Over the past thirty years I have worked to understand how it might be possible to see both the design and management of horticultural plantings through an ecological science lens.  I first started with this work in Melbourne, Australia in the early 1980’s on native grasslands, but my research really came to fruition after joining the Department of Landscape, University of Sheffield in 1995. Being in a highly multidisciplinary environment gave me the intellectual freedom and stimulus to explore new approaches to planting and planting design.  I had already developed the idea of designed plantings as ecologically functional communities, irrespective of plant origin, by the late 1980’s. At Sheffield, I went on to apply these ideas to the design and management of real vegetation, to test how such communities actually behaved post-creation.  What survived and what disappeared and why? Step by step this growing understanding was incorporated into the next generation of increasingly refined, designed plant communities.  My work is unusual in that I am equally comfortable dealing with these ideas at  both strategic scales through to highly detailed design scales. Being able to do this creates a much wider range of opportunities for my research and practice. Design rhetoric about nature like vegetation is of little value if not matched by understanding of how to reliably create and manage such vegetation.


An important part of my design work deals with how randomness can be controlled in naturalistic planting to create long flowering and structurally complex herbaceous vegetation, from either sowing seed in situ or planting.  I have developed methods that, even with seeding, allow control of the density of individual species and hence the aesthetic and functional performance of designed plantings as a whole.  I am particularly interested in the design of herbaceous vegetation composed of 2-3 overlapping canopy layers, and the visual effects associated with species with basal foliage and naked emergent flowering stems.

Typically I have commenced work on a new plant community by adopting a biogeographic approach, for example working with the species drawn from a particular semi-natural plant community, such as for example, North American prairie or South African Altimontane grassland.  This provides intellectual connections with more conventional plant ecology. Once the community is fully understood as a designed vegetation, the component species can be mixed in practice with species from other compatible biogeographic communities to create more obviously cultural, but still ecologically functioning communities. For example, prairie plants species work very well with many Altimontane South African grassland species because they are both selected to be burnt in spring by indigenous people. This increase in biological understanding has been paralleled by work into how human beings see the  designed, naturalistic vegetation that I have strove to design. This essentially environmental psychology research has, in combination with my experience of practice led to a deep understanding of how various design elements impact on human preference for designed vegetation.


My professional work with designed vegetation is strongly informed by a lifetime of intense scholarly interest in plants and their cultivation. The distinction between my professional and amateur interests in plants are pretty difficult to establish.  Since being a teenage I have grown and trialled large numbers of species every year, most commonly grown from seed collected in the wild by highly expert professional botanical seed collectors, such as Jim and Jenny Archibald, Ron Rathko, Rod and Rachel Saunders, and Volubec Holubec.  Over a 40 year period this has given me a scholarly understanding of a huge range of plant species with high aesthetic and functional merits but which are currently unknown in designed landscapes. Relatively few designers have an understanding of these new horizons, and how collection location profoundly affects both plant genotype and phenotype, but this has been an invaluable inspiration to me when looking for species for new roles in urban landscapes at a time of rapid climate change.

One of the hallmarks of my work is to look for plants that have no history of use in cultivation in Britain and Western Europe, but which clearly have either functional or aesthetic characteristics that make them potentially useful new landscape plants. It is only by doing this that it is possible to continually re-invent planting genre.

For those who are unfamiliar with the world of wild collected seed and the profound level of understanding of ecology and horticulture held by the collectors, I include some examples of the field notes in the 2008 Catalogue of Ron Rathko, doyen collector of the Western North American flora.  Ron is discussing the morphological and environmental variations associated with various subspecies of  a desert rosette Yucca.   This level of  observation and understanding is incredibly useful in the development of new types of planting.

05-696 Hesperoyucca whipplei ssp. caespitosa syn.  Yucca whip. ssp. caes. Transverse Ranges., San Bernardino Co., CA. 5750'. 50 seeds $3.50. I am offering several collections of this yucca commonly seen on the dry chaparral slopes of the Southern Sierras and Southern California. They have been segregated into 5 subspecies. Which I have followed in identifying my collections. This subspecies can be found at higher elevations bordering the Mojave Desert. Several equally-sized rosettes are crowded into spreading mounds up to 6' across. The glaucous blue leaves are shorter and narrower than other spp. and triangular in cross section. The impressive panicle, reaches 6-8' in height with hundreds of large, white globular flowers. Many rosettes develop each year, allowing the possibility of several inflorescences each year. Pinyon/juniper and chaparral. I expect that the winter temperatures drop consistently below 10°F. Yucca whipplei has recently been returned to its own genus: Hesperoyucca.

