Every garden plant looks back to a wild ancestor — sometimes several. Some are separated by lineages extending for thousands of years, and others came in from the wild only yesterday. Cultivation changes plants. No plant that produced tightly sheathed and impenetrable cobs, as cultivated maize does, would survive in the wild; the development of cultivated forms from the small, loosely wrapped heads of their wild ancestors is a long and fascinating tale. The size, diversity of
shape, and range of colour of garden dahlias and roses exceed beyond imagination any variations to be found among their wild relatives. One of the world's major staples, wheat, is an intricately complex hybrid between three species from two genera. Heads of wheat, as seen today, are unrecognisably different from those of their wild forebears and a true example of genetic engineering — though whether by divine intervention or nature's happenchances, who can say?
These changes are skin deep. They are not necessarily, or even usually, accompanied by extensive changes in the ways the plants work or respond to their surroundings. The stratagems by which their wild ancestors survived persist through generations in cultivation often with little or no change, including hardiness, responses to day length through which plants control the seasons when flowers and fruits are produced, the conditions in which seeds do or do not produce seedlings, and the ability of shoots to form roots when attempts are made to propagate them from cuttings.
Propagation is concerned with how plants work — not what they look like. Cultivated dahlias with peaches-and-cream petals on flowers as big as dinner plates may not look much like their wild ancestors, whose faded pink flowers fit easily into the palm of a hand, but both are equally sensitive to frost; both produce tubers, flowers and seeds in the same way; and as far as propagation is concerned, what works for one works equally well for the other.
Recently developed techniques ofgenetic engineering (transgenics) by which genes from unrelated organisms can be incorporated in a plant's genotype may fundamentally change the responses of some, and eventually many, of the plants we grow. One day we will fill our gardens with hardy kinds of petunias and cease to worry about spring frosts — but for many years to come, the best guide to the conditions most of the plants in our gardens depend on to flourish and reproduce will be the responses that enabled them to survive in the places where they grew naturally, inherited from their wild ancestors.
No plant lives forever, though some make impressive efforts to do so, and under natural conditions a species survives only when and where it can reproduce. Reproduction is a risky and uncertain process, during which the offspring must separate from their parents and succeed in establishing themselves elsewhere. Plants have two basic choices: to reproduce sexually from seed or asexually through vegetative reproduction.
When plants ceased to be free spirits, opting instead to live rooted to the spot, they had to find solutions to the problem of having sex without being able to move around in search of partners. Water provided a way for mobile niale cells to reach nubile ovules, and for eons mosses, ferns and similar plants depended on complex life cycles to make that possible. It was, and is, an uncertain process, limited by the availability of water and more likely to result in self-fertilisation than in cross-ferrilisation, but until something better turned up it made gene exchange and the benefits of sexual reproduction possible.
Pollen and seeds were the great invention of the gymnosperms, transforming the efficiency of sexual reproduction. Pollen carried from one plant to another on the wind provided opportunities for cycads and later conifers to matchmake over considerable distances. Seeds enabled their offspring to be distributed across the countryside by the wind, animals and birds. The production of brightly coloured fleshy arils, like those found on yews and podocarps, heralded the invention of succulent fruits, increasing the attractions of seeds to birds and animals.
Then, more than 100 million years ago, angiosperms refined the process by producing conspicuous, brightly coloured, often intricately constructed flowers to attract pollinators. Alliances with birds, animals, insects, snails, the wind and a dozen other agencies led to an astonishing variety of ways of ensuring pollination and seed distribution. Such improvements should have consigned mosses, ferns and other less sophisticatedly endowed plants — even conifers — to obsolescence and oblivion. Failing to get the message, they continue to coexist and compete effectively with the flowering plants.
Reproduction by seed is a sexual process as a rule, though not always. It ensures every plant in a population is an individual with its own particular genetic constitution. The differences between individuals provide the diversity that enables populations of plants to adapt to changing circumstances by selection over the generations. However, the odds against any seed growing up to produce flowers and seeds tinder natural conditions are not ten to one and seldom even as favourable as a thousand to one. They can be tens or hundreds of thousands, not infrequently millions to one against survival, and a population of plants survives under natural conditions only for so long as it is able to produce so many seeds that some individuals still remain after all the agents of death and decay have done their work.
A single sallow, or goat willow tree (Salix caprea), in a single year can produce sufficient seeds to give rise to 1300 seedlings per square metre (1100 seedlings per square yard) over an area of some 3 hectares (7.5 acres) — about 15 million seedlings in all. That same tree may enjoy 20 or 30 fruitful years during a lifetime in which it produces 200 to 300 million seeds. And the result of this enormous output? On average, just two trees, one male and one female (because this is a dioecious species), grow up to maturity, bear seed themselves and maintain the population.
Willow seeds have brief lives. Their minimal storage reserves sustain life only for a few days, and they make minimal demands on the plants that produce them. Most species provide more generously for their offspring before launching them into the world — correspondingly improving their chances of survival — so they do not need to produce so many. Nevertheless, extra provisions put greater strains on the resources of the parents. Balances must be struck between producing
millions of poorly endowed seeds with almost infinitesimal chances of individual success or producing fewer, better endowed seeds with correspondingly better prospects.
The carbohydrates, oils, fats and waxes with which seeds are stuffed are energy-rich foodstuffs, expensive to produce and highly sought after by birds and mammals, insects, bacteria and fungi. Larger seeds provide greater support for seedlings but are even more desirable to predators and easier to find than smaller ones. Small or large, numerous or few, there is no clear-cut best option in the equations linking seed size and numbers with reproductive success.
An alternative strategy is to produce genetically better tuned seeds so that higher proportions establish themselves successfully and grow into mature plants. Pursuing this strategy has led to some curious results — even to second thoughts on the value of sexual reproduction itself.