open fields prevail in the east and south. Yet the enormous value of a wind-screen to vegetation can be readily observed, and the market-gardeners both in England and the still dryer districts of the south of France make great use of them. Lastly, we must have more knowledge about the relation between transpiration-water and growth: we do not know if the high ratios we have spoken of hold for all plants. Xerophytic plants are supposed to be possessed of protective devices to reduce loss of water: Are they merely effective in preserving the plant from destruction during the fierce insolation and drying it receives? and do they enable a plant to make more growth on a given amount of water? Wheat, for example, puts on its glaucous waxy bloom under dry conditions: Is this really accompanied by a lower rate of transpiration per unit surface of leaf? and is it more than defensive, connoting a better utilization of the water the plant evaporates?
The cultivation of these soils with a minimum rainfall necessitates varieties of plants making a large ratio of dry matter to water transpired and also with a high ratio between the useful and non-useful parts of the plant. Mr. Beaven has shown that the difference in the yields of various barleys under similar conditions in England is due to differences in their migration factors: the same amount of dry matter is produced by all, but some will convert 50 per cent, and others only 45 per cent, into grain. This migration ratio, as may be seen by the relation between corn and straw on the plots at Rothamsted, is greatly affected by season; nevertheless Mr. Beaven's work indicates that under parallel conditions it is a congenital characteristic of the variety and therefore one that can be raised by the efforts of the plant-breeder. The needs of dry-land-farming call for special attention on the part of the breeder to these two ratios of transpiration and migration.
Closely linked up with the problems of dry-land-farming are those which arise in arid climates from the use of irrigation-water on land which is either impregnated with alkaline salts to begin with or develops such a condition after irrigation has been practised for some time. The history of irrigation-farming is full of disappointments due to the rise of salts from the subsoil and the subsequent sterility of the land, but the conditions are fully understood and there is no longer any excuse for the disasters which have overtaken the pioneers of irrigation in almost every country. Sterility may arise from two causes—overmuch water which brings the water-table so close to the surface that the plants' roots may be asphyxiated, or the accumulation by evaporation of the soluble salts in the surface layer until plants refuse to grow. The annual cutting off of the cotton crop in Egypt as the water-table rises with the advance of the Nile flood affords a good example of asphyxiation, but in the neighborhood of irrigation canals we also find many examples of sterility due both to the high water-table and an accom-