05-697 Hesperoyucca whipplei ssp. intermedia Santa Ynez Mtns., Santa Barbara Co., CA. 3925'. 50 seeds $3.50. This supspecies is more coastal in its distribution. The darker green leaves are thinner and much longer forming much larger rosettes. New rosettes are formed only after the main one produces a flower stalk. As a result, only one inflorescence in a year and only 1-2 rosettes develop every year. The dense panicles reach 8-10' in height. Edges of coastal chaparral. Sandstone outcrops. Substantial winter rains, 40-60" is not uncommon. Freezing temperatures are frequent in the winter, however, the temperature rarely drops below 25°F.

05-699 Hesperoyucca whipplei ssp. parishii San Gabriel Mtns., San Bernardino Co., CA. 7525'. 50 seeds $3.50. Appearing like a white candle-flame on the chaparral slopes of Southern California the dense panicle of this subspecies can reach up to 15'. Here, at the upper elevational range, the scapes are a modest 8' tall. The blue-green rosettes with broader, stiff leaves are up to 4' across. Off shoots are rarely produced. S-facing slopes with manzanita in a big-cone Douglas fir/Coulter pine woodland. Heavy winter snows are frequent here. The winter temperatures are not as severe as at the ssp. caespitosa location.

05-700 Hesperoyucca whipplei ssp. percursa San Rafael Mtns., Santa Barbara Co., CA. 2850'. 50 seeds $3.50. A rhizomatous form slowly developing large patches of rosettes of various sizes and ages from 6" to 4' across. The older rosettes flower and die only to be replaced by more the next year. The rhizomes are about 1" thick with the offspring appearing within a few feet away from the parent. The stiff glaucous blue leaves are longer than ssp. caespitosa. The flowering panicles are only 6-10' in this subspecies. This population of about 100 individuals is more than likely one clone covering about 5000 sq. ft. W-facing serpentine outcrops with Eriogonum fasciculatum. Not restricted to serpentine. I have encountered populations further north, in the Diablo Range, on sandstone.

05-701 Hesperoyucca whipplei ssp. whipplei San Jacinto Mtns., Riverside Co., CA. 5175'. 50 seeds $3.50. The typical form of Y. whipplei is the most southern subspecies. This population is at the northern distribution of ssp. whipplei. Like ssp. parishii, it too only produces a single rosette. The flowering stalks rarely exceed 8'. The leaves are a pale yellowish green to dark green. They are narrower and more flexible than ssp. parishii. Red shank chaparral with a few scattered four-leaved pinyon pines. Loose granitic soils.

Some of the genera and species that I have worked with to try to find a meaningful role for in contemporary designed urban landscapes are shown below.  Castilleja are a largely North American genus of semi (hemi- to ecologists) parasites, with brilliant, normally orange or red bracts.  I initially wanted to see if they (and related genera such as Pedicularis could be used to reduce plant biomass in designed meadows). They cannot unfortunately do this, but they are valuable flowering plants in their own right.  I have started to use Castilleja integra in dry meadow-like plantings established by seeding.  I have undertaken much work on wild occurring species Penstemon (Hitchmough and Khaliq Khawer (2013) Long-term survival of Penstemon in northern England, The Plantsman, 12,4.

Staying with the dry theme, is Salvia pachyphylla, the Mohave Sage, an incredibly good shrub for very dry soils is again becoming a characteristic element in my work. I am currently working with some of the highly xeric Turkish Salvia for use in rubble substrates.


Salvia pachyphylla in August in the authors green roof microcosm experiments in Sheffield

Close up of the bracts of Castilleja integra

The same species sown as part of a Penstemon dominated dry meadow

Some plants I work with are essentially unknown in cultivation in Europe, for example Geranium brycei, a sub-shrubby lavender blue species that grows in Drakensberg Grasslands in South Africa.

Penstemon barbatus and Penstemon cobaea in dry meadow plantings in the new meadow borders designed by the author at the Oxford Botanic Gardens.