A big reason to avoid buying topsoil is that it has usually been heaped in a stack. When this happens many of its organisms die, from lack of air and food.

This was unintentionally proved by the UK National Coal Board (NCB) in the 1940s–70s.  Some of their mines had coal near the surface, and they would scrape off soil to remove the coal, called ‘opencast mining’.

After removing the seams of coal, they put the soil back so that farmers could ‘carry on farming as before’. Except they couldn’t: growth was so poor, and nobody understood why.

The NCB sought advice and struggled to discover what the problem was. Finally it was the insights of a soil biologist in the 1970s, Professor Victor Stewart, that revealed why plants were struggling.

He discovered that the soils contained enough nutrients, but were slumping and ‘dead’ from being stacked too long in heaps. They needed bringing back to life with additions of organic matter. Stewart’s work, on these farms and elsewhere, highlighted the importance of soil biology and structure to maintain a balance of air and moisture, with a free flow of both.

I was privileged to meet Victor and discuss this with him in the 1980s, and we agreed that his work was still ahead of its time. In other words, not many farmers or scientists were asking him for advice.

In the 1980s, and even for another 30 years or so, soil research majored on nutrient status and chemistry, rather than biology. Soil was seen as a ‘bank balance’ for an NPK chequebook, receiving nutrients in and sending/allowing nutrients out.

Chemists and farmers controlled growth through managing the input–output equation, for which you needed to know which plants are heavy feeders, which are light feeders, and when they all like to feed. As Bower commented, it was getting mighty complicated.

More confusions were happening on the cultivation front. Roots were seen as accessing their nutrient ‘bank’ by free movement in a loose soil. In this scenario (but not in reality), if soil was not mechanically loosened, growth would be compromised.

Compacted soil

The word ‘compact’ is overused and causes confusion, linked to the false belief that soil must be opened up by tools for growth to happen. This misunderstanding results in the incorrect labelling of firm and healthy soil as ‘compact’.

However, firm soil is not the same as true compaction, which happens rarely. A definition of compacted soil is that air and moisture cannot pass through any free channels. Soil then changes colour to orange, grey, or even pale blue. And it becomes smelly, like a mildly sulphurous swamp. Plant roots can’t survive in compacted soil.

Nonetheless plant roots do like firm soil, into which they can anchor themselves and be stable. Ideally, there is an ‘aggregation’ of soil into lumps of different sizes. Roots and associated fungi travel within this matrix of soil aggregates, both between and into them, in search of moisture and food.

• Aggregated soil particles are coated in organic matter, the glomalin that we looked at in Lesson 14.
• There is a slow flow of air in and out of soil, enough for roots to breathe. Soil with air smells ‘sweet’.
• Aeration is helped by the good drainage of no dig soil, so that air is not displaced by water.

The orange clay in these photos was probably a result of earlier work by builders, because it is close to Homeacres’ house wall. Some compaction at this depth is not good for growth, and there was a slight smell of putrefaction. However most of the soil at Homeacres is healthy so this was not a major problem, and aeration improves when soil organisms are active and can penetrate any compacted layers.

As ever there are exceptions, such as where excessive passage of heavy machines in wet weather has squashed soil into such density that almost nothing can pass. In this case using a broad fork once makes sense, but just once. Then mulch the surface, and allow soil life to work quietly.

You can avoid much of the difficult work of dealing with new weeds by filling beds with compost rather than soil, and always making compost the surface layer.

• All soil contains weed seeds, and sometimes roots of perennial weeds.
• After soil is moved, weed seeds germinate when exposed to light.
• Soil is sticky and heavy compared to compost, and weeds pull out less easily.

Below are some photos of a trial I did, just to see which growing medium led to the most weeds growing: soil, homemade compost, fresh horse poo or old horse manure.

It was in winter and growth was slow, but there was still extra germination of weeds in the disturbed soil. (We’ll look at this again in Lesson 16.)

Most soil you can buy is of a lower quality than any you may have on site, say from building work at your property. Only use soil from the top 30 cm (12 in) or so of darker topsoil, depending on how deep a layer you see.

Soil holds fewer nutrients than most composts, especially potting composts. Hence, for propagation, it’s worth buying commercial products designed for that use. You can use multi-purpose compost, also called all-purpose, for every stage of propagation, including seed sowing.

Incidentally, in the US I have noticed that potting compost is called potting soil. Until I realised this, I was often having confused conversations with American gardeners.

I did a trial on soil for potting, and you can see the results in the photos below. The soil pots had one third of cow manure mixed in, to see how viable that might be as a mix. It would have been interesting to grow some plants in soil only, without the nutrients of the old manure. Even so, my conclusion is to stick to compost for propagation, for the extra nutrients and better retention of moisture.

