0)0

"•
•,•
• -•
)•
.EEP
C.7 7:i

~~m -
EXPERIMENT
IN WATER
DOWSING
EEDavid

1.Til(Gaisford.
I"SIS
AFIT/GEE/F.NS/94S-O
1).
E.
DEPARTMENT I

OF
THE
AIR
FORCE

.
.
AIR FORCE INSTITUTE
OF TECHNOLOGY
AIR UNIVERSITY
Wright-Patterson

Air
Force
Base,
Ohio
AFIT/GEE/ENS/94S-01

EXPERIMENT IN WATER DOWSING

THESIS

David I. Gaisford, P. E.
AFIT/GEE/ENS/94S-01

Approved for Public Release; distribution unlimited

=TC QUALITY pSP-ECTED 3

S0694-30622

V4HI HI
AFIT/GEE/ENS/94S-01

The views expressed in this thesis are those of the author and do not reflect the official
policy or position of the Department of Defense or the U. S. Government.

Aooession 7or
NTIS GRA"
DTI" TAB "

JuY;i., Cation

By-
Di~trut lo.
Avhbllb111ty Codes
~aiad/or
'AVi
Dist~ Specils
 THESIS APPROVAL

Student: David I. Gaisford Class: GEE 94S

Thesis Title: Experiment in Water Dowsing

Defense Date: 11 Aug 94

Committee Name/Department Signature

_ _ _ _ _ _ _ _ _,_,.
_____ _.
A dvisor/co-A dvisor
__

(circle appropriate rolc)

co-Advisor/ENS Representative J, 4mbw-fE$ lJowEJal ZNJA 62dargftv-je

(circle appropriate role)

Reader " .R__ _ _

___ IV&

Reader M
AFIT/GEE/ENS/94S-01

EXPERIMENT IN WATER DOWSING

THESIS

Presented to the Faculty of the School of Engineering

of the Air Force Institute of Technology

Air University
In Partial Fulfillment of the
Requirements for the Degree of

Master of Science in Engineering and Environmental Management

David I. Gaisford, P. E

September 1994

Approved for Public Release; distribution unlimited

 Acknowledgment

By continuing in the middle of a program as I have done, time to select a thesis

topic was limited. The topic of this research was easy to comprehend, but the lessons

learned by the process will be carried for the rest of my life.

This work was the result of the time of many people. I wish to thank everyone
involved and especially recognize the time and effort of Mr. Lewis Carl. As the subject
of the research, he has given himself freely to any request and provided the data to

conduct the research.

I wish to thank my thesis comnuittee, Lt Col Kenneth Bauer, Dr Matt Kabrisky. Dr

Steve Rogers, and Maj Andy Howell. Your inputs will have made this a much more
respectable product. I'm sure we will face a future of stout criticism for the results. I can

only say that we reported what we saw.

My humblest thanks go to my family, Genia, my wife, Toni, and the one on the
way. You have followed me on this trying adventure for the past two and one-half years.
I've got my master's, now it's your turn, Genia. Go get 'em.

David [. Gaisford

ii
 Table of Contents

Acknow ledgm ent ......................................................................................................... ii

Table of Contents ............................................................................................................. iii
List of Figures .................................................................................................................. iv
List of Tables .......................................................................................................... v
Abstract ............................................................................................................................ vi
Chapter 1 Introduction ..................................................................................................... 1
Problem Statement ......................................................................................... 3
Research O bjective ........................................................................................ 3

Scope and Lim itations ..................................................................................... 3

Chapter 2 Literature Review ...................................................................................... 4
Q uick H istory of Dow sing ............................................................................. 5
The D ow sing A rgum ent .................................................................................. 5
The D ow sing Problem .................................................................................... 7
Im portance and U sage of D ow sing .................................................................. 9

Interview s w ith Dow sers ................................................................................ 9

Mr. Lewis Carl .................................................................................... 9

Mr. Jesse Aites .................................................................................... 12

H ow D owsing W orks ....................................................................................... 12
Geology, G roundw ater and Dow sing ............................................................. 13
Chapter 3 M ethodology ............................................................................................... 16
Proposed Experim ents .................................................................................... 16
M ethodology for the Dow sing Experim ent ................................................... 17
Establishing Statistical Validity ...................................................................... 21

iii
 Binom ial Probability Distribution ...................................................... 21
Hypothesis Testing .............................................................................. 22

W ald's Equations ................................................................................ 26

Analysis of the Param eters .................................................................. 28
Determ ining the Experim ent Param eters ........................................................ 32

Chapter 4 Results ........................................................................................................ 35

Proposed Water Well Site, State Route 35, Urbana ........................................ 35

Selecting the Site.............................................................................................. 38

