Silverweed Musings

A more free form revisit of past impetus

For convenience, here is an index of my Silverweed related content.

---

I try to keep my substack approachable, not least because I myself am a newbie enthusiast to plant breeding (and botany) in general, but this entry is more free form and information dense and will revisit some of previous assumptions updated with my current understanding. 

I am attempting to domesticate Argentina anserina syn. Potentilla anserina (Common Silverweed, or just Silverweed) using two methods: landracing and intergeneric hybridisation.

Can we get Silverweed roots to carrot-like thickness? Only one way to find out!

In any case, let’s drill into what I’m doing here. Common Silverweed has a denser starch and nutrient content than potatoes. They propagate themselves via stolons, basically like strawberries, and wherever they root into the ground, there will inevitably be a long pencil thin storage root. This is the edible part. Silverweed also propagates itself via seeding, and this mechanism should not be underestimated as I’ve had this happen in my own garden after just one year of breeding. From my observations, Silverweed will grow nearly anywhere, but prefers ground that has a certain degree of moistness. You will see mention of Silverweed preferring seasides, but my experience is that it prefers to be near any body of water. Locally, I have seen it grow near reservoirs, canals and seasides. If you bring up the plant distribution on iNaturalist, you will accordingly find very little presence in certain regions that are known to be arid, like most of Australia, or large parts of the Mediterranean. In the US, there is a “sister” plant called Argentina pacifica or Pacific silverweed which is famous for being harvested by the First People’s of America. For more information about A. anserina, see the late Gordon Hillman’s excellent entry on the matter. 

If you’re wondering why Common Silverweed was switched from the Potentilla to the Argentina family, this was a result of genetic and otherwise unused morphological analysis revealing that a certain group of Potentilla were pretty distinct from the others, P. anserina and pacifica being two but there are 64 species in this family - though not everyone accepts this distinction. I would love to get my hands on some of these other specimens. Kew Garden’s has a more readable list. A fair few of these species are considered by some to be a subspecies (I think basically a cultivar) of P. anserina, and most don’t have common (english) names. Arctic Silverweed is a lovely common name for a subspecies of Common Silverweed (Argentina anserina subsp. groenlandica) and must exhibit some extreme adaptations to be endemic to Greenland.

Taxonomy tangent aside, wild A. anserina is extremely nutrient and carbohydrate dense - but there’s one sticking point. It’s hard to harvest and hard to clean. The storage roots reach deep and you’ll be hard pressed to not snap them midway if you’re harvesting in the wild. I grew mine in pots last year and was able to side step this issue, but some sources mention growing silverweed in loose, loamy raised beds. Once you’ve got hold of the storage roots, which will have lots of stubborn, hair-like roots attached to them, you’ll have to clean it. It’s hard work because the hair-like roots cling to soil and stone very strongly. I assume that these secondary roots are responsible for any nutrient gathering and the storage root we’re interested in is the result of their hard work. One nice benefit of these storage roots is that the specimens travel well. I’ve had tiny fragments of storage root regenerate. The secondary hair-like roots burn easily when it comes to cooking too (they also stick in your teeth), which makes preparation and consumption of Silverweed so burdensome that it’s got an - undeserved! - reputation for being a famine food.

So I hope I’ve led you to the conclusion that Common Silverweed deserves to be domesticated. And when I say “domestication”, I mean in much the same way as the humble carrot was domesticated from the wild carrot: fattening the storage root and reducing the hair-like roots. Silverweed has something going for it that the carrot does not though, which is the self propagation. What some view as the “weedy” property of Silverweed can be harnessed to great good - just picture a bed of self propagating Silverweed; minimal effort, maximum gain, pulling up wheelbarrow loads of carrot-thick Silverweed come autumn-time.

Now the dream has been cast, there is an obvious way to achieve it. We can selectively breed Silverweed, crossing only those with the thickest storage roots. In time, we will tend towards thicker and thicker storage roots and we’ll be able to “stabilise” the genome into what would probably become a new species, Argentina megarhyza (thanks Nancy!) or maybe Anserina domestica. In practical terms, this is how any plant has been domesticated for food use. Some genomes probably have less latent morphological adaptability than others, but then who has really plumbed the depths of every genus? Brassica oleracea, as one popular example, has been forced to produce cabbage, broccoli, cauliflower, kale, Brussels sprouts, collard greens and on and on. These wildly different forms are all technically “cultivars” of the wild cabbage. So perhaps our domesticated Silverweed will just be Argentina anserina var. Megarhyza.

