A.R.T. part three. Martin Crawford is honest (and philosophical) that not everything at the Agroforestry Research Trust works out quite how he anticipated. Case in point he devised an irrigation system which worked fine- for a while.
 The various structures like the greenhouse are at the lower end of the site and rainwater off the roofs is collected then pumped uphill by solar powered pumps. At the top of the site the water passes through a large filtration tank then into several large reservoirs. The water gravity feeds through pipes as needed onto the site as a whole.
 It all worked according to plan for several years then the membrane used to line the reservoirs began to perish and leak. Actually it's all too common that products and materials don't last the way they should but it means that Martin has to re-make the reservoirs.

 A few set-backs are inevitable but the scale of Martin's achievements is remarkable. The Littlehempston site is where he has his nursery beds to grow trees and shrubs for both the Trust itself and to sell. He uses no chemicals and improves the soil by alternating growing areas with areas given over to "green manure" cover crops for a year or two, as above where he has sowed Black Medick and Phacelia.

 Martin does all his propagating in two polytunnels from seeds, cuttings and division. A lot of commercial plant nurseries buy in plug plants and grow them on but he does it using the classic methods of botanical horticulture. The potted perennials are laid out alongside the polytunnels awaiting sale or use.
 Many garden centre plants are mass produced in "optimum" conditions i.e. kept in polytunnels, overfertilised, sprayed regularly and too tender to resist pests, disease and weather. By contrast the plants that Martin Crawford grows are clearly very healthy and hardy.
 The working environment at A.R.T. is a utilitarian one. As far as I can see no attempt is made to prettify it but everything large and small is there for a reason.

 Martin Crawford plays a long game. On the way out I went to have a look at a planting of six Monkey Puzzle trees that he had referred to earlier in the day. Monkey Puzzles are dioecious i.e. there are 'male' and 'female' trees whereas the great majority of flowering plants have both 'male' and 'female' flowers on them for the purposes of pollination.
 An edible nut is produced by the female of the species but you don't know if a tree is male or female until it's at least twenty years old!