eDowser is currently available as a web-app. Soon as an app for Google Andoid and Apple iOS.
to enable a first-hand experience of what dowsing is and isn't
to make dowsing convenient and non-geeky to use
to make a useful technique generally recognised
to put dowsing on a scientific basis as far as possible.
the app has these sections:
the dowsing device itself
change various internal settings like font size etc. Includes a button to re-run the intro demo screen.
what the app is and does
how to use it
example questions about things you know you know, like facts about yourself
example questions about things you don't think you know well enough
example questions about things you don't think you know at all
example questions about things you didn't know even existed
question types that don't work, found by reason and experience
this page also has these sections:
the theory of dowsing, so far as it exists
the classic coathanger+biro L-rod
tests of reliability
videos, articles, books about dowsing
The theory of dowsing, as far as it exists.
The movement is from you but not intentionally, which is why some
people confuse it with magic. It's called the ideomotor response
– a psychosomatic phenomenon.
First inferred by Georgius Agricola in
1556, then by Athanasius Kircher in 1641, then named & described by William
Benjamin Carpenter in 1852.
There is a question whether it is a learned behaviour or not.
In 2018 it can now be said there's a consistent response (at least to 3 Sigma
– 99.7%) for Show Yes & Show No which means the ideomotor response is mechanically
The wrist/arm movement is in fact rotational, which is how an L-rod doesn't
"push uphill" and is stable if held just under horizontal, i.e. it will centre
The second puzzle is how you get meaningful answers to other questions,
which evaluate to Yes & No and are shown as before.
It can't be a science – objective knowledge –
because it uses personal judgement and sensory experience (which is sometimes wrong).
It could be a technology – a mixture of objective and subjective –
as with: sports, cooking, language, memory, colours, riding a bike, and driving a car.
It's unexplained and is an open problem. It isn't magic but is extraordinary.
It's susceptible to confirmation bias – you recall the successes
more easily than the failures – record everything!
The earliest clear instance was in 16th century Germany of miners dowsing for minerals.
Divining is the general category of judgement by intuition,
going back to the Ancient World and beyond,
e.g. Consulting the Oracle and Reading the Auguries.
Divining includes dowsing – and is confused with it – but is much wider.
If you prefer not to get the app but wish to try dowsing, here's
the classic coathanger+biro dowsing device, which cost me about 17p or 34p to make a pair.
You'll need some pliers to bend the wire and to extract the biro inner.
You can use thick drinking straws instead of biro outers if you prefer.
You do need new ones to make sure the movement is smooth.
You don't need a pair unless it's windy.
The dowsing response is towards or away from the centre whereas wind will push both to one side,
which a pair is able to detect.
The other thing a pair does is increase the visual feedback (i.e. they cross or diverge) which increases the general response.
You hold it like a pistol but pointing slightly down. The closer it is to horizontal the more sensitive it is.
It's stable in the sense that if you disturb it it will return to centre.
Your wrist & arm movement is rotational so you're not "pushing it uphill" in moving it.
Instead you make a new point of balance and the rod falls to that.
starting at asking questions.
The 3-sigma rule of thumb is a convention in the empirical sciences that if
nearly all values lie within three standard deviations of the mean –
which is 99.7% – then that's considered proven. Social sciences
use only 2 sigma, particle physics uses 5 sigma.
3 sigma test of the ideomotor response for the coathanger+biro procedure:
Hold the l-rod vertically.
Level up rod until just below horizontal.
Test that rod is stable i.e. it will return to centre if disturbed.
Pick a point in the background the rod end aligns with and use that repeatedly.
Relax hand & arm so the response can happen.
For each test question:
Balance & align the l-rod.
Say the test question.
Wait for a response.
Record the response if any expected observation:
For right-handed use: Show me Yes means the rod moves at least an inch to the left,
Show me No to the right, both within a few seconds.
For left-handed use the directions are reversed. post-test remarks:
With some tests there was a slowness to move.
It would be better if tests were random to eliminate movement out of habit. conclusion:
No difference between expectation and result was observed for 370 tests,
so by the 3 Sigma Rule of Thumb the ideomotor response for coathanger+biro l-rod is a fact. results:
185 "Show me Yes" and 185 "Show me No" tests, in mixed order.
3 sigma test for eDowser.
