What to Grow in a Small Vegetable Garden: The Right Test Is Not the Obvious One

Two square metres, or about twenty-two square feet, is a standard raised bed, a long balcony planter, or the corner of a lawn you finally dug over one weekend in March. The question that follows is always the same: what goes in it? And the instinctive answer is just as predictable: whatever you like eating.

That is a perfectly sensible test in a large plot or on a full allotment. It becomes misleading the moment the area shrinks, because it ignores the one genuinely scarce resource in a small bed: space, multiplied by the length of time it stays tied up.

Yield is measured per square metre and per month of occupation

A vegetable does not just cost you space. It costs you space for a duration. Two crops each taking a quarter of the bed can come out on opposite sides of the ledger, depending on whether one clears the ground in six weeks while the other holds it from April to October.

In a plot of a hundred square metres, roughly a thousand square feet, a sprawling crop goes unnoticed. In two square metres, a plant that takes half a square metre for six months has claimed a quarter of your growing space for the whole season.

The useful mental tool is therefore a division: what the area returns, set against the number of months it is blocked. It is not a measurement and should not be given a precision it does not have; it is a way of comparing two choices, and it settles most of the arguments.

What that test rules out, including some much-loved crops

Pumpkins and trailing winter squash are the clearest case. A single plant runs stems several metres across, holds the ground from May to October, and delivers a handful of fruit — of a vegetable that stores well, is available everywhere in autumn, and costs very little. Maximum sprawl, maximum occupation, low practical value.

Sweetcorn fails for an additional and structural reason: it is wind-pollinated. A few plants in a line pollinate poorly and the cobs fill unevenly. Proper pollination calls for a block of several close-set rows, which is incompatible with a raised bed.

Potatoes in quantity present a different problem: they grow perfectly well, but they are among the cheapest vegetables to buy and among the bulkiest to grow. A few earlies for the pleasure of turning tubers out of the soil in June, by all means; a storage row in two square metres, no.

Heading cabbage, finally, illustrates the cost of time better than any other crop. Planted out in early summer, it holds its position into autumn or winter and produces, at the end of it, a single head cut in one go. That is a structural observation rather than a measured yield: several months of occupation for one indivisible harvest. Leafy brassicas, kale above all, behave quite differently, since you pick them leaf by leaf for months.

None of this makes these plants bad: they are badly placed. If a pumpkin makes you happy, grow one, knowing what it costs.

What genuinely earns its space

Cut-and-come-again leaves are the first thing to reach for. Loose-leaf lettuce, salad mixes, chard, spinach, kale: the plant is established once and harvested repeatedly. The ground stays occupied, but it produces continuously instead of producing once.

Next come the very short crops, the ones that finish their cycle in under six weeks. Radishes, rocket or arugula, baby leaves, early turnips. Their value is not their bulk but their speed: they slip into the gaps in the calendar and release the space before they have got in anyone's way.

Herbs probably offer the best pleasure-per-square-foot ratio in a small garden, for a reason that is commercial rather than agronomic: you buy them in small quantities at a high price by weight, often in a plastic packet half of which ends up in the bin. A clump of parsley, chives or thyme takes up very little and can be cut for months. Basil and coriander or cilantro need repeat sowings.

That leaves the vertical dimension, which is the real space reserve of a small garden. Cordon cherry tomatoes, climbing or pole beans, peas, a cucumber trained up a trellis: all of them exploit a height that would otherwise do nothing. Climbing beans deserve a mention against dwarf ones, which take up as much ground, crop all at once, then leave the space empty.

Interplanting, staggered sowing, and what they actually cost

Interplanting means slipping a fast crop between the still-small plants of a slow one: radishes among carrots, lettuce among young tomatoes. It works on one condition that is rarely stated — pull the fast crop on time. After that point it stops being a bonus and becomes a competitor for water, light and nutrients.

Staggered sowings, every couple of weeks for radishes and salad leaves, are more reliable still: they head off the wall of lettuces that bolt together and end up going three-quarters to the neighbours.

As for growing upwards, it carries a cost you discover in July: a cordon nearly two metres, or six feet, tall casts a real shadow, and whatever grows behind it suffers. An underbuilt support gives way under the weight at peak cropping. And all of these techniques share one constraint: the more heavily a bed is planted, the more water it uses and the faster it dries out. An intensive small bed needs more watering than a sparsely planted large one, not less.

The overcrowding trap

There is a threshold beyond which packing plants in costs more than it returns, and it arrives sooner than you would think. Crowded plants stretch and go leggy reaching for light. Foliage stays wet for longer after rain or watering, which favours leaf disease. Demand for water and nutrients rises on the same volume of soil, and you eventually cannot reach the middle of the bed without treading on something.

An overcrowded bed often produces less than a properly spaced one, and produces it less well. This is not a matter of discipline: you grow two square metres for pleasure, and a tangled, diseased bed you cannot harvest offers none.

On two square metres, planning matters more than on two hundred

That is the paradox of the small garden: the less ground you have, the more planning counts. In a large plot a mistake in occupation gets diluted — the failed bed is one bed out of twenty. In two square metres, a badly chosen or badly timed crop ties up a large fraction of the garden, often for the whole season, and there is no slack elsewhere to make up for it.

This is not about building a twelve-month plan. It is about knowing, before you sow, what occupies which square, at what spacing, and until roughly when. It is a small thing, and it is exactly what is missing at the garden centre display.

Where Marmanda comes in

Marmanda lays crops out on the space you actually have, holds to the spacing each plant needs along with its sowing, planting and harvest windows, and shows the source and the limits of every figure it uses — including where a number is common practice rather than a measured value.

The useful question, in the end, is not "what do I fancy growing?" but "what am I willing to tie up, and for how long?". On two square metres, that is the same question asked honestly.