Setting Up the Equipm ent ............................................................................... 41
Conducting the Experim ent ............................................................................. 42

Chapter 5 A nalysis and Conclusion ........................................................................... 45

Analysis of the Experim ent ............................................................................. 45
Experim ent Process ............................................................................. 45
Failure to Explain Random Procedures ............................................. 47

The Break at Trial 17 .......................................................................... 48

Changes in the Param eters ................................................................. 49
Implications of the Results ............................................................................. 50
Conclusion ...................................................................................................... 51

Appendix A ...................................................................................................................... 53
A ppendix B ...................................................................................................................... 66
A ppendix C ...................................................................................................................... 75
Bibliography .................................................................................................................... 76
V ita ................................................................................................................................... 78

iv
 List of Figures

Figure Page

1. Mr. Lewis Carl 3

2. Lewis Carl's Depth Gauge I1

3. Schematic of Proposed Network 18

4. Manifold 'A' 19

5. Experiment Test Procedures 20

6. Graphical Test with a = 0.05, 13 = 0.10, PO = 0.10, Pj = 0.50 29

7. Graphical Test with a = 0.05, 03 = 0.10, PO = 0.18, P1 = 0.25 29

8. Wald's Equations Parameter Analysis 30

9. Experiment's Graphical Test 33

10. State Route 35, Urbana Site 36

11. Typical Well Screen 37

12. Site Location Plan 39

13. Site Layout Plan 40

14. Graph of Experiment Results 44

15. Experiment Data Plotted with Removed Trials 48

16. Experiment Data Plotted with Changed Parameters 49

V
 List of Tables

"Table Page
1. Odd of Successive Binomial Trials 23
2. Experiment Results 44

vi
 Abstract

Do'v .'ag is a folklore process used to locate an unknown, such as the best location
for a water well, by the use of a hand-held device. The process is commonly know as
water witching, divining, dowsing or radiesthesia. The practice continues despite the lack

of a proven scientific basis.

This research develops an experiment to test the claims of a dowser. Specific
procedures are established and statistical theory is applied to determine if one man can

identify which of five water lines has flowing water in it better than a chance operator
could achieve. The statistical analysis uses Abraham Wald's sequential analysis
procedures for establishing when to accept a hypothesis in a binomial situation. The
dowser's performance proved to be better than chance. Further research is recommended.

vii
 EXPERIMENT IN WATER DOWSING

Chapter I

Robert McNamara appealed for the help of his Marine audience in a 1966 film.
He was seeking new and innovative ways to solve military problems of the day. Louis

Matacia, "a professional dowser with an impressive record," answered the call and
presented himself to the Marine Corps Landing Force Development Center at Qantico,
Virginia. Though he met many skeptical people, they let ,roceed to their mock

Vietnamese village where he located an underground tunnel system by dowsing (2:199).

He subsequently trained some young Marines and asked to show his talent to higher

command. His first trainees were shipped to Vietnam and soon had reported success in
locating tunnels. An article appeared in the 13 March 1967 edition of The Observer, a
weekly publication for US Forces in Vietnam, attesting to the success of the Marines
(2:206). With increased reports of success reaching the United States, the Marines started
to investigate dowsing further.
Dowsing or divining is the process of using a device to locate a known object in

an unknown location. It has taken on many forms and may have been practiced since
history has been recorded (15:12). The most familiar forms of dowsing are the use of
forked branches to locate water and angled rods used to locate pipes underground. Other
forms use pendulums as the device.
 The Marines' interest in dowsing ended in 1971 when the Commanding General at
the Marine Corps Development and Educational Command decided there was no
"scientific basis" for the phenomenon (2:213).

In 1990, Lewis Carl , Figure 1, approached Lt Col Kenneth Bauer to offer his
services to the United States. He had been dowsing for water since he witnessed a

dowser when he was 12 years old. For the past 25 years he has been earning money as a
professional dowser and felt he could find water in a harsh desert environment for the
troops deployed to the Persian Gulf. As it turned out, finding water was not a problem
for the US Forces and his services were not needed.
Lt Col Bauer's curiosity was piqued, however. If Mr. Carl's claims were true,
would it not be to our advantage to understand and learn this skill? Much as the Marines
who met Louis Matacia in 1966, Lt Col Bauer's meeting with Lewis Carl created more

questions than answers. Of the questions that surfaced, the primary one was "Could this

man's claim be true?"

A quick review of the literature showed a mixed reaction of believers and non-
believers. Mr. Carl presented possible evidence of his success, and Lt Col Bauer decided
that there was enough evidence that the subject could not be brushed away. The

possibility of success seemed too great to dismiss.

