Why a Soil Test Beats Guesswork

Most smallholders begin with a bag of general-purpose fertiliser and a fair amount of hope. It can work for a season or two, but after a while you notice odd patches — a strip of stunted brassicas, a bed where carrots fork, a corner that stays yellow however much you water it. That is usually the point at which a soil test starts to look like money well spent.

A laboratory analysis tells you two things you cannot reliably judge by eye: the pH of your soil and the levels of the major plant nutrients it holds. Get those numbers right and you stop wasting money on fertiliser the plot does not need, and you stop losing crops to deficiencies you never knew you had. On a smallholding, where every bed is precious and there is rarely a spare hectare to experiment on, that is worth an afternoon of your time.

Taking a Sample That Actually Represents the Plot

The single biggest source of misleading results is a poor sample. A handful of soil from one spot tells you about that spot and nothing else. You need a representative composite sample, built from many small cores taken across the area.

• Divide the site into blocks of roughly uniform soil type, history and cropping. A vegetable plot, an orchard and a pony paddock should each be sampled separately. So should any area that has been heavily manured or limed in the past.
• Use a clean stainless steel trowel, a soil auger or a hand corer. Avoid galvanised buckets and tools, which can contaminate the sample with zinc.
• Walk a staggered W pattern across the block and take 15 to 20 cores, discarding surface litter and stones.
• For vegetable beds and arable ground, sample to about 15 cm. For permanent pasture, 7.5 cm is the standard depth.
• Tip every core into a clean plastic bucket, mix thoroughly, then take out about 500 g of the mixture for the laboratory.
• Sample at the same time each year — late autumn or early spring are both sensible — and never within three months of liming or six weeks of a fertiliser application.

Label the bag clearly with the field or bed name, the date and the depth. Send it off promptly, or air-dry it in a cool, shaded, dust-free spot if there will be a delay.

What to Ask the Laboratory to Test

Most standard agricultural packages cover the essentials, and for a smallholding that is usually enough:

• pH — the master number. Most vegetables, and most of the clovers in a grazing ley, want somewhere between 6.0 and 7.0. Below 5.5, and phosphorus becomes locked up while clubroot and other problems thrive.
• Phosphorus (P) — reported as an index, usually 0 to 5. Index 2 is the target for vegetable ground.
• Potassium (K) — again as an index. Root crops and fruit take a lot; index 2 is a reasonable aim.
• Magnesium (Mg) — index 2 is adequate. Low magnesium shows up as interveinal yellowing on beetroot, potatoes and tomatoes.
• Organic matter and texture — optional extras, but genuinely useful if you are trying to build fertility or improve a heavy clay.

If you suspect a specific problem — boron deficiency in swedes, say — mention it when you send the sample and ask for the relevant trace element test.

Reading the Results Without Panicking

The report will look forbidding at first, but read it in order. Start with pH. If it is low, that is your first job, because liming changes how everything else behaves. A slightly acid soil that is limed can release nutrients that were already present but unavailable.

Then look at the nutrient indices. Index 0 or 1 means the soil is short and will need feeding. Index 2 is a comfortable working level. Index 3 and above means you already have enough — adding more is a waste of money and, in the case of phosphorus, a genuine risk to nearby watercourses through runoff. Phosphorus accumulates slowly in the soil and is far easier to add than to remove.

Remember the numbers describe the soil, not the crop you intend to plant. A hungry leafy crop on index 1 soil may still need a top dressing; a low-demand crop on index 3 may need nothing at all.

Acting on the Numbers: Lime First, Then Nutrients

Lime is best applied in autumn, on bare ground, and worked into the top 10 cm or so. The laboratory generally gives a recommendation in tonnes per hectare based on your soil type and existing pH — follow it, but split very large applications over two seasons rather than dropping it all at once. Never lime and manure the same bed in the same month: they react together and you lose nitrogen from the manure and the benefit of the lime.

For nutrients, choose a fertiliser that matches the shortfall. A soil low in both phosphate and potash suits a balanced compound; a soil needing only potash can take sulphate of potash on its own. Nitrogen is rarely tested routinely because it moves so quickly through the soil — supply it instead through compost, manure, green manures and modest top dressings when the crop is actively growing.

A Simple Annual Routine

Test each distinct block once every three or four years, and rotate which block you sample so that by the end of the cycle you have a full picture of the holding. Keep the reports in a folder or a notebook, with the date and the plot name, so you can see trends rather than isolated figures. Record what you applied and when, then look back at harvest notes the following year.

Combine laboratory results with your own eyes: earthworm counts, water infiltration after rain and the smell of the soil tell you a great deal about structure and biology that no test measures. The numbers give you the foundation; the rest is observation, patience and a decent rotation.