From the archive…

Permaculture: a story from the dawn time

In the interests of documenting our shared permaculture story, I’m reprinting a story first published in Grass Roots magazine, edition 13, Summer 1977. It was credited to David Holmgren and Bill Mollison.

Grass Roots (and Earth Garden magazine, of similar vintage) grew out of the alternative lifestyles subculture that emerged in Australia in the late 1960s and blossomed to fullness in the 1970s. The magazines informed a youthful subculture seeking better ways to live, a populous subculture that fed into the early permaculture design concept.

I found the magazine in a secondhand bookstore in Campbell Town, the largest town on Tasmania’s Midlands Highway, while on an unplanned stopover.

The text, from permacuture’s dawn time of the 1970s, is reproduced as it appeared in Grass Roots with only minor modifications to improve readability.

This is Part 1 of a two-part article. Part 2 is unavailable.

Reproduced with permission.

Today’s fruit, vegetables and flowers are weaklings when compared with the origins of creation. Without man standing guard, his developed strains would quickly be erased from the face of the earth.

Annual cropping is humus depletive and energy expensive, and to my mind leaves much to be desired.

Why grow grain and lettuce if we can farm grain trees and lettuce shrubs? Can we establish a food-plant system that allows planning for a thousand years ahead? Where are the walnut forests of tomorrow? If we are able to establish a permanent tree crop culture, what are the implications for the waste society and an economy based on quick return profits and the maintenance of disease?

Looking around at the lawn-rose culture of suburbia, and the neglect-erosion of the outback, we can see the morality of land use.

The time is coming when. once again, man will have to rely on tree crops for his very survival.

Bill Mollison.

1: Some theoretical considerations.

The tree-crop concept is an advance on conservative and traditional agriculture, but is limited in that it considers only planted dominant species, trees, as the productive unit over sub-dominants.

We have devised, and are practicing, a more complex cultivated forest of three or more tiers, integrated with domestic and useful wild animal species, in an involving and dynamic perennial system, for which we coin the name ‘permaculture’.

Species chosen for the system have the following basic criteria:-

  • they are useful to man, or to his animals as fodder
  • they are self-perpetuating, either as perennial plants, self-seeding species (borage), replaceable from harvest (oca), or self-breeding (most animal species)
  • they are suited to southern Australian temperate regions, more specifically to foothill Tasmania (cool Temperate).

The system is being designed by one of us (D.H.). (ed: David Holmgren) as a study for the T. C. A. E. Hobart (ed: Tasmanian College of Advanced Education, Hobart, later absorbed into the University of Tasmania), and both authors are evolving the system on approx. 1ha. (2.5 acres) at Strickland Avenue, Hobart.

In a total cultivated permaculture, dominant species (oaks, black locust, walnut) are planted over sub-dominants (hazel, feijoa, almond) and understory (berry fruit, vines, smaller guava), and the system closed with a herb layer (lavender, comfrey, borage, strawberry, root crops) at ground level.

Edge, aquatic, rock crevice and bog species are selected for specific habitats, and the whole assembly integrated by means of fencing and other structures for animal use as evolution proceeds.

The permaculture system is designed for self-sufficiency without heavy annual input of energy beyond that available as manpower in the community which it is intended to serve (family, farm, village, town). Initial establishment energy is not intended to be greater than for annuals raised under the bio-dynamic system. After establishment, the energy functions turn to control (pruning, cutting) and gathering-storage functions.

Unlike agribusiness, the aim is to distribute, rather than concentrate, the season of yield. Unlike commercial operations, the aim is to diversify product rather than to simplify it. Unlike crop farming, the aim is to produce stable self-regulating systems (as in some pastures, some forests) and, unlike any other system, the aim is to select species to suit all niches and to buffer short and long term climatic change by the selection of species at either end of the hot-cold, wet-dry tolerance limits for the area selected.

Man has used plants for oil, fibre, construction, music, cosmetics, utensils, medicine, fuel and forage (to limit the range). He obtains fats, fibre, sugars, starches, proteins, minerals; in short, complete food needs, from plants — as do his domestic animals. We have tried to consider all essential needs of a group in selecting species.