With this background, this research was recommended to try to substantiate if Mr.
Carl's claim is valid. If one can establish that he has a skill that can be learned, today's
technology may now be ready to measure and quantify its sources. What stopped the

Marine investigation into the subject was the inability to scientifically prove the skill
exists. In the twenty years that have passed since that time, technology has advanced to
the extent that methods exist that can measure many previously undiscovered aspects of

hu.,ian nature. A dowsing sense, if it exists, could be a new aspect.

2
Problem Statement
Does Mr. Lewis Carl possess the ability to locate subsurface water better than one

could expect a person to find water by chance? In this preliminary study the only interest
is in discovering if the man has an ability that can be proven through a scientific process

to provide a statistically valid result.

Research Objective
This research will be done to study the ability of one man to locate flowing wvater

by dowsing in a controlled experiment. If the dowser can pass the initial study,
additional research into the physics of the process may be warranted.

Scope and Limitations

As stated above, this research will be limited to one individual who claims to
possess an ability to locate water for landowners who are preparing to drill for domestic
water. Dowsers claim to have many more abilities than just locating water, but none of

these reported abilities will be tested here.

Figure 1: Mr. Lewis Carl

3
 Literature Review

The ability to dowse for water is a controversial subject; as such, there are
believers and non-believers. This literature review required the study of both sides of the
story and analysis of the literature to determine the underlying assumptions. Technical
experience or education does not determine who falls into each category. It seems that
authors are either set upon proving or disproving dowsing with predetermined notions.
Hence, most work appears subject to a prearranged conclusion.
Very little objective information is available on the subject. Most written
materials support dowsing but lack hard data to uphold the conclusions presented. Only
one book critical of dowsing was found and it presented arguments that followed
scientific methodology. Of other critical items found, most were periodical articles.
They supposedly reported on experiments, but lacked sufficient data to support this
classification. This lack of critical material appears to stem from a typical rift between
believers and non-believers of any such sensitive subject. Believers continue to preach
their beliefs. The non-believer dismisses what the believer says and does not see much
need to continue arguing. Once dowsing has been disproved, in their viewpoint, nothing
more needs to be said.
For the dowsing enthusiast, the critical literature does nothing to "dowse" their
beliefs. As one author put it:
It might be much more comfortable to have an explanation that looks
something like what other people accept as 'fact': but we don't have one.
None that would stand up to any real scrutiny, anyway...
...But in a way there's no point. As we've seen we don't really need to
know how it works, as long as it does actually work. Dowsing is far more a
technology than a science; and all we need to know in any technology is how
it can be worked (8:151)

4
 The literature reviewed was from a selection of pro-dowsing literature, the limited

items documenting contrary viewpoints, and interviews with the players in this research.

This information will be presented in a historical format, following the development of

dowsing from both points of view. Additionally, interviews with the dowser of this study
and his clients and employers will close the discourse. From these reviews and
interviews one can draw definite conclusions about what dowsing is, how dowsers say it
works, and how to test the skill.

Quick History of Dowsing

Numerous texts have been written on the subject of dowsing and provide a well-

documented history of the subject (Appendix A). The practice of dowsing may be as old
as 7000 years (15:12). Barrett and Besterman provide a good introduction into the

history (1:1-20). Most other books on the subject reference Barrett and Besterman or
provide similar accountings for dowsing's history. Vogt and Hyman summarize Barrett
and Besterman's work earlier work (15:12, 15, 16, 23). Dowsing as is commonly
practiced today is linked to German miners (1:19) Little has changed in dowsing since

these early descriptions.

The Dowsing Argument

Research into the subject of dowsing starts around the turn of the 20th century.

At that time, many books had been written on the subject. But without scientific
validation of the process, it remained a folklore method. Barrett and Besterman

attempted one of the most precise accountings of the success of dowsing. They authored
several volumes from work conducted from the late 1890's through the 1920's which
culminated in their book The DiviningRod, printed in 1979'. They trace the history of
dowsing, in the form discussed here, to sixteenth century Europe (1:7).

1Barrettand Besterman's book published in 1979 was the final work completed by Besterman long after
Barretts death. The entire text is based on work completed during the period mentioned. Barren and

5
 Many books were written supporting dowsing in the intervening years. Then in

1959, Vogt and Hyman published Water Witching US.A., the most thorough scientific

study of the subject written to date. They were interested in discovering why the process
remained "as vigorous as ever" (15:22). They surveyed 500 County Agricultural
Extension agents throughout the United States about dowsing in their counties (15:7).
Vogt and Hyman explain that the "strongest argument for water witching comes from

case histories" (15:40) Their conclusion, from a scientific perspective was:

...we don't have to resort to prejudice to dismiss water witching as invalid.

