The difficulty of color-driven landscapes

The most eye-catching part of a garden plant is its flower, and the most captivating element of a bloom is its color. You might think then that designing a garden should be an exercise in painting with flowers.

This idea once held a lot of sway, but color-driven garden design is, by and large, a dead duck.

Gardeners today are more relaxed about their plantings and are driven less by color schemes than the desire for naturalistic effects. We are still drawn to flowers and have our own color preferences, but the need for elaborate, color-coded borders has generally vanished.

There are ways to pinpoint plant color - the most famous is the Royal Horticultural Society Color Chart, essentially paint charts with holes in them for matching chips directly with a flower - but I have never seen a gardener in the United States use one.

This retreat from overt color design doesn't mean that we should abandon our interest in color theory. Every gardener needs to know how color works.

To that end, we mark this week the Smithsonian's publication of "Werner's Nomenclature of Colours," a reproduction or facsimile of an 1821 manual that is slender physically but a giant in its significance. It was devised by a Scottish art teacher named Patrick Syme and based on a system of color classification by a German mineralogist, Abraham Werner.

This book standardized the color descriptions of scientific specimens in a pivotal era of discovery. One of its users was Charles Darwin.

But color systems are needed by artists as well, and by the end of the 19th century, color science had made the leap from botany to horticulture, most famously with the work of the Arts and Crafts garden writer and designer Gertrude Jekyll. She started out as a painter but turned to gardening after her eyesight deteriorated.

While Claude Monet was capturing his garden on canvas, Jekyll was turning her unrealized paintings into gardens.

She put together planting plans for borders of hot colors and cool colors. Her favored approach was to compose a plant border that started with cool colors, moved to hot ones and then receded to the cooler ones.

This coherent artistry had great appeal and was adopted on both sides of the Atlantic. One of Jekyll's admirers was Vita Sackville-West, whose renowned garden at Sissinghurst Castle, south of London, includes a white garden aped in private gardens around the world. I prefer Sissinghurst's Purple Border, which, as I recall, is a medley of reds, pinks and blues as well as purple, all set against a high brick wall.

The desire to group plants by color is thrilling when done well, but it leads you into a maze - you must master color theory before moving on to high-level gardening.

First, the theory. If you've taken an art class, you know that the appearance of a color is controlled by three components: hue, brightness (or value) and saturation.

A pastel color - seen in a pink Oriental poppy, perhaps - has high value and high saturation, making it light and bright. The pale color of a blushed peony has high value but low saturation. The rich color of a crimson gallica rose has low value and high saturation. This is explained in a book by the late Sandra Austin, who was an instructor of landscape design at George Washington University. "Color in Garden Design" was published in 1998 but still can be found online.

Austin hoped that if gardeners understood the technical attributes of color, they could use it more effectively in the landscape.

But mastering color theory is one thing; having the proficiency to create a season-long color-coordinated garden is something else.

Even if you include foliage as part of the color plan, as Austin suggests, you'd still need an encyclopedic knowledge of plants and how they grow in your garden. Sorry, but you can't Google that; such knowledge takes years to accumulate through trial and error.

Another factor working against color gardens is the mid-Atlantic climate, which is colder in winter than England and certainly a lot hotter in summer. This alters the plant repertoire. You can't just crib a planting scheme from an English book.

I can think of a few instances where color-driven gardening still commends itself. The first is in garden areas of light shade, where you could put together plants in considered shades of green and white with a little blue thrown in. Foliage color would be a major element. I might suggest various hostas and ferns, grasses and sedges, Satsuki azaleas, smooth hydrangeas, fothergillas, sasanqua camellias, the native fringe tree, foamflowers, wood asters, foxgloves, Japanese anemones, rue anemones, white varieties of wood anemone and Grecian windflowers, and lots of little white daffodils followed by Virginia bluebells.

Start with containers

The easiest, cheapest color playground is the container, where you can pick long-flowering annuals and tropicals that conform to a given three-or-four-color scheme (or a single color).

Another simple way to play with colors is to mass-plant three or four tulip varieties in a considered color scheme. The show lasts for only a couple of weeks, but it's a delightfully luxurious way to celebrate the arrival of spring.

How should you piece together a planting plan? It is far more satisfying to compose gardens in terms of textures, forms, heights and blocks of plants rather than color. Such compositions still pack a flower punch, but they aren't reliant on a constant floral parade for effect. Besides, there are times when the color wheel and rules about complementary and harmonious hues seem irrelevant. Color combinations often take care of themselves, and there will be happy accidents. I am thinking of a tulip named Dordogne, which by rights should be a gaudy disaster, marrying a peachy orange ground with a flame of bubble-gum pink. It looks fabulous.