
Growing guide
The one fruit where the most important decision is made before you plant, is invisible on the label, and cannot be changed afterwards. It is not the variety. It is the roots it is grafted onto.
Malus domestica · Rosaceae — with pear, plum, cherry, strawberry and raspberry
Above: Sun & Rain Works · © Sun & Rain Works
moderate. Undemanding once established and easy to get permanently wrong at planting. An apple tree is a twenty-year decision made in an afternoon, and the two things that decide it — rootstock and a pollination partner — are both settled before the tree goes in.
Almost every apple tree sold is two plants joined together: a variety grafted onto a rootstock chosen from a different plant entirely. The variety decides what the fruit tastes like. The rootstock decides how big the tree gets, how soon it fruits, how long it lives, and whether it needs staking for ever.
This catches people out constantly, because two trees labelled with the same variety can finish at eight feet or thirty, and nothing on the label may say which you are buying.
Very dwarfing stocks — the M27 and M9 family — give a tree six to ten feet, fruiting in two or three years, needing permanent support because the root system is too small to hold it up. They suit a small garden and a container, and they are unforgiving about drought and weed competition.
Semi-dwarfing stocks such as M26 and MM106 give twelve to eighteen feet, fruit in three or four years, and are the sensible default for most gardens. Vigorous and standard stocks such as MM111 and M25 make the twenty-five-foot trees in the photograph at the top of this page — decades to reach full size, a century of life, and a ladder for every harvest.
So decide the size you want first, then find your variety on that rootstock. Doing it the other way round is how people end up with a tree that shades the whole garden.
The old orchard shape is the open centre or goblet: a short trunk, three or four main limbs spreading outward, and a hollow middle so light and air reach every branch. It suits vigorous rootstocks, it makes a beautiful tree, and it needs a ladder.
The modern commercial answer is the opposite. Dwarfing stock, planted close, trained flat against wires so that every fruit is within reach from the ground and every leaf gets light. The photograph below is one version — a V-trellis, where two rows lean apart from a shared post line.
For a garden, the useful middle is a cordon, espalier or fan against a wall or a wire fence. A single-stemmed cordon on dwarfing stock takes eighteen inches of border, crops in its second or third year, and lets you grow four varieties in the space a standard tree would shade out.
Whatever the shape, the principle is identical and it is the only pruning rule worth memorising: light and air must reach the fruiting wood. Everything else is detail.
Prune in winter while dormant for shape and vigour, and in summer for restraint — summer pruning slows a tree down, which is exactly what you want on a wall-trained form.

Walla Walla, Washington, the same season as the photograph above and a century of thinking apart. Every tree here is on dwarfing stock, tied flat to wires so nothing is out of reach and no fruit is shaded. The whole block is a machine for putting light on leaves.
Walla Walla, Washington, dormant season
Sun & Rain Works · © Sun & Rain Works
Most apple varieties cannot fertilise themselves and need pollen from a different variety flowering at the same time. A single tree in an isolated garden may set a light crop or none at all, and the owner reasonably concludes something is wrong with it.
Nurseries handle this with pollination groups, numbered roughly one to seven from earliest to latest flowering. A tree needs a partner in its own group or an adjacent one — group 3 works with 2, 3 or 4.
In a town this is often solved already: apple pollen travels a few hundred feet, and a neighbour's tree or even a crab apple in a hedge will do the job. In an isolated garden it will not solve itself.
Two complications worth knowing. A few varieties are triploid — Bramley is the famous one — and their pollen is sterile, so they need two other varieties nearby and give nothing back. And a family tree, with three varieties grafted onto one trunk, is the neat answer for a garden with room for exactly one tree.
Do not rely on a self-fertile label. Self-fertile varieties do set alone, but nearly all of them set a heavier, better crop with a partner.
A tree carrying too much fruit makes small apples, breaks limbs, and — worst — exhausts itself so thoroughly that it sets almost no flower buds for the following year. That is biennial bearing: a huge crop, then nothing, then a huge crop.
Breaking the cycle is a job in June. The tree drops some fruitlets itself — the June drop, which is normal and not a problem — and then you thin what is left to one fruit per cluster, spaced four to six inches along the branch. It feels like vandalism and it is the difference between a bowl of large apples every year and a glut followed by a blank.
Feed and mulch in early spring, keeping the mulch off the trunk. Keep a clear circle of bare ground or mulch around a young tree — grass growing up to the trunk competes hard and noticeably slows establishment.
And check the graft union, the kink low on the stem. It must stay above soil level for the life of the tree. Bury it and the variety roots directly into the ground, escapes the rootstock entirely, and grows into the full-size tree you were trying not to plant.
Plant a pip from an apple you enjoyed and you will get a tree — and its fruit will not be that apple. Apples are extreme heterozygotes: every seed is a genuinely new individual, and the overwhelming majority are small, sharp and not worth the twenty years.
This is why every named apple in the world is a clone. A variety is one original seedling, grafted onward indefinitely — so a Northern Spy today is a cutting of a cutting of a tree that appeared in New York around 1800.
It also means every named variety was once a chance seedling that somebody noticed and thought worth propagating. That is the entire history of apple breeding until the twentieth century: not design, but attention.