There are some basic ecological concepts to be considered, and perhaps the primary one is: that a complex system, in terms of structure, species and varieties, is more stable and in general higher-yielding than any simple system.

Another: that a complex system, in terms of structure, species and varieties, is more stable and in general higher-yielding than any simple system.

Another: that a complex system with ‘edge’ (or interface) as pasture-forest, water-land, pine-eucalypt, is far more productive than a system with little or no edge.

Another: that any system evolves; some species compete and banish others, while until plants have established, sheltered niches may not be available.

Another: that species do not necessarily ‘compete’ — they may in fact assist or add to others; eels will exist with trout, companion plants interact to aid each other. It follows then, that the ‘best’ oil tree or nut tree for your area should be more commonly planted but that all others (if there is room) should also be planted.

Finally, we doubt that a fully developed system with all possible species could be evolved in less than about 5 ha (2.5 acres), but the smallest plot of land presents opportunity for some application of the above principles, using only some species.

Now, we may have more to say on theory later, but for the present what did we do?

  • we selected, from the world literature on ‘useful plants’ and animals, a list of species for all uses suited to our area
  • we located most of these growing locally, or sent overseas for seed and specimens via sources such as botanic gardens, international seedsmen, and friends
  • we designed a permaculture using ecological principles, some of which are stated above
  • we commenced to put it into practice, having sorted our selected species into a variety of classes.

While so doing, we stumbled on accounts of some ancient low-energy systems which have survived to the present day as simple permacultures, as in the ‘cork-pork’ forests of Portugal and the date-melon-goat assembly of oases. All showed great stability and complexity with just local species to build on.

We have the world’s plants at our call.

Consider ‘dominants’ alone — large trees of almost unlimited life, often slow to yield, but increasingly bountiful as they age.

  • walnuts and black mulberry, in Tasmania, survive as early examples
  • the oaks (620 species) have supported pigs, turkey, pigeon and man for centuries and provided the only bread of many men in the middle ages
  • olives, ancient and hardy, provide light cooking oil, food and timber, as do many unconsidered trees of non-European origin
  • many pines (at least 20 species) give edible nuts, oils, timber and fuel (as cones) annually
  • we have all the fruit trees now cultivated or wild
  • several trees (black and honey locust, coprosma, osk, carobs, for example) rain down seed or sugars for animal fodder. And so on.

But looking lower down, the understorey species are just as valuable. Hardy perennials such as the guavas, vines, ‘small fruits’ give a similar variety of yields, and the herbs of the lowest layer repeat and elaborate that yield, with underground tubers or rhizomes for variety, and fungi as a bonus.

2. Design In the system

Selection of species is a prelude to arrangement in a specific design. We could consider a mandala of earth with man’s habitation at the centre, favoured animals and delicate species kept close and hardy, self-sufficient species at a distance.

There are other considerations.

From the north, comes fire. From the south-west come icy winds. From the east and north, light and warmth. From the west, snow. Arrange a mandala for your site.

(Mandala diagram here)

Now, it is obvious that fuel-storing plants (pines, gorse, broom, brambles) should not be in the fire sector, that wind-tender plants (vines, citrus) should not be to windward; that trees of almost unlimited height (black walnut, oak) should not block the sun unless you have a very hot site; and that many structures (roads, ponds, stone walls, paths) help in the fire sector, while others (paling fences, brush shelters) do not. We are on the way. Draw another mandala.

  • in Zone 1 we are often about, so high-use plants and labour intensive systems (annual garden, herbs, sheet mulch, pruned trees, hens) belong close-in here
  • in Zone 2 we can put less-needed species (geese, citrus, artichokes, spot mulch, vines)
  • in Zone 3 even more self-reliant species (sheep, wallaby, nut trees)
  • in Zone 4 really wild and tough plants (pines, oaks, comfrey, mulberry) and animals (goat, donkey, turkey).

Thus, the placement of structures, species and habitation become self-evident and you become less-uncertain about where to place your plants or pen your animals.

