thistle emerged in large quantities. Lamb's quarters were particularly easy to control, however, since I let them grow to eight or ten inches, then picked, boiled and ate them.

Through mid-July, I continued weeding about an hour per day, often in the morning before work and the heat of the day.

By July 14, average corn plants were forty inches tall. During the day, an inch and a half of rain softened the ground, and heavy western winds flattened much of the plot. I felt discourged as I removed five plants that were totally broken off and surveyed the damage. It was hard to believe that the garden wasn't ruined.

When I did more weeding in the evening of July 18, I was working particularly on the hills. The sun was low to the horizon, and small details of the hills were conspicuous. Embedded in the side of one hill on the north side of the garden, I found a large fragment of a broken spear point, about one and one-half by one and one-quarter inches.

I consulted William Ritchie's A Typology and Nomenclature for New York Projectile Points which indicated that the Brewerton side-notched spearpoint was common in New York State between two and five thousand years ago. The mottled grey-brown flint warmed to my touch, perhaps like the handshake of its maker.

Even though I had felt that the garden was ruined, by July 20 most of the corn plants had regained their feet, and by July 21, the corn canopy was complete again. Corn plants measured fifty-four inches tall. Squash had run about four feet across the ground below. At this time, I began to observe that some of the lowest corn leaves were dying back, and some squash and pumpkin leaves showed some etiolation (leaf elongation caused by light starvation). Successful plants ran to edges of the plot to set fruit. Beans likewise seemed suppressed by the shade, and those on the edges showed some advantage. At about the same time, I noticed yellow stripes in the corn leaves, perhaps indicating a shortage of potassium.

On July 22, the first corn tassels began to appear. At the same time, heavy infestations of aphids and some corn borers and Japanese beetles began to show up on the plants. I did nothing to control these insects.

On July 28, the squashes were blossoming heavily wherever they were in the light, pumpkins had run eight feet from their mounds and were blossoming, and beans were growing luxuriantly and climbing but had not blossomed.

Despite the wider spacing, access to the garden was closing down in late July. On July 30, I ate the first small yellow crook-neck squashes from the garden. On August 1, I carried out another major weeding. On August 5, my evening meal included sliced crookneck and pattypan squashes fried up in butter, salt and pepper with a few tiny ears of corn

By mid-August, the garden was beginning to produce heavily. On August 11, I picked four pounds of crooknecks and four pounds of pattypans. On August 15, I harvested five pounds each of crooknecks and pattypans.

On August 16, the corn had reached full vegetative maturity, at a height of six and a half to seven feet. Like my previous planting of Tuscarora corn, it was showing a disposition to break at the joints and to have problems with smut (Ustilago maydis), but the ears seemed to be filling out and maturing without a problem. The tillers coming up beside the main corn stalks generally developed tassels but no ears. On August 16, for the first time, I noticed white blossoms on the Genuine Cornfield beans.

On August 17, three-tenths of an inch of rain and high winds again broke down some of the corn patch, but I noticed that with some of the cover removed, the beans seemed to renew their growth and flowering. On August 23, the first small bean pods were forming. I tried to weed the patch but, with the exception of some sparse quackgrass, very little was growing in the center. The combined shade of corn, beans and squash suppressed almost all growth. I pulled a few weeds around the edges.

On August 24, I picked about two pounds of pattypans and three pounds of yellow crooknecks, some for my own use and some to give away at the office. Soon after, the weather turned cold, and temperatures fell to 43 degrees F. on the morning of the 25th.

On September 1, I picked and sauteed some immature Genuine Cornfield beans. The beans and squash were regular parts of my meals through early September. From the 11th to the 20th, the weather turned cold and wet and the ground never really dried out between showers.

In late September, the weather cleared and warmed enough to dry the ground. Frost was in the air when, on October 2, I picked most of the Acorn squashes and pumpkins and brought them inside.

The corn plants were already dying back naturally. Between October 5 and October 16, I picked about a hundred pounds of the Tuscarora corn, pulled back, tied and braided the husks, and hung the ears to dry from the porch rafters. Early in the corn picking, I also harvested some of the beans and laid them out in shallow baskets on the porch to dry.

On October 11, without a preceding killing frost, about two inches of wet snow fell on the garden and lasted all day. When it melted off, I picked, shelled and dried the remainder of the beans, not all as mature as I would have liked. Dried, there was four and three-quarters pounds of Genuine Cornfield and four and one-quarter pounds of Wren's Egg beans.

Counting those eaten and those stored, the garden yielded twenty pounds of yellow crooknecks, twenty pounds of pattypans, thirty pounds of acorn squash, and forty pounds of pumpkins.

Converting all the figures to per-acre yields, one acre of garden would have produced about twenty-eight bushels of dry corn, two hundred and twenty-five pounds of dry beans, six hundred and forty pounds of summer squash, four hundred and eighty pounds of winter squash, and six hundred and forty pounds of pumpkins.

Doing some layman's nutritional calculations, I estimated that one acre of Three Sisters garden with production like mine would supply over one thousand, four hundred pounds of carbohydrates and over a hundred thousand grams of protein per year. Factoring in average nutritional requirements, an acre of Three Sisters garden could support three or four people, without other sources of hunted or gathered food.

Considering my gardening experience, I came to some general and some specific conclusions. Most of them should be tested further, but I'll share my thought with you, in case you are the sort of curious people who'd test these conclusions for yourself.

Three Sisters garden practices embody a number of values and efficiencies. For example, the various characteristic leaf postures and positions make a thick carpet of vegetation for the capture of solar energy and thoroughly suppress weed growth below. The plant leaves intercept and buffer the impact of raindrops on exposed soil, cutting soil erosion to nothing. Various rooting depths of the plants draw water from different layers of the soil, and the vegetation's shade reduces evaporative losses.

The Three Sisters was part of the Seneca strategy of land use that involved rotation of fields rather than crops. Probably only crop residues were added to the hills as fertilizer, but the various rooting depths of the plants would have been efficient in retrieving leached nutrients. Abandoned garden fields with regrowth of shrubs and trees would be re-used as sources of building poles, wildlife, and crops such as berries.

Pests were present in my garden but not at harmful levels. High populations of aphids, some cornborers and some smut were minor problems for the corn. No borers were present on the squashes, nor were the beans bothered by thrips, beetles or cutworms. Possibly the mixture of crop species acts to dampen the build-up of harmful pest populations.

Three Sisters gardens require much labor, but their needs aren't constant. Because of the variety of tasks involved, they could engage the skills of whole families. The gardens are expressions of family and community life.

The Three Sisters adapted to local climate, and the Senecas adapted to the Three Sisters culture for over five hundred years. My short experiment with the Three Sisters led to an admiration for the people and the feeling that their agriculture was efficient, sophisticated, and altogether adequate to their needs.

© 1994, Stephen Lewandowski
Index to articles by Stephen Lewandowski
CLR Blog | Site Map | Contact CLR