‍

However, if you are a garden landscaper making permanent features and new contours, it does make sense to use mostly soil. The shapes and undulations will endure!

In contrast for vegetables you want mostly compost, but with two exceptions:

1. For making beds more than 25 cm (10 in) high, a bottom layer of up to 50% soil serves to maintain the volume needed to keep it full. For beds of 60 cm (24 in) and higher, use two thirds of soil at the bottom.
2. To ‘lose’ unwanted soil, it can be the base of any bed. This happened at Homeacres when the builder was digging trenches for my greenhouse. They dropped soil on the pasture where I was about to make new beds, as a 5 cm (2 in) base layer. It took a lot of effort to spread the sticky soil into a bed shape.

To discover more about the differences between growing in soil and in compost, I filled two beds at Homeacres in November 2012 – one with compost and one with soil. The compost bed was actually my very first Homeacres bed, next to where I would soon establish the dig/no dig trial beds.

The beds both measured 1.2 × 2.4 m (4 × 8 ft), and the sides were 2.5 cm (1 in) thick planks of softwood, of 15 cm (6 in) depth.

Each bed had 15 cm (6 in) of ingredients, with no cardboard on the weeds. There were enough dense ingredients that only a few small dandelions, buttercups and weak blades of couch grass made it to the surface. However, annual weeds were another matter.

• The soil was fertile and had been growing strong weeds. I pulled out any roots of dandelions and couch grass, but a lot of weed seeds remained.
• Because the soil was freshly moved, many of its organisms survived to carry on their work, although some would have perished in the move.
• It was true topsoil, from the surface 15 cm (6 in).

My method was to grow and compare the same vegetables in each bed. However, I now realise that perennial kale was not a good choice for the middle section, where the beds were adjoining, and only separated by a plank of wood.

The kale plants grew to over 2 m (7 ft), and their extensive root networks could access the adjoining bed, as well as the one they were in. I could not say that either pair of kale plants were growing exclusively in a soil-filled or a compost-filled bed.

‍

The main, and almost immediate difference between the beds was a massive quantity of new weeds germinating on the soil bed, while there were only a few on the compost bed. The soil had a lot of weed seeds and they wanted to grow, a healing job I imagine.

The soil weeds were time-consuming because there were so many to pull, one by one – the only option in early 2013 when it was too damp to hoe. This mass of seedlings were even more difficult to remove as the soil was extra sticky from having recently been moved.

Twice, in early spring, Steph and I spent a whole hour on the soil bed. We pulled speedwell, grass, celandines, buttercups, nettles and bittercress. Although small seedlings, their roots hung onto the soil as though glued to it, and our fingers became less nimble, coated in sticky soil.

You rarely suffer this stickiness with compost mulches. In fact no weeds grew in the compost bed, thanks to two factors:

1. I had added few weed seeds when filling the compost heaps, thanks to my garden almost never having seeding weeds.
2. The compost heaps had mostly attained a high enough temperature to kill any weed seeds.

By late spring the surface of the soil bed was clean and drier, and I could hoe lightly. Superficial hoeing is quick, and from then onwards it was easy to keep the soil surface clean. By autumn there were almost no new weeds. At that point in a dig garden, the spade or fork would have returned things to square one: disturbed soil with many weed seeds germinating.

The growth differences were smaller than I had expected, as the year one results show in the table below. In 2013 some vegetables even grew larger in the soil bed, mostly first plantings such as early potatoes, garlic and shallots.

Possibly this is because working and moving soil admits air, which stimulates bacterial activity, causing conversion of organic matter into nutrients. Hence, when diggers start with a reasonable level of organic matter in the soil, they can ‘use’ that for a while to have reasonable growth, while reducing soil organic matter.

Growth was good for all plantings on the compost bed. By autumn, most of its second plantings were more productive, and had glossier leaves than the same plantings on the soil bed.

I think that the high quality of the soil used here made a big difference. If I had bought soil, it would have been different in these two ways:

1. Older, and therefore less full of microbes and organisms, which diminish in heaps of soil.
2. Less aggregated/structured in lumps, from being less fresh, and having a higher proportion of subsoil compared to topsoil.

In subsequent years, because the soil was not diminishing in volume, it became difficult to add compost to feed the soil bed – the surface rose higher and above the boards. The compost bed gave stronger growth throughout 2014 and 2015, until I stopped doing comparative plantings in 2016. We removed the wooden sides, now eight years old, and discovered many slugs in rotten cavities inside each plank.

In 2013, the total yield of vegetables was 20 kg (44 lb 1 oz) from the compost bed, and 19.1 kg (42 lb 1 oz) from the soil bed.