Roads, tracks, ponds and fences are special features serving at least two functions:

  • a road is a fire break, a water shed, an edge
  • a pond is a water store, an animal (fish) pen, a mirror to reflect sun, a fire break, a garden
  • a fence is an edge, a support, a sieve, a shield, a windbreak, a reflector
  • a house is a green- house, a shelter, an edge, a shade, a fence, a source of nutrients, an incubator.

This sort of thinking aids greatly in your design, and is a great help in deciding on materials as well as placement.

Now, visualise in three dimensions (we will discuss the fourth — time — next). Together with mandalas, the height-limit diagram helps you select further; for example, a stone pine of twenty metres, providing fuel, needing few visits, can only be placed in the S. W. of our diagram. There, one may search for a special rocky site.

It is here, too, that you can allow for landscape. A gully to the north allows taller trees, a hill there restricts height even more and so on.


We have evolved some rules for placement and materials, for landscape and species. What of succession? As long as there is ‘edge’, succession looks after itself to a great extent — there is always room for strawberries along a path, and filberts in a hedgerow.

Apart from shade and crowding effects, new niches evolve as plants grow. Dominants become shelters and vines become trellises; shady sites develop and more rather than less species can be introduced in time, even if the number of plants grows less by the processes of crowding and either controlled or natural thinning.

Fuel yield and natural mulching increases with time and energy inputs decrease with a net gain in the energy available from the system. Control functions (weeding, mulching) decrease and gathering and pruning functions increase.

Succession also decides how soon, and how many, animals can be introduced. If we start with pasture we can use geese, guinea fowl, wallaby (given some shelter). If we develop oak and seed forest we can introduce turkey, pig, hen, pigeon, but geese will decline in numbers.

We may never develop a sufficiently hardy system to survive possum, goats, pea fowl or cattle, but may have to use ‘soft’ browsers such as ducks, bantams, and guinea fowl unless we have a very large area and very good fences or hedgerows.

3: Suitable Plants, their culture and uses

Planning for complex permaculture involves a general knowledge of economic botany. The information that is available on useful species other than commercial crop plants is rather sparse. Much work needs to be done on yielding characteristics and requirements under local conditions of many species, but it is not necessary to fully research the needs, problems and characteristics of a species or variety before attempting cultivation. Every system becomes an experiment to refine and extend knowledge.

We can consider these lists (ed: see plant list graphic at end of article) in the light of certain principles basic to the system. With many species serving the same function or use, complexity of yield is increased. Many species of nut trees, for example, will give:

  • a long yielding period with no heavy work load in harvest
  • variety of produce with slightly differing uses
  • differences in ability to store
  • at least some yield in years of bad spring frosts, poor summers or drought (due to variation in climatic tolerances or different spp).

Also, each species serves many functions. The walnut, for example, provides culinary nuts, pickled fruits, dye from the nut husks, citronella (mosquito repellent) from the leaves and husks. With large numbers of trees, it can give walnut oil (expressed), animal forage (nuts) and valuable timber (old or thinned trees). Bergamot can be used to significantly increase available bee forage due to its excellence and ease of propogation. As well, the scarlet flowers are used in salads, the leaves a culinary herb and its tea is an antiseptic for sore throats.

The species in each list show great variety in cultural requirements, so plants can be fitted into appropriate niches rather than attempting to adapt the land to a ‘chosen’ crop. The nut trees as examples again:

  • almonds need sunny, well-drained sites but are draught resistant
  • hazels grow well in thickets and hedgerows
  • beeches need cool, moist forest conditions and frost
  • stone pines grow well on steep, dry, exposed, rocky sites
  • macadamias need a warm, sheltered location
  • the Butternut is adaptable to many soils, tolerates a high water table and is extremely hardy to frost and cold.

Permaculture systems should be designed to utilise all available resources (light and heat, space, nutrients and water), minimise competition and maximise symbiotic associations. This is ecological engineering and can be exceedingly complex, but much can be done quite simply with an appreciation of ecology.

