One: that you must dig

You do not. The beds I stopped digging four years ago drain better, dry out
slower, and grow better carrots than the beds I turned every autumn.

Two: that compost needs turning

• Turned heap: ready in four months, and a weekend of work
• Unturned heap: ready in eleven months, and no work at all

Both produce compost. Only one produces back pain.

Three: that watering daily is good for plants

A daily splash keeps the roots in the top two inches where the soil dries
first. A long soak twice a week sends them down. The plants that survived
last August were in the beds I neglected.

Four: that you should grow what you like eating

Better rule: grow what is expensive or bad in the shops.

1. Salad leaves, which are both
2. Broad beans, which are never sold fresh
3. Herbs, which cost more per gram than most meat
4. Not potatoes, ever, which are cheap and take a quarter of the plot

Five: that a plan survives April

It does not. The plan’s job is to be there in January when there is nothing
else to do.

Six: that the shed needs organising

The shed does not need organising. The shed needs less in it.

Ranked, best first

1. Nematodes. Expensive, needs warm wet soil, and works. Six weeks of cover
   per application, so three applications a season if you mean it.
2. Beer traps. Cheap, effective within about a metre, and you have to empty
   them, which is genuinely unpleasant.
3. Copper tape. Works on pots and only on pots, and stops working the moment a
   leaf bridges the rim.
4. Night patrols with a torch. Free, effective, and it costs you the evening.
5. Pellets. I do not use them, because the plot has hedgehogs.

What does not work

• Crushed eggshells
• Coffee grounds
• Wool pellets, in my soil, though people I trust swear by them
• Hoping

The thing that changed most

Not a defence at all: raising transplants to a decent size before they go out.
A slug can strip a two-leaf seedling in a night. It can only annoy a plant
with six leaves and a root system.

This season’s plan

1. Nematodes in the brassica bed, mid-May
2. Beer traps around the hostas, refilled weekly
3. Nothing at all for the beans, which have never once been touched
4. A torch by the shed door

The ledger entry I keep forgetting

Slug damage is worst in the week after transplanting and almost irrelevant a
month later. Everything on this list is really a way of buying one week.

The change

Sow a short row every fortnight instead of a long row once.

The crops it suits

• Lettuce and other cut-and-come-again leaves
• Radish, which is ready in a month and bolts a week later
• Carrots, early varieties only
• Beetroot, up to midsummer
• Dwarf French beans, three sowings, six weeks apart

The crops it does not

• Parsnips, one sowing, that is the whole season
• Leeks, which want a long single run
• Anything you grow for storage rather than eating fresh

The fortnight, in practice

1. Sow row A on the first Saturday
2. Sow row B two weeks later, alongside
3. When row A is cleared, that space becomes the next sowing
4. Stop six weeks before the first frost date, which for me is early August

What actually happens

I miss a fortnight in June, always, because June is when everything else needs
doing. The gap shows up in the kitchen in August and by then it is too late to
fix.

The fix is a note in the calendar rather than a note in a notebook.

The method

1. Take five samples per bed, one from each corner and one from the middle,
   a trowel deep
2. Mix them in a clean bucket and pull one test sample from the mix
3. Dry the sample indoors overnight; a wet sample reads low
4. Follow the kit, which is always: soil, powder, water, shake, wait
5. Read it in daylight, not under a kitchen bulb

What I got

• Bed one: 6.4, the brassica bed, which explains the club root
• Bed two: 6.8, fine, leave it alone
• Bed three: 7.2, the old bonfire corner
• Bed four: 6.6, fine

What each result means

Brassicas want 7.0 and above. Potatoes want 5.5 and will take scab at 7.0.
Everything else is happy between 6.0 and 7.0 and will not thank you for
fussing.

The plan

1. Lime bed one this month, at the rate on the bag, and do not plant into it
   for six weeks
2. Move the potatoes to bed three next year, where the high pH will be a
   problem, so lime nothing there and add leaf mould instead
3. Retest in September, before the autumn manure goes on
4. Write the numbers on the shed door where I will actually see them

The step I skip and should not

Step three. Every time. The sample goes in wet, the reading comes back low,
and I lime a bed that did not need it.

Under cover, this month

• Broad beans, Aquadulce, in root trainers
• Onions from seed, in modules, kept cold
• Sweet peas, soaked overnight first

Direct, once the soil is workable

1. Garlic, if the autumn planting failed
2. Shallots, pushed in rather than buried
3. Early peas, under cloches, two weeks after the shallots
4. Parsnips, last, because they will sit and sulk in cold ground and then rot

The rule behind the order

Everything that needs a long season goes first, everything that resents a cold
start goes last, and the gap between the two is exactly as long as your soil
takes to warm.

The parsnips are the test. Sown too early they germinate at a rate that makes
you think the seed was old. It was not. The ground was cold and the seed sat
in it for three weeks going soft.

What I am not sowing

• Tomatoes, because there is no heat until March
• Courgettes, for the same reason and also because four plants is three too
  many and I do it every year anyway
• Anything brassica, until the netting is repaired

The wrapped item above is the honest one. A list item that runs to two lines
is still one item, and a diary that pretends otherwise is a diary nobody
wrote in.