A farmer in Georgia dowses for and digs his own well, 2017:
Hans-Deiter Betz 1995, Department of Physics, University of Munich
"Unconventional Water Detection: Field Test of the Dowsing Technique in Dry Zones: Part l",
Journal of Scientific Exploration, Vol. 9, No. I, pp. 1-43, 1995
In this study, a team of scientists investigated the ability of dowsers to
find drinkable quality water with a useful flow rate, over a period of ten
years in ten different third world countries in which geological conditions
were particularly difficult for finding water. In Sri Lanka, expert dowsing
found 691 sites with a 96% success rate. In a direct comparison of
conventional detection and dowsing, a team of hydro-geologists were given the
same task as the dowsers but had a success rate of only 21% against 86% for
dowsing. Similar work in the Congo produced a success rate of 90% for the
Arthur Bailey PhD, former President of the British Society of Dowsers
"Anyone Can Dowse For Better Health".
Out of print but available second-hand; if you need it the publisher is .
This is the best book about dowsing since it is by someone with a scientific outlook
(he was a lecturer in Electronic Engineering in Bradford, Yorkshire).
background to dowsing
not an objective science
relies on the subjective so can't be
tests of dowsing have shown it doesn't work objectively but anecdotally it does work.
there is a level of belief which mixes the objective and subjective
such as cooking, sports, language, riding a bike, driving a car
thing missing is meaningful connection with thing dowsed
It isn't magic or mediaeval witchcraft nor supernatural
Started in the North German renaissance amongst miners, first reference in the Harz mountains
synonym divining - ambiguous so we prefer the word dowsing
dowse from Middle English duschen - to strike (Modern English related to to dash) translation of Middle German Schlag in Cornwall
Two phenomena: ideomotor, which is below the normal detection threshold, and we know more than we are aware of.
It works using the ideomotor response (wikipedia) to provide
the muscle tension showing which idea is in mind, Yes or No. The basis of dowsing is a different response to Yes and No.
The source of what idea is in mind is straightforward when you ask about things
you already know you knew, but gets more mysterious with questions that are less so.
"Ideo" means the meaning of what you think, so it works in any language.
Dowsing in general is semantic not syntactic.
ideomotor effect & meaning - v subtle so can confuse as if not us
how to use
How do I use it?
Place the phone lying flat in your dominant hand.
Start eDowser using the button at the top.
Balance the pointer: tilt the top of the phone down,
centre the pointer,
and then level up slowly keeping the pointer still
until it's inside the arc and you see "ready" displayed.
This takes practice.
To make it easier to keep the pointer still, lower the sensitivity in
though that also reduces the dowsing response.
Keeping your hand still but relaxed, and arm free, slightly away from your body,
ask "Show me Yes" to see which way the pointer moves, without deliberately trying to.
Then re-balance and ask "Show me No".
to swap the yes/no response if necessary.
The main thing is Yes and No being different.
The difference in responses is the basis of dowsing.
It's normal to get no response at first, but keep trying. Hint: let it happen, rather than make it happen.
Practice strengthens the response.
You can also get no response meaning Don't Know; to check, ask "Do I know the answer here?".
There are 3 techniques for getting answers.
They can be combined:
see which way the pointer goes, as for "Show me Yes/No".
You can add the questions you ask to the log. The answers can then be recorded.
estimate an answer lower than you expect.
Ask "Is the answer more than my estimate?"
If a Yes add a step to the estimate and ask again, until you get a No.
If necessary refine by going back one step and adding a smaller step, and so on.
This works even when the answer to your particular question is a fraction.
see where the pointer goes, ignoring the Yes and No directions. You may need to turn to follow it round.
If you get an answer that appears nonsensical or even dangerous you should over-rule it with common sense.
Test your answers with alternative methods to build confidence in your dowsing ability.
select a line by tapping it. If it turns black it's now active.
sel'd shows only the selected lines. You can set the window length so you see the current queue on the whole screen.
all shows the whole log and the various buttons to do things.
day headers will select a whole day's lines at once, in a cycle of taps.
One is added automatically when the date changes and there is a new line (add or copy).
You can have more than one for the same date, i.e. they can be copied.
current answer is the line where the answer will be put. The answer box is a different colour from the rest.
add to enter a new line.
copy makes a copy of all selected line and adds them to the end.
edit enables changing the first selected line.
del deletes all selected lines, with an accident-preventer.
up moves all selected lines up 1.
down moves all selected lines down 1.
send emails the log to the address in settings, with an accident-preventer.
It's in CSV format so you can import it into a spreadsheet
settings has various things the user can set:
intro re-runs the intro screen.
font size affects all text; default 12px.
start button position should you be left-handed; default right.
sensitivity changes the skitteriness of the pointer but also the strength of the dowsing response; default 90%.
yes/no response sets what is your particular Yes/No direction; default "left is yes, right is no".
yes/no angle so the app knows when you've detected a Yes or No; default 30°
log window max size to keep the queue visible in the screen; default 3.
log time format is either 24 hours HH:MM:SS or 12 hours HH:MMam/pm; default 12 hours.
email address to send log to.
reset to default values.