  • plants such as blackcurrants, bergamot and mulberries can grow and yield on the small amount of light available under large top-storey trees
  • stone pines and other windfast species can be planted out in exposed positions to create shelter for wind tender Arancaria species and macadamias
  • the Bunya Bunya Pine and macadamia occur naturally together in dense sheltered forests on deep moist soils — this situation could be partly reproduced in the cultivated situation
  • nitrogen-fixing legumes such as lucerne and black locust aid nearby plants
  • in hot, dry climates carobs provide understory conditions favourable, in terms of shade and moisture, for many understory species
  • deciduous trees with deep roots draw-up subsoil nutrients and make them available to surface feeding understorey plants by the annual leaf fall.

Permacultures cannot be designed in the sense of a landscaped park, there is no finished product but, instead, an ongoing process.

For example, pampass grass could be planted in belts to provide shelter and forage, comfrey-lucerne pasture established for fodder, forage and sod improvement. Oak trees and sugar pod trees (honey lucust and carob) would not become significant yielders for some years, by which time they shade the pasture out, but provide acorns and pods for the animals.

As the trees mature (maybe thirty years later), crowding may require thinning, giving timber. Removal of groups of trees and the stock would leave sheltered clearings with well-manured and mulched soils for more intensive cultivation.

In orchestrating ecological succession, the aims are to have some yield available at all stages and for one stage to improve conditions for the next, while moving towards greater nett productivity and, thus, support capacity.

End of Part 1.

Plant list

Plant list continued.

Editor’s notes

The article is attributed to Bill Mollison and David Holmgren. The writing is suggestive of Bill’s style.

Design principles already evident

Even though the article is an early articulation of permaculture, some permaculture design principles are evident and other selections from the text can be interpreted as ad hoc design principles.

The principles below are in bold type, the selections from the text are not bolded:

  • PRINCIPLE: Create diverse systems
    A complex system, in terms of structure, species and varieties, is more stable and in general higher-yielding than a simple system.
  • PRINCIPLE: Plan functions in zones out from the dwelling for access according to need of visitation
    Establishing plants, animals, structures and infrastructure in zones at a distance from the dwelling according to their need for monitoring and maintenance creates efficiency in system management and effectiveness in system functioning (this is permaculture’s zoned landuse design methodology).
  • PRINCIPLE: Make use of the edges where ecosystems meet to increase productivity
    A complex system with edge (or interface) with some other system is more productive than a system with little or no edge.
  • PRINCIPLE: Consider the passage of time in designing systems to allow for their evolution
    All systems evolve. Succession of plants/animals occurs over time. Time is a design consideration in permaculture design.
    There is no finished product in permaculture. Instead, the design is an ongoing process.
  • PRINCIPLE: Permaculture is nature-assisted design
    Permaculture design uses ecological principles.
  • PRINCIPLE: In planning permaculture systems, consult the research. Accept feedback to improve productivity and knowledge
    Planning for complex permaculture involves a general knowledge of economic botany.
    Every system becomes an experiment to refine and extend knowledge.
  • PRINCIPLE: Design for multifunction
    Each species serves many functions.
  • PRINCIPLE: To increase resilience and productivity in a design, make best use of all available resources
    Permaculture systems should be designed to utilise all available resources.
  • PRINCIPLE: Obtain a yield
    The aim is to have some yield available at all stages and for one stage to improve conditions for the next.
    In a permaculture food production system the aim is to distribute, rather than concentrate, the season of yield.

Expression of an evolving permaculture

The article was one of the first expressions of permacuture’s zoned landuse design methodology.

Authors David Holmgren and the late-Bill Mollison developed the permaculture design system while living in Hobart, Tasmania, in the 1970s.

An article on permaculture history written for New Internationalist magazine by Steve Payne and I can be found here. See also: It’s been 40 years

The first expression of the permaculture design system in book form was Permaculture One by Bill Mollison and David Holmgren, published in 1978. Permaculture Two, by Bill Mollison, followed around a year later.

David Holmgren continues to offer permaculture education and design services from Melliodora, his home-base at Hepburn, Victoria.



Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
Russ Grayson

I'm an independent online and photojournalist living on the Tasmanian coast after nine months on the road in a minivan.