I work a full time job - writing about food forest adjacent subjects and plant breeding projects must fit around that, so I’ve opted for a low effort “open pollinated” breeding program. What this boils down to is gathering as many specimens of Argentina anserina as I can from around the world, allowing pollinators to cross these specimens and growing on the seeds until I have a nice healthy and wide gene pool. Once I have broadened the gene pool as much as I can, then intensive, selective breeding can begin. Why do this open pollination step first? When we domesticate a species (be it animal or plant) inbreeding and back breeding is used to encourage traits that we want. We might find some progeny in the third generation that has excessively many hairy roots, but also has big storage roots too, so we’ll cross it with a second generation child that has slightly smaller storage roots but less hairy secondary roots, and so on. When this happens, the resulting children are weakened, so we want to start with as vigorous a gene pool as we can. Silverweed is eminently suited to this scattershot approach because each plant is self-infertile, so we do not need to worry about inbreeding of vegetatively propagated specimens. This means that we can be pretty sure that the seeds are always a healthy recombination of genes from the round-the-world specimens that I’ve gathered. I’m not sure I would have begun this quest if they were self-fertile; think of the extra administrative overhead!

If we think further along these lines, we might end up at crossing A. anserina with other members of the Potentilla family. The same logic applies; by having a deeper and more variable gene pool, any domestication efforts will hopefully retain a high degree of vigour. Crossing two genera is also known as intergeneric hybridisation. I owe a lot to Shane Simonsen for pointing me down this path. This type of crossing is criminally under-researched - there is so much scope for variety and vigour - but there are famous examples like Leylandii which is a cross between the Monterey cypress (Cupressus macrocarpa) and the Nootka cypress (Cupressus nootkatensis). The Leylandii exhibits remarkable vigour but are generally thought to be sterile (like a mule is generally sterile) - though I would be willing to bet money that a serious program of seed germination would result in viable offspring from which a new “species” could be stabilised - and the same would apply to the mule. In fact, any new “cultivar” of Leylandii is really just different genetic combinations of the host parent whenever they are cohabiting.

I am but one man, and there are many many (500, actually) Potentilla species. I attempted to filter the easily available species, and narrow them down by interesting traits (such as extreme cold adaptation) and or root edibility. Bonus points if the plant has both types of traits. The hope is that we will have the deepest possible genetic pool from which we can narrow down to desirable domesticated traits.

When I speak of intergeneric hybridisation, I really do mean between distinct genuses. Luther Burbank successfully crossed Malus domestica (the humble apple) with blackberry (Rubus) over five thousand times. If you think about how different apples and blackberries are then it becomes more plausible that two plants from species that were only recently split - potentilla and argentina - could produce viable offspring. Stabilisation of these offspring implies that there is morphological consistency from one generation to the next and it is arguable that this is how to create a new species. If we trace back the scientific taxonomy between apple and blackberry, we arrive at the common Rosaceae family and that is actually the commonality between Silverweed and apple or blackberry and it’s a fun thought experiment to think about what the offspring would look like. The blackberry apple hybrid was reportedly thornless, apple tree in form and bore pink flowers. Imagine a vegetatively propagating apple tree!

Steering back to the task at hand, we could have a guess at, for example, the offspring between A. anserina and P. nepalensis (Nepal Cinquefoil). They are both similar in form and function, and both have roots that are allegedly edible. Nepal Cinquefoil root is meant to be an effective red dye so that would be my major guess. I’ve spent quite a few hours researching prior work of hybridisation between Silverweed and other Potentilla species and to my joy discovered that someone has recorded a couple of attempted crosses. P. recta x A. anserina (two failed attempts) and P. argentea x A. anserina (one failed attempt). At least someone tried, but not nearly enough in my opinion to show that it’s impossible. Luther Burbank, the famed plant breeder, would attempt thousands of crosses - a paltry couple will not do unfortunately.

Right now, I have two parallel routes towards a broader gene pool, before “domesticating” to thicker roots. I likely do not have the land (or bandwidth) to bring about this full domestication and this is where you, dear reader, might step in. A group of people located around the world, using their free time to grow and select thicker and thicker Silverweed roots, could rapidly domesticate this humble plant. There is precedent for this, with a (currently paused) citizen science project attempting to remove the daylight dependence of Oca (Oxalis tuberosa).

It’s still early days for Silverweed though. I have the first generation offspring of these mixed Silverweeds, and I hope by this autumn I will have - thanks to some hardworking pollinators - a decent batch of seeds. When that time comes, I will send a call out for some willing individuals.

Together we can bring Silverweed to more dinner tables.