The start/stop button is to prevent random movements being recorded as answers, and also to save battery power.
You may need to increase your screen timeout delay, and to switch off the screen auto-rotation.
The balancing is so that the next movement is from a subtle level.
Anonymous dowsing: everyone walks around looking at their phone, you can look entirely normal :)
The same movements happen with other
such as a forked twig (a.k.a. Y-rod) held in tension,
and with a bent wire (a.k.a. L-rod) held just off horizontal.
Some people find they can programme their response by deliberately moving the pointer in the expected direction,
then reverting to letting it happen.
Sooner or later you will be able to read a Yes or No from just a twitch without strictly starting in the centre.
Always make the question as clear as possible, else you'll get answers that make no sense.
Always apply a plausibility test in case your question thought isn't as clear as you believed.
Can I, May I, Should I? is a useful check on the viability of dowsing a particular question or situation.
"Can I dowse about/in X?" means "Will my dowsing give a meaningful answer here?".
"May I?" is asking permission: it's not clear of whom, but it puts you into the passive, which is effective against willing a particular answer.
"Should I?" means "Is it OK for me to dowse this here & now?".
Additional checks include "Am I ready to ask this?", or the follow-up to No of "Is that true forever?".
It's best practice as recommended by the British Society of Dowsers.
You may need one of those
power packs for a long session.
A rubber band can help hold the phone in place.
The response direction can change over time, so confirm "Show me Yes" and "Show me No" for each session.
What can I ask?
Things you know you know, like facts about yourself:
"Is my name <insert your name>?"
"Is my name [say or think your name]?"
"Is my name Rumpelstiltskin?" (to test getting a No).
"Am I in England?"
Things you didn't know you knew well enough:
"Show me north".
"Is my weight <guess your weight>?". Or count it.
"Is this food good for me / needed / ready to eat?"
"Where's that thing I lost?" Ask a series of questions to home in, always imagining the object. It works sometimes.
Things you didn't know you knew at all:
"Show me the way to that buried leaky pipe, with a Yes when I get
there... Now show its direction... Now let me follow it with a Yes when I'm over the leak".
"Which cup is the penny under?" Children can be surprisingly good at this, adults less so.
"Has this meeting had a good result for all?"
"Is that road the best way home?"
Things you didn't know even existed:
Starting about 30 feet away from a tree, walk slowly towards it with "Show me a Yes whenever I move into a band of this tree's aura."
To dowse for a water source you do this:
Ask to be shown the direction of a source of underground water, with a "yes" when you're standing on top of it,
When there ask if the water is drinkable (and not sewage!), whether seasonal, what the flow rate is (use the counting method),
Dowse how deep it is (use the counting method), what materials are above it (i.e. you don't want to drill through granite),
whether it's in clay (which is not much use),
Mark it on a map with the details and see if there are any more to dowse,
But there's little point unless you've a drilling rig on hand, which costs thousands.
What can't I ask?
Some question types have problems (try them!).
You can get apparently confident wrong answers.
Questions that don't have a Yes or No answer.
Questions you want to be true:
"So where's the gold hidden?" Personal gain seems to obscure or switch off how it works,
as if there's a greater strength or priority.
Remember it's your own intuitions that are being amplified even though it can feel as if you're consulting the Oracle.
Questions you can't know yet:
"Will this work?"
The future doesn't exist yet so the answer will be a probability.
In general the future is difficult to predict even without the element of dowsing.
Questions that are circular:
"Does dowsing work?" This seems to affect all attempts to test dowsing empirically.
Subjecting dowsing to an empirical test doesn't generally work.
If you try to dowse when you could just look, the dowsing gives entirely plausible but wrong answers.
Try it! One example is whether the top card is a picture card or not.
There's something extra weird happening here.
the logo is a hazel leaf. Hazel was traditionally used for dowsing rods because it has just the right amount of flexibility.
and represents a forked twig that used to be the standard dowsing device.
It was found at briarpress.org and edited.
You can see a bigger version on the eDowser Facebook page
It's useful to dowse with others: you dowse better and you get confirmatory results.
There is a close relationship between dowsing and the metaphysical world. It
is that dowsing is useful to investigate the metaphysical, because it is
generally subtle and subjectively appreciated, but dowsing doesn't entail the