Page:Tree Crops; A Permanent Agriculture (1929).pdf/360

 acre per year were added to this thirty-two million acres—a rather large result to follow from an endowment of one or two millions.

The variation of plants of the same species has received frequent mention. One more point needs to be presented in connection with the idea of breeding desert trees. They sometimes vary fifty per cent, or more in the amount of water required to produce a given result. This variation within the species suggests an interesting line of experimentation with different species and strains, to find the most efficient for particular places.

Lastly I wish to submit the thesis that the trees now unused or little used as crop mediums may be the best kind of crop for some of our levelest and most arable lands. I have in mind a two-story agriculture with tree crops above and tilled crops beneath. By analogy I would recall the French practice (page 166) of scattering walnut trees all over the farm while going on with the farming. This does not sound alluring to the machine-using American, but let us consider it. Suppose you had a farm on the sandy plain somewhere between New York and Galveston. You are growing hogs and letting them