>
> Actually it does: (a) your odds of hitting buried
> water are between 75% and 100% no matter where you
> drill; and (b) if you do this a lot, you might well
> have developed a sense for correlations you aren't
> consciously registering.
>
> But I've never done it, and you have, so if you have
> reasons to think he's wrong about those odds, I'm
> all ears.
I've never dowsed for a site to drill a water well. The only dowsing I've ever done was to locate water lines buried less than six feet underground. The chances of digging at a random spot in the right-of-way and striking a buried water line are a lot less than 75-100%, more in the range of 2-5%.
When I was first working as a land surveyor I was sent out to locate buried water pipes for engineers. If you can't find any valves or blowoffs you have to probe and dig blindly. A guy I worked with showed me this trick where you take two pieces of wire, about 12" - 15" long, bend them 90 degrees in the middle, and hold them loosely in your hands so that half the wires hang down through the middle of your fists and the other half points ahead of you. Walk where you think you will be crossing the buried water line at right angles, e.g. if the water pipe is running parallel to the road, walk at right angles to the road. When you cross over the water pipe the wires rotate away from pointing ahead and cross one over another. At that spot you probe in the ground with a probe rod and see if you hit something (if it's Christmas, you _break_ the water line with your probe rod, what fun!). If the probe rod hits something, then you uncover it with a shovel, and if it actually is the pipe you're looking for and not some random chunk of concrete or rock, locate it and shoot the elevation on top of it with a level.
I watched him do it and to my surprise he did hit a buried water pipe under the spot where the wires crossed. So the next time I had to look for a buried water line I gave the technique a try. The first surprise was having the wires repeatedly twist in my hands as I walked across a certain spot; the second surprise was when I probed at that spot and actually hit the water pipe I was looking for. (As a kid I read about a "law" of "parapsychology," I think it might have been called "Rhine's Law," which basically states that "if you don't believe in ESP-like razzmatazz it won't work for you"; maybe despite my skepticism, I was just credulous enough to pass "Rhine's Law.") At any rate I used this voodoo technology at least a few dozen times over the years and I recall fair success with it.
I have no idea how this works. As water lines often run parallel with roads, it may be, as you suggest, that I subconsciously assessed the locale, consulted memory of having locating or staking water lines in the past for how far from the pavement or fence line or tree line to expect a buried water line to be, and then as I walked across that area the ideomotor response made me twist the wires in a way similar to how one unconsciously pushes the "planchette" when using a Ouija board. But I also had some successes which were hard to explain, such as locating an irrigation line across the middle of a large plowed field, or locating lines crossing a road. If I'm looking for a buried pipe parallel with a road, I know it falls in that 10 to 20 foot wide strip between the edge of pavement and the right-of-way line, but looking for a pipe that crosses the road somewhere down the entire length of a block, I'd have to pick one point on a line hundreds of feet long, yet I've successfully done that a few times with my little wires.
I had failures too, both negatives and false positives. Where I screwed up is that I failed to document every occasion when I used this pipe-finding technique. If I had I'd have numbers to see whether something was really happening or whether I was just fooling myself with rain dance magic ("Ooh, see, first I danced like so, and then lo! it rained! QED, my rain dance has great magic power").
yrs WD "conjuh" K