The drawing below is what recording those varieties looked like before colour photography. The USDA employed artists to document fruit, and this sheet is a graphite cross-section of a Northern Spy — the core drawn as carefully as the outline, because a written description could not settle an argument about identity and the inside of the fruit was part of the evidence.
The handwriting records everything a modern database would: the variety, an accession number, the grower who supplied it — Wm. Andrews of South Haven, Michigan — the artist, and the dates.

Northern Spy, drawn in graphite by Royal Charles Steadman in 1918 from a specimen supplied by Wm. Andrews of South Haven, Michigan. The USDA's pomological collection is usually described as watercolours, but it also holds line drawings like this one — and a cross-section shows what a colour portrait cannot: the shape of the core.
Specimen from South Haven, Michigan, drawn 9 November 1918 · National Agricultural Library
Royal Charles Steadman, USDA Pomological Watercolor Collection · Public domain
Grouped by what each is for rather than by catalogue order, because the useful question is which one suits your ground.
Best for: Eating straight off the tree
Discovery, Katy and their kind. They do not keep — a week or two at best — which is exactly why they are worth growing: this is a category the supermarket cannot sell you.
Best for: Storage, and the deepest flavour
Northern Spy, Ashmead's Kernel, Newtown Pippin. Many improve for weeks or months in store, which is a thing modern supermarket varieties largely gave up in exchange for shipping well.
Best for: Sauce, pies and baking
Bramley is the archetype and is triploid, so it needs two pollination partners and gives nothing back. High acidity is the point: it survives cooking, where a sweet dessert apple turns to bland mush.
Best for: Fermenting
Bittersweets and bittersharps carry tannin that eating apples were bred away from. Grown in the Northwest again after a long absence, and genuinely inedible fresh — which surprises people who pick one.
Best for: Growing what already works here
The Hood River valley in the photograph at the top of this page has been an apple and pear district for well over a century. Regional varieties and local nurseries are worth far more than a national list, because scab pressure in a wet spring is a local question.
Listed by what you can see, not by what the plant has — you arrive here holding the evidence and without the vocabulary. Where the answer is that nothing can be done this year, it says so.
Why: No compatible pollination partner flowering at the same time.
What to do: Plant a second variety in the same or an adjacent pollination group, or a crab apple, which pollinates almost everything. Check whether yours is a triploid — those need two partners.
Why: Biennial bearing, brought on by letting the tree overcrop.
What to do: Thin to one fruit per cluster in June, four to six inches apart. In the blank year, thin especially hard on any tree that does flower, to reset the rhythm.
Why: The June drop — the tree shedding what it cannot carry.
What to do: Nothing. It is normal and helpful. Thin what remains afterwards rather than before.
Why: Apple scab, which thrives in a wet spring — so, here, most springs.
What to do: Rake up and remove fallen leaves in autumn, since the fungus overwinters on them. Prune for airflow. Choose scab-resistant varieties when planting; that decision matters more than anything you can do later.
Why: Apple canker, entering through wounds, worse on wet heavy soil.
What to do: Cut affected wood back to clean growth and remove it from the garden. Improve drainage. Prune in dry weather so cuts close cleanly.
Why: Codling moth larvae.
What to do: Hang pheromone traps in late spring to catch the flight, and pick up windfalls promptly rather than leaving them as a nursery.
Why: The graft union was buried, so the variety rooted above it and escaped the rootstock.
What to do: Not reversible. When planting, keep the kink in the lower stem several inches clear of the soil, and check it again after the soil settles.
Worked out from the botany rather than from a hand-written list, so it stays correct as the library grows.
Also Rosaceae. Shares much of this crop's pest and disease pressure and belongs in the same rotation group.
Apples do not come true from seed, and unusually the reason is worth understanding rather than just accepting.
The species is self-incompatible and highly heterozygous, so every pip is a genuinely novel individual carrying an unpredictable shuffle of its parents. Plant ten pips from one apple and you get ten different trees, nearly all of them small, sharp and not worth the decade it takes to find out.
So every named variety in the world is a clone, maintained by grafting. Northern Spy is not a type of apple; it is one tree that appeared in New York around 1800, propagated onward ever since.
If you want to propagate a tree you like, take a cutting of one-year wood in winter and graft it onto a rootstock — whip-and-tongue in late winter is the classic method, and it is far more learnable than it looks. That is a genuine skill worth having, and the only way to reproduce an apple.
The one reason to grow from pips is the one that produced every variety we have: chance. Somebody noticed a seedling worth keeping. It takes years, nearly always disappoints, and is exactly how Northern Spy and every other apple came to exist.
We're working through the vegetables one at a time, with dates counted from your own frost station and the old books read properly. Occasional notes from Eugene, Oregon — an art business with agricultural roots, working its way toward farming.
We'll email you to confirm before sending anything. After that, occasional seasonal notes from Sun & Rain Works — we won't share your address, and every email has an unsubscribe link.
Growing advice here is current horticultural practice written in our own words. Where it quotes a book, the book is named and linked. Where practice is genuinely contested, or local enough that we would be guessing, the text says so — your county extension service beats any national guide, including this one, on pest timing.
Photographs carry their author and licence beneath them. Nothing on this page is AI-generated.