1. When I started to create the small garden, my first job was to remove the concrete posts and wire fence. This entailed removal of the roots of stinging nettles (Urtica dioica) which were entangled in the wire. Normally I mulch over nettles, after cutting them very short to ground level.

2. I used a sharp spade to remove the top 10–15 cm (4–6 in) of the few docks I saw (Rumex obtusifolius). There were no other perennial weeds, perhaps because chickens had been running here for a year or two. Prior to that, in the 1960s–70s, this area was covered by a shed with boilers for the nursery, and the soil has slabs and pieces of concrete below the surface. This matters less in no dig, it is just annoying when pushing in posts and canes to support vegetables!

3. I cut back all growth from the neighbour’s shrubs, as close to the fence as possible. I also pulled out many fat white roots of hedge bindweed (Calystegia sepium). It is common in his garden and continually spreads into the small garden (as does Euphorbia), but both are manageable.

4. I looked at the plot and decided that three beds of 1.4 m (4.6 ft) would make best use of the space, with one side 70 cm (2.3 ft) from the tin shed, and the other side 40 cm (16 in) from the brick lean-to on Homeacres’ east wall. The two paths between the beds are 30 cm (1 ft) wide. If I had chosen to have four beds, each would have been 1 m (3 ft) wide, and that would have meant an extra path, resulting in a small loss of growing space and more maintenance per area. For anyone liking narrow beds this would be possible here; my preference is for wider beds. You can be creative in making such choices – it’s about what you like to do, not what is right or wrong.

5. I kept a path at either end: it’s easier to maintain a path near to a neighbour’s garden than a bed, because the invading weeds make less progress into a path than into a bed – the path is like a quarantine zone.

The small garden gave harvests of 93.7 kg (207 lb) of vegetables in 2018, with a steady 2–4 kg (4.4–9 lb) every week from mid-May, less before that. No dig makes it easy to keep resowing and planting throughout the season.

The photos below show how beautiful a vegetable garden can be. This is an important aspect, and helps you to stay involved. I strongly advise that you garden a small area to its full potential, for high productivity and full enjoyment, compared to a less intensive and larger area.

(See Lesson 18 for a detailed plan of 2018’s plantings in the small garden, and the results.)

‍

The other side of the tin shed was a different challenge, because I mainly wanted it clean, clear and looking nice. In 2012 I was seeing Homeacres as a teaching and trials garden, more than one for producing and selling vegetables.

Originally, I also thought to have my new polytunnel closer to the shed, until I discovered the concrete under 10 cm (4 in) of soil and weeds. The whole area had been a shed or hard standing 40–50 years ago. I then started to worry about what else I might find under Homeacres’ soil!

Every cloud has a silver lining, and I scraped off the soil (which had formed on the concrete) to fill a bed, which was then used in a soil/compost trial (see Lesson 15). Before filling the bed, I removed the roots of perennial weeds from the soil (such as nettles, buttercup and bindweed), and they went on the compost heap.

The concrete surface has proved ideal for stacking crates and washing salad leaves. The concrete is not thick, about 7 cm (3 in), and could have been cleared if I had wanted the area for growing. I could probably have made beds on it, because it is broken enough to allow water to drain through (although removal would be worthwhile before growing).

Drainage is best where materials are of the same or similar densities. For example, the various stages of compost decomposition are similar enough in nature for water to flow freely. Likewise in undisturbed soil, where changes in texture are gradual and do not impede drainage.

However, if you laid gravel on concrete as a ‘drainage layer’, water would resist flowing from the compost and into the gravel, because it makes a capillary layer across materials of different densities. This is why gravel or shards of clay are not a good idea for the base of pots before they are filled with compost. I fill pots and containers with compost only, for good drainage and more food plus moisture than if you had put gravel at the bottom.

If you should inherit a greenhouse with a concrete base, and you want to grow vegetables, it’s probably worth hacking out the concrete to expose the soil. (See the pictures below for what the Catlows did, after attending a day course here. They are in Lancashire in the north west of England.)

‍

I was advising a gardener, who lamented how her squashes had wilted every afternoon of a hot summer despite her regular watering. Was that part of the garden just too hot for them? I was immediately suspicious of tree roots, which travel further than one might imagine. I asked her, and yes, they have tall acacia trees in the pavement just near their front garden. She had even seen suckers growing up – new little acacia trees from these roots – a long way from the trees.

Her squash plants wilted every afternoon because the trees could suck the moisture faster than vegetables; squash need to grow fast, from good access to moisture.

Many of us have borders and edges with our own or neighbours’ trees nearby. Unless you can pollard the tree, or cut a hedge very short, there is not much that can be done. Cutting surface roots with a sharp spade is possible, but many roots run deeper than a spade can reach, and that rarely answers the problem.

Don’t consider a membrane to ‘keep roots down there’. They will eventually find a way through, and you will have poisoned your soil with a layer of plastic, restricting movement of soil life and of vegetable roots growing downwards.

