Let me introduce you to one of my garden characters, one with pros and cons depending on where you find it. Soursob is that weed that looks like it has clover leaves – but isn’t a clover. Instead it has bright, almost fluorescent yellow flowers on thick juicy stems, which are well known to children around the country as being great to eat. In my experience adults find them not nearly as good as we remember. The sourness comes from oxalic acid, something not great to consume in quantity but in the stems and leaves it’s fine. It’s used in a traditional popular water-flower stew in its home country of South Africa, and I’ve tasted cordial and brewed lemonade flavoured with it.

Oxalis flowers and leaves. The grassy weeds are trying to get through with only minimal success.

Oxalis is one of those really pesky weeds as far as bush rehabilitators and landcare groups are concerned. Once it’s in an area, it’s *really* hard to remove. It doesn’t usually set seed, but underground it makes a white fleshy tuber that has many bulbils attached. When drought triggers the die-off of the plant, that fleshy tuber dehydrates – but not all at once. It’s faster from one end, causing the tuber to slew sideways – and spread its little bulbs just a few centimetres further. Weeding by pulling up the plant often leaves several if not all of the bulbils behind. So once you have one plant somewhere, the next season you have many. It can’t spread fast on its own as a patch can only expand as far as the sideways drift of one of those tubers, typically around 10 cm. But if any soil moves from an area that’s had oxalis in it, you can guarantee that at least one bulbil’s come along too, and that means you’ve got a new area of infestation.

In the garden, oxalis isn’t so bad. It doesn’t usually get higher than 25 cm, often not that high. And it’s a beautiful spring green, with lovely reliable flowers. The effect en masse – and it’s always en masse – is of a lovely meadow. In that regard I don’t mind it. It acts as an iron concentrator too, so mowing or cutting it and using the leaves and above-ground stalks to make green manure or fertiliser tea is worthwhile (and safe, as you can’t spread it that way). The problem is first that it’s hard to contain when soil moves around a garden – as it does in muddy boots and children’s digging pails, let alone when I’m actually trying to reconfigure the microgeography – and second that it grows so thickly it’s very hard to get anything else growing in its patches.

Removing it is notoriously difficult. There is a golden period, somewhere in May or June, that lasts one to two weeks, when the bulbs have sent up their plant stalks and exhausted their resources but have not yet formed the next year’s crop of bulbils. If you spray the plants then, or gently and thoroughly lift them out with a digging fork if you’re avoiding sprays, you can knock the infestation over completely.  Outside of that period, you get partial knockouts but you don’t get them all and maybe even none. The problem is knowing when that period is. If you can see flowers, it’s way way too late.

Last year I tried removing all the sand from a few sections of the garden to a depth of around 30 cm, and sifting it with shadecloth back into position (adding compost and sand-to-soil conditioner at the same time to save double-handling). It was a lot of hot, heavy work and I only got two of three sections finished. The results were noticeable – I really cut the numbers of plants down – but it wasn’t enough. A few bulbs remained, which have thrived this year despite my attempts to weed them, and next year it’ll be hard to see where I sifted. The third section stayed open to the sky, sand piled around it – and to my surprise, it was as thickly full of oxalis this year as it had been last year. So, in these sands-I-mean-soils, the oxalis tubers can leave their bulbs deeper than 30 cm. It makes digging not seem terribly worthwhile. The other problem was of course that in sands like these “you can’t dig a small hole” – I had a lot of sand (and bulbs) sliding back into the pits as I dug. So the first area, a narrow trench which I cleared quite well, despite my best efforts had a lot of side infiltration before the trench was refilled.

The three sections I tried to physically dig and sift all oxalis bulbs from are all in this picture – but not visible enough to see where they are (tip: there’s lots of yellow flowers in each)

I’m reluctant to use herbicides for several reasons, and in the case of oxalis knowing that it’s a wasted effort unless I hit that window has stopped me giving in and trying poisons anyway. And digging it out just moved the bulbs to new spots without clearing the old spots. So this year I’ve tried two new tactics.

The first is just mowing it – there’s a lot of garden area I’m not doing much with anyway yet so I don’t need it cleared. My theory is that I can use the oxalis as a green manure and return organic matter to the soil (which desperately needs it), reduce its vigour and hopefully the concomitant number of bulbs formed by regularly stealing its energy, and otherwise just enjoy its meadow effect.

Oxalis meadow. In front is a mowed swathe of oxalis, behind is uncut.

The other is the blanket approach. And I do mean blanket. I’ve put down cardboard, a tarp, an old leaky ex-air mattress – and I’ve left them there for the oxalis growing season. Those spots have received no light and virtually no water all winter, so while some oxalis has attempted to sprout it hasn’t been able to do much. I’m doing this because one of the sources I read mentioned that oxalis bulbs have no dormancy – they sprout the next year or they die. I am hoping that this means I can knock out large swathes of it at once by blocking it for a year. I can’t use this method in areas I’ve already planted up, unfortunately, but I can use it in areas I plan to plant. The downside of this approach is it costs me a year in planting and soil reparation – but if it works, it’s worth it to clear an area before planting.

If it doesn’t work, my strategy next autumn will be to try the window approach. Get an old glass window, and when the oxalis begins to sprout, lay the window over it for six to ten days until everything under the window is scorched. There’s probably enough sun to make that work through March and April. It can’t get any bulbs that haven’t sprouted by that time though so I’m less hopeful for that method.