The first diagram (Fig. 49) is for a sandy loam soil, 7 to 9 feet deep 

 underlaid with a sandy soil. The observations were made in a trench 

 excavated across sixteen furrows. The furrows were 660 feet long and 

 the trench was half way down the furrows; it required about five hours for 

 the water to travel half the length of the furrows and about 12 hours more 

 to reach the end. This diagram shows results which are very surprising 

 to the irrigator. It is commonly believed that the water spreads laterally 

 so as to give a fairly uniform distribution throughout the soil, but the 

 diagram shows that the water spreads only a small distance laterally, 

 usually less than 3 feet from the furrows, which is not sufficient to wet the 

 soil uniformly. In the first foot 77 per cent, of the soil was wetted, in the 

 second foot 78.75 per cent, in the third foot 71.13 per cent., in the fourth 

 foot 40 per cent., in the fifth 27.50 per cent., and in the sixth 5 per cent. 

 This shows the necessity for placing the furrows close to the trees and for 

 having sufficient number of furrows to bring them close together. 



The next diagram (Fig. 50) is for a gritty clay loam with a subsoil 

 which is very compact, but which quickly absorbs water and becomes soft. 

 The trench was made across five furrows half way down the furrows. At 

 the end of 54 hours the water was cut off. The distribution of water in 

 this soil was much more uniform than in the previous case. This is due 

 to the soil being more compact which produces a slower downward perco- 

 lation and a greater sideways motion because of the greater effect of 

 capillarity. 



The third and fourth diagrams (Figs. 51, 52) show the limit of percola- 

 tion in a sandy soil from single furrows 10 inches deep and 5 inches deep 

 for a period of seven hours. With the deep furrow there is a greater 

 depth of percolation and the water spreads farther sideways. The deep 

 furrow also has the advantage that less of the surface mulch is wetted than 

 with shallow furrows. 



The fifth and sixth diagrams (Figs. 53, 54) show the outlines of perco- 

 lation in heavy loam for four deep and four shallow furrows under the 

 same conditions of soil and water. The deep furrow gives a more uniform 

 distribution of water, a greater depth of percolation, a greater sideways 

 absorption, and a smaller percentage of moisture rises by capillarity to the 

 surface to be lost by evaporation. 



The seventh diagram (Fig. 55) shows the probable distribution of moist- 

 ure in a sandy loam lengthwise with the furrows, from the head to the 

 foot of the furrows as determined by a few borings. The furrows are 660 

 feet long and the water was run in them for a period of three days. Suffi- 

 cient borings were not taken to determine the curve accurately. Other 

 experiments showed that the curve may ascend quite abruptly toward the 

 surface as indicated by the dotted line. This diagram shows clearly that 

 the water is not evenly distributed; the depth of percolation near the head 

 ditch is much greater than the average depth of percolation along the fur- 

 row; a large part of the water passes beyond the reach of plant roots and 

 is wasted. With heavy soils the difference in percolation would not be so 

 great. This experiment shows the necessity for short furrows, especially 

 in sandy soils. 



