Soil Matters More Than Sun

A client rang me last spring convinced her garden had a light problem. Her Cistus had died, her lavender had gone woody and collapsed after two winters, and she wanted to know whether felling next door's sycamore would fix it. I went round. The garden faced south. It got seven hours of sun in June. The problem was not light. The problem was that she was standing on heavy London clay that had been compacted by decades of foot traffic and puddled after every rain like a stock pond, and she had planted a Mediterranean hillside palette straight into it without anyone, including me on a previous visit two years earlier, thinking to check.

I mention the previous visit because it was mine. I'd specified the Cistus. I looked at the aspect, noted the sun, and didn't dig a hole. That's the whole essay, really, but I'll earn the length.

Aspect is the easy question

Sun is the first thing every client understands and the first thing every designer measures, because it's visible, photographable, and requires no tools beyond a compass app. Soil requires a spade, twenty minutes, and a willingness to get your hands wet in front of a client who is paying you to look competent. I understand why aspect gets the attention it does. It's just that aspect tells you what a plant can tolerate above ground, and soil tells you whether it will survive below it, and the second question kills far more plants than the first.

Cistus, lavender, rosemary, most of the grey-leaved Mediterranean palette that has become the default aesthetic for anyone wanting a "low maintenance, drought-tolerant" garden in the south of England: these plants evolved on free-draining, often alkaline, often stony ground where their roots never sit in standing water. Put them in clay that holds moisture through a wet winter and they don't die of drought. They die of drowning, of anaerobic root rot, of the same waterlogging that would kill a Camellia in dry sand for the opposite reason. Sun has nothing to do with it. I've seen this exact failure on south-facing sites and north-facing sites with identical frequency, which should have been a clue sooner than it was.

Read the soil before you read the plant list

You don't need a laboratory to know roughly what you're dealing with, and I'd argue that if you're waiting for a lab report before making basic decisions, you're not yet reading a site properly. On a site visit I do three things, and none of them take more than half an hour.

First, the ribbon test, unglamorous but genuinely useful. Take a handful of moist soil and try to roll it into a ribbon between your fingers. Sandy soil won't hold together at all. Silt holds briefly and breaks. Clay will form a long, smeary ribbon that holds its shape, and the longer the ribbon before it snaps, the higher the clay content. It's crude. It's also more honest than guessing from the plant list of the previous owner, which is what I suspect happens more often than anyone admits.

Second, dig a hole and fill it with water. A percolation test, essentially, though that phrase makes it sound more technical than digging a 30cm hole and watching what happens. If the water's gone within a couple of hours, you have free drainage and the Mediterranean palette is back on the table. If it's still sitting there the next morning, you have a drainage-limited site regardless of what the aspect promises, and you plan accordingly: moisture-tolerant species, raised beds, or serious structural drainage work that most residential budgets won't stretch to, in which case you change the plant list, not the site.