One of the very earliest experiments on transpiration, conducted by Woodward in 1699, showed that it required less water to produce a pound of dry matter if the soil solution were of the proper concentration and contained the elements necessary for plant growth.  Little more was done to answer the above questions for over one hundred and fifty years.  Perhaps the question was not even asked during this period, for scientific agriculture was just coming into being in countries where the rainfall was abundant.  However, Tschaplowitz, in 1878, investigated the subject and found that the increase in dry matter is greatest when the transpiration is the smallest.  Sorauer, in researches conducted from 1880 to 1882, determined with almost absolute certainty that less water is required to produce a pound of dry matter when the soil is fertilized than when it is not fertilized.  Moreover, he observed that the enriching of the soil solution by the addition of artificial fertilizers enabled the plant to produce dry matter with less water.  He further found that if a soil is properly tilled so as to set free plant-food and in that way to enrich the soil solution the water-cost of dry plant substance is decreased.  Hellriegel, in 1883, confirmed this law and laid down the law that poor plant nutrition increases the water-cost of every pound of dry matter produced.  It was about this time that the Rothamsted Experiment Station reported that its experiments had shown that during periods of drouth the well-tilled and well-fertilized fields yielded good crops, while the unfertilized fields yielded poor crops or crop failures—­indicating thereby, since rainfall was the critical factor, that the fertility of the soil is important in determining whether or not with a small amount of water a good crop can be produced.  Pagnoul, working in 1895 with fescue grass, arrived at the same conclusion.  On a poor clay soil it required 1109 pounds of water to produce one pound of dry matter, while on a rich calcareous soil only 574 pounds were required.  Gardner of the United States Department of Agriculture, Bureau of Soils, working in 1908, on the manuring of soils, came to the conclusion that the more fertile the soil the less water is required to produce a pound of dry matter.  He incidentally called attention to the fact that in countries of limited rainfall this might be a very important principle to apply in crop production.  Hopkins in his study of the soils of Illinois has repeatedly observed, in connection with certain soils, that where the land is kept fertile, injury from drouth is not common, implying thereby that fertile soils will produce dry matter at a lower water-cost.  The most recent experiments on this subject, conducted by the Utah Station, confirm these conclusions.  The experiments, which covered several years, were conducted in pots filled with different soils.  On a soil, naturally fertile, 908 pounds of water were transpired for each pound of dry matter