The self-sowing of plants in borders has always fascinated me – and increasingly I’m inclined to think that it is crucial to the long-term stability of naturalistic plantings. There is a real thrill in suddenly seeing the seedlings of desirable plants popping up in the border.
Most of us are familiar with those plants which always seem to produce seedlings, like aquilegias and verbascums. The latter are a good example of self-sowing as too much of a good thing – how many of us have frantically hoed off thousands of their tiny seedlings, alarmed at the prospect of a take-over bid by big furry rosettes. Aquilegias though are generally better behaved, their behaviour illustrating what many of us welcome about self-sowing – the spontaneity of desirable plants occasionally popping up of their own accord.
A lot more species do self-sow however, indeed in theory any reasonably genetically diverse plant population should. Many garden plants though are not ‘genetically diverse’ but genetically identical clones (i.e. cultivars), which don’t self-fertilise. Even if plants produce viable seed, the likelihood of the seed germinating does seem to be contingent on soil conditions – but what those conditions are – well who knows? The unpredictability of self-sowing is one thing which makes it so fascinating. In theory, self-sowing is more likely on lighter soils, but then there is the case of a friend whose heavy clay produced remarkable crops of seedlings of just about anything. And then there are my hellebores – my last and present gardens are both on Old Red Sandstone, although this is quite a varied geological formation; in the last garden there was virtually no self-sowing, but in my new garden, almost every seed which hits the ground turns into a seedling. Most have to get hoed off!
One ‘rule’ of self-sowing is the inverse relationship between lifespan and seed production – the longer-lived the plant is, the fewer seeds it produces (and very often the slower they germinate). Short-lived plants put far more resources into seed production, and those seeds tend to be rapidly-germinating. The reasons are pretty obvious – short-lived species need to make sure they leave plenty of youthful replacements around to keep the species alive, long-lived plants don’t need to, and producing lots of seeds might even be counter-productive, taking resources away from more effective methods of reproduction in a competitive environment, like producing running roots or new shoots.
If things go well, seedlings of desired plants fill gaps, producing a steadily denser plant community, which helps to limit weed infiltration, and is probably better invertebrate habitat. A dense plant community, with a near complete canopy is far more ‘natural’ than the traditional border with big gaps between plants at ground level – even though there may be no gaps at foliage level. Self-seeding helps to produce a nature-like visual continuity; in my last garden Geranium sylvaticum ‘Birch Lilac’ self-fertilised and spread everywhere; I didn’t know this was going to happen, but the results were delightful, a continuous drift of purple in May. Ideally, several species will self-sow, so that one does not dominate, and a relative balance develop between them.
Self-sowing is a chance to see natural process in action, a sharing of the design and management of the garden with the energy and life-process of the plants themselves.