Near to deciduous trees, an option would be to have vegetables that grow in winter and spring, such as garlic, spinach, autumn-planted salads and spring cabbage. Even broad beans sown in November. They do much of their growing when tree leaves are not present and roots are therefore dormant – it’s more about roots than shade.

‍

Much of the conventional reasoning given for deciding how wide to make beds is based on a false assumption: that you cannot or should not walk on them. I have already explained how this is untrue, so deciding bed width is your call, according to what you are comfortable with and what will fit into your space.

Beds of 1 m (3 ft) width are my minimum, and are ideal for rows of tall peas, or one row of courgettes/zucchini; plus they are easier to pick leaves from. Bed widths of 1.2 m (4 ft) use space more efficiently, from a higher proportion of bed to path area, and up to 2 m (7 ft) is possible. At this width you can be creative with spacings of vegetables such as climbing beans, by making wide teepees for example.

Another consideration is square foot gardening, a method for growing a lot of variety in a small area. Your square feet can be other sizes, square 40 cm (16 in) for example, and on beds of varied width. Have fun trying out a few things.

I caution however, that the common bed ingredients advised for square foot gardening are one third compost, one third peat and one third vermiculite. This combination is weak in nutrients; I had a mail from a gardener who had grown for ten years using this recipe and who had suffered declining harvests. He copied my advice to simply fill beds with compost of any kind, including old animal manure, and said that for the first time ever he now had more parsnips than they could eat.

‍

How long to make beds? They are easier to manage when small, depending on how much you grow. The small garden beds, which are 4.5 m (15 ft) long, grow four crops in each bed at any one time. My dig/no dig trial beds are slightly longer (see Lesson 4), and grow up to ten kinds of vegetable at any one time.

Homeacres has bed lengths ranging from 2 m (6.6 ft) to 12 m (39 ft), and I like them all. The garden is prettier when not totally geometric, and many of the longer beds have mixed crops at any time of year. Not being slave to a four year rotation helps to achieve this exciting patchwork effect.

Orientation

The oft-advised north–south orientation, suggested to minimise shading effects, applies only if most plants are tall ones. On east–west beds these will shade plants on the north side at midday.

This happens in my polytunnel during summer, which I aligned almost east–west to be parallel with the neighbour’s hedge, and to reduce wasted space. In summer I grow tomatoes and cucumbers in the middle beds, so the vegetables on their north side are somewhat shaded, but still good!

More inconvenient is hot afternoon sun stressing cucumbers in the southernmost row, because it dries the soil exposed to sun at their roots, more than elsewhere in the tunnel. However, this happens for only four months of the year; mostly the east–west orientation is fine, such as for salad plants and garlic in winter.

• The majority of vegetables are low growing, making a north–south factor redundant. Plus it is outweighed by important considerations of access and convenience – the entrance to a path needs to be where you enter the plot. Bear all this in mind when making your first bed, because the layout of all the others will usually follow these lines.

A final consideration is slope. Conventional advice recommends running beds across a slope, rather than up and down. This follows its own assumption that soil has been loosened by tools, which makes it less stable.

This does not apply to no dig, where erosion is minimised. In my garden at Lower Farm there were slopes of around 10 degrees, and I found it worked better to run beds up and down, rather than across the slope.

When beds are oriented up and down the slope, newly applied compost then runs down the bed (if going anywhere!), rather than into a path below. Similarly for water. Plus my access point was at the bottom, so it made sense for me to run paths straight up the hill.

Only in extreme rain did I find some of the paths’ surface material run down, and, in just two years out of 13, I needed to load compost and soil at the bottom, and redistribute it at the top. It came to two wheelbarrows full for the 1440 m^{2  (}0.35 acres).

If your slope is more than, say 12–15 degrees, it’s a different matter:

• Either run beds across the slope, and use wooden sides on their bottom edge only, to hold a reasonably level surface for each bed.
• Or, for larger areas, you could hire a digger to create terraces, as you see in the photos. This means you have steep slopes between each terrace, which will grow weeds unless you have a growing idea for them.

‍

Creating new beds is simple, once you are clear about the layout you want and which width will work for you. You need mulching materials, and at least some must be organic matter.

• Organic mulches keep light off existing weeds and emphasis feed soil life, creating fertile soil even when it wasn’t fertile before.
• Polythene/plastic mulches are quick to lay, but do not feed soil. I suggest not to use polythene mulches unless:

1. The weed problem is severe.

2. Time is limited.

3. You don’t need to crop for a while.

For most beds at Homeacres, I have laid temporary wooden sides to contain the compost. I made an exception for the two dig/no dig trial beds, whose oak sides are permanent.

In the area of the garden shown in the photos below, the beds could have run either way, but generally it makes sense to follow existing path lines, which we did. This gives clear access from one end of the garden to the other.

As it happens, we often carry watering cans across the line between the old and new beds, from the water butt by the greenhouse to the polytunnel. Before the new beds were in place, this had been grass. Since it’s fine to walk across the mulched surface of no dig beds, this is what we do, without there being a proper path. Which leads on to the next lesson about paths.

‍