In conducting the experiment, after the apparatus has been arranged as described, the small lateral tubes are, with one exception, closed with stoppers. On the open one, _d_, a rubber tube is fixed for the purpose of removing the water. The required water pressure is secured by taking the lateral opening corresponding to the pressure required. Water is introduced into the apparatus slowly through the glass tube _f_. The water rises to _d_ and then any excess flows off through _e_. By a proper regulation of the water supply the pressure is kept constant at _d_. The water flowing off through _a_ is collected by the funnel and delivered to graduated flasks where its quantity can be measured for any given unit of time. Since the rate of flow at first shows variations, the measurement should not be commenced until after the flow becomes constant. In general, the experiments should last ten hours, and, beginning with a water pressure of 100 centimeters, be repeated successively with pressures of eighty, sixty, forty, and twenty, centimeters, etc. In coarse soils, or with sand, one hour is long enough for the experiment. =176. Statement Of Results.=—In the following tables the results for ninety centimeters, seventy centimeters, etc., are calculated from the analytical data obtained for 100 centimeters, eighty centimeters, etc. MATERIAL—QUARTZ SAND. │ │ LITERS OF WATER PASSING IN TEN │ HOURS. No. Diameter of│ of sand │ Exp. particles │ Water in │pressure in Thickness of Soil Layer. mm. │ cm. 10 cm. 20 cm. 30 cm. 1. 0.01–0.71 │ 10 0.244 0.187 0.151 „ „ │ 20 0.282 0.198 0.154 „ „ │ 30 0.320 0.209 0.158 „ „ │ 40 0.358 0.220 0.161 „ „ │ 50 0.396 0.231 0.165 „ „ │ 60 0.434 0.242 0.168 „ „ │ 70 0.472 0.253 0.172 „ „ │ 80 0.510 0.264 0.175 „ „ │ 90 0.548 0.275 0.179 „ „ │ 100 0.586 0.286 0.182