(or nearly) distant from a centre, and kept gradually turning until it pointed directly downward. To assure myself, I repeated this experiment many times, and arrived at the conclusion that the switch commenced to turn at an angle of forty-five degrees from the edge of the water, and that the distance from my hand to the water would be measured by the distance from the point where the switch commenced to turn to the point of absolute turn-down, and so it seems to be. The following diagram will show more clearly my meaning: A B, B A is the surface of the earth; W, a stream or pool of water below the surface. Walking along toward A B the switch begins to move at A, and turns down at B; the angles B A C and B C A being equal, the distance from A to B is equal to B C. Measure the distance, therefore, from the point of commencement of turning to the point of turn-down, and you have the depth from your hand to the water. I have verified this over many water-courses, upon bridges, etc., and I am satisfied it is correct, at least for the latitude in which my experiments were made. Upon this basis I made my first estimate of the depth of the water at Coloma, and gave it as from twenty-five to thirty feet. I employed an experienced well borer and had a two and one-half inch pipe driven into the ground at the exact point my switch indicated, and found water at twenty-seven feet exactly. I had the pipe driven down forty feet, and found that I had thirteen feet of water in it. I then had a windmill erected and a large tank. Up to the time of my leaving the road, the engines were supplied with the water, which, besides, proved to be of excellent quality for drinking. My well borer, who was a doubting Thomas, said he believed that he could get water at the same depth anywhere. Fortunately for my theory, a neighboring store-keeper tried the doubter and failed to get water under fifty-nine feet.