Why we grow seedlings alongside grafted trees

We grow seedlings because every seed is a new tree, and the way to find trees fitted to cold, dry Montana is to grow a lot of new ones and keep what lives. A grafted tree is a clone you can count on. A seedling, like our Zard apricot seedlings or Siberian C peach seedlings, is a new genotype: living genetic material, not a finished product.

What a seedling is, and what it isn't

A seedling is a tree grown from seed, so it's a new genetic individual and not a copy of its mother. A named variety is a clone of one original tree, kept going by grafting. Most of our seedlings are open-pollinated, so we know the mother but not the father. Our Fred's Mac seedlings are half Fred's Mac, and the other half came from whatever pollen the bees brought. The mother tree stands next to Liberty and Goodmac.

Apple seedlings vary wildly

An apple seedling is a brand new variety, and it can land anywhere. Apples carry a genetic system that blocks most self-fertilization, so apple seed almost always comes from a cross. McIntosh shows both sides. Seedlings from the original tree were on sale by 1820 and didn't bear fruit as good as the original, and the family later propagated it by grafting. Yet a 1996 study found McIntosh was a parent of 101 of 439 cultivars, more named apples than just about anything, and Luke has found McIntosh seedlings that kept a lot of what makes the original good.

In Montana, we find seedling apples often beat expectations on flavor. When one doesn't, it isn't wasted. Press it for cider, feed it to animals, or top-work it by grafting a named variety onto it.

Stone fruit seedlings are less of a gamble

Stone fruit seedlings are less of a gamble than apple seedlings, for two reasons, though neither is a guarantee. First, many apricots are self-fertile, so an open-pollinated seed lot can hold selfed seedlings, where the mother is also the father, next to crossed ones. Selfing raises the odds that a seedling shares more genes with its mother, which is why apricot seedlings can show more family resemblance than apple seedlings. Second, Luke's working theory: centuries of seed-grown landraces are part of why peach, nectarine and cherry seedlings, and to a lesser extent plum and apricot seedlings, rarely taste bad. It's a theory, not a finding.

They still vary. Bob Purvis describes Manchurian apricot fruit as small and highly variable in quality, and on our Manchurian apricots some trees lean sour while others run sweeter. Our Morden 604 and Westcot seedlings may turn out better or worse than their parents.

The landrace lines we're keeping going

A landrace, as we use the word, is a line kept going from seed, where each generation gets sorted by the place it grows. These are the ones we're working with now:

• Siberian C peach. Miekal of Driftless Sacred Grove in La Farge, Wisconsin has bred this line for 30 years for bigger, juicier, less mealy fruit. His trees took -35F in 2012 and 2018 without winter damage (the flower buds were hurt). Ours are from his fifth-generation selections, and we're carrying the work on.
• Purvis hardy walnuts. English walnuts grown from nuts collected on Bob Purvis's farm in Homedale, Idaho in fall 2023. The mother tree took -20F without injury.
• Doug's Big Red Juicy nectarine. Our friend Doug's tree bore heavily in Missoula for over 10 years and took at least -26F before it was cut down for an alley. We saved it by grafting and are growing out its seedlings.
• SnowCzar almond, our own introduction, still in trial.

How we select: Antifreeze Protein Selection

We let the cold do the first round of culling, on purpose. Luke's Antifreeze Protein Selection method takes two winters:

1. Cold-stratify the seeds by deliberately freezing and thawing them.
2. Plant only the seeds that survive.
3. Leave the young trees in pots through a natural freeze-thaw winter, with no protection.

After two years of culling, apricot seedlings whose roots have taken -25F or colder unprotected get listed as zone 3. The idea is to favor trees that make antifreeze proteins, which many overwintering plants use to keep ice between their cells from growing, the way native chokecherry and serviceberry do.

SnowCzar shows how hard the sort is. We started with 800 almond seeds, 40 germinated after the freezing, and after two years we culled 32 of those, leaving 8. SnowCzar came from those 8 and showed no tip dieback at -25F. When we ran Siberian C peach seed through the same process, none of them died.

Let the site do the sorting

Plant more than you need and let the site sort it out. Mark Shepard of New Forest Farm in Wisconsin, author of Restoration Agriculture, calls his version STUN, for Sheer, Total, Utter, Neglect: "Plant a zillion seedlings and let the losers die. The only ones that can survive on your site will have some sort of resistance trick." The catch, he admits, is that wild seedlings are built for their own survival, which "doesn't always mean producing the most nuts or the biggest fruit."

Systems ecology gets to the same place. H.T. Odum, who coined the term ecological engineering, listed "maintaining biodiversity with multiple seeding" among its tools in a 2003 paper with B. Odum. Donella Meadows wrote that "the ability to self-organize is the strongest form of system resilience." Ernst Götsch plants his syntropic systems as a succession, from short-lived pioneers through to climax species, each stage creating the conditions for the next. Two cautions we'd add: selection only sorts what you planted, so start with diverse seed, and a run of mild winters selects for mild winters.

Good trees are already growing near you

Some of the best genetics for your place are already bearing down the road. Purvis's advice from his apricot talk goes for any fruit: look in your neighborhood for bearing trees, and if the fruit is good, propagate it, at first on Manchurian apricot seedling rootstock for its compatibility and anchorage. Our Florence Plum seedlings come from a purple European plum Luke found in Florence, Montana, itself almost certainly a seedling, standing since the 1980s through winters into the mid -30s. 3rd and Myrtle began as seedlings of a 40-year-old Missoula apricot, and its best performers are now grafted.

Seedlings make rootstock, too. Our Manchurian apricots can be bud-grafted with Westcot, Morden 604 or another cold-adapted apricot, best done in early August.

What you're buying when you buy a seedling

You're buying living genetic material, not a finished product. As we put it on the Zard seedling page, part of the value is in observing, selecting and learning over years. Go in knowing this:

• On its own roots it grows full size. Plan on 18 to 20 feet between apple seedlings and about 15 feet between apricot seedlings, which can reach 20 to 25 feet unpruned.
• Its hardiness listing is an estimate from the parents, and some seedlings will be hardier than others.
• Bloom time and pollen quality vary, so plant at least two.

Our Vermont Beauty pear seedling page says why we bother: "When people grow trees from seed, we allow biology to continue collecting information (through genes) rather than locking it in a biological time capsule."

Related guides

• The orchard as a system
• Growing apricots where winters hit -30F
• Rootstocks for cold country, and why we use ours

Where this comes from

• Our product descriptions for every seedling and landrace line named here, plus our Kerr Crab and Spartan x Mac apple seedlings.
• Our own notes on the Antifreeze Protein Selection method and our landrace grow-outs.
• Miekal, Driftless Sacred Grove, La Farge, Wisconsin, for the Siberian C peach line.
• Bob Purvis, Apricots: A Wider Geographical Possibility (NAFEX conference lecture, 2006).
• Mark Shepard, Restoration Agriculture (Acres U.S.A., 2013), and his posts on permies.com.
• H.T. Odum and B. Odum's 2003 paper on ecological engineering.
• Donella Meadows, her 1999 essay on places to intervene in a system.
• Ernst Götsch's syntropic agriculture, as described by Agenda Götsch and ARC2020.

Updated September 2026.