The evidence for it, when assembled and examined, is not merely
insufficient; according to current scientific standards (parenthetical omitted),
it is appallingly negative (15:82).

Despite these findings, the practice continues. Vogt and Hyman suggests dowsing

continues because of"...the difference of opinion between skeptic and believer is in the
interpretation of the facts. Each side draws a different moral from the same story
(15:81)." A pro-dowsing author supports this conclusion when Christopher Bird
criticizes their findings, calling Vogt and Hyman's conclusion "feeble" and based on a
bad assumption, that divining can not be "100 percent [accurate] (2:10)."
In 1977, the U.S. Geological Survey entered the picture through a pamphlet on
water dowsing. After explaining dowsing, the authors define groundwater and how

hydrologists locate it. Based on the data required by the hydrologist to locate water,

including "hydrologic, geologic and geophysical knowledge (14:10)," and that "no single
technique suffices to locate favorable water-well sites" they concluded "that the expense

of further tests of water dowsing is not justified (14:11)." The argument continued.
In 1979, James Randi discussed a $10,000 reward for any dowser who could

prove, in a controlled experiment, he had dowsing abilities (13:16). In the experiment, he

had a IOm x 1Om area constructed with three flowing pipelines. Four dowsers attempted

Besterman published other books on the subject in early 20th century. See Appendix A for some of their
works.

6
to locate a vein and none of them succeeded. Randi kept his reward, but did not convince
any of the dowsers that they lacked their reported skill (13:20).
In 1984, Randi reported on a subject who claimed to have proved the dowsing
sense (12:329). By this time, the reward had increased to $110,000. Under Randi's

scrutiny, the dowser's claim was found to be linked to "auditory and visual clues." Randi
concluded the man had no dowsing sense and did not reward the prize.
In 1987, Norman Eastwood reported on a "human magnetic sense" in an English

medical journal. He reported using a pendulum over portions of his body to locate "north
and south pole reactions" as those observed in a magnet. Upon his 'test' he decided that it
was "conclusive evidence for the existence of a human magnetic sense and that dowsing
reactions are associated with it (6:676-7)." Randi called his conclusions unlikely (11:88).
Today, one continues to see articles appearing in popular magazines and local
newspapers reporting on the work of dowsers. No amount of scientific work has stopped
those who believe from believing in dowsing--or vice versa. A person who has
witnessed a dowser, especially one who has received setbacks in the form of dry wells,

from "professional" water people, will not be easily convinced that the dowser did not do
something special. Conversely, the non-believer continues believing the dowser to be a
charlatan or con man.

The Dowsing Problem

Dowsing, in a traditional sense, is the "practice of using a forked stick, rod,
pendulum, or similar device to locate underground water, minerals, or other hidden or lost
substances (14:3)." According to Graves, it is "a way of using your body's own reflexes
to help you interpret the world around you: to find things, to make sense of things, to

develop new ways of looking and seeing (8:11)."

7
 From these definitions, two classes of dowsers can be discerned. Those who

dowse purely for water or other buried objects and those who dowse for anything. These
two groups will be termed conventional dowsers and metaphysical dowsers.

Conventional dowsers, under this classification, are those that use a diving rod
and traverse the ground to locate subsurface matter like water and minerals or buried

objects like piping. Their divining rods are branches, wire and sometimes implements,
but most commonly a forked twig (15:26,27). They can be two branches taped together
to form a V-rod or a natural V-rod from a tree (8:77). Typical branches used for these
rods are willow and peach, however; cherry, apple, elm, hickory and others are also used
(15:26). Metal rods are L-shaped rods whose dimensions vary according to the dowser's
preference.

Metaphysical dowsers are those dowsers who attempt to dowse from remote
locations or utilize dowsing as Graves has suggested. They may use a pendulum, any
weight suspended from a string, or diving rods. Their beliefs hold dowsing as the answer

to everyday questions. For example, Graves suggests using dowsing to determine what
you are going to drink. You would ask your pendulum if you wanted to drink coffee, if
you wanted sugar and milk in the coffee, and the pendulum would give you the answer.
He suggests that the pendulum reflects what the body needs and not what the conscious
mind tells you. Your mind says you want coffee because you always drink it in the
morning. Graves says the pendulum may tell you that your body does not want the coffee

at all (8:104).
Remote dowsing fits into this area because the dowser believes he/she can locate
objects from a map or by asking the dowsing rod questions about the issue at the remote
area. Bird relates several stories of remote dowsing in his book, The Divining Hand.

There will be no attempt here to test any metaphysical claims of the dowser.
These claims, even if valid, are much too controversial for empirical experimentation

8
when the subject has not been proven, from a scientific standpoint, to provide better odds
than chance at solving a particular problem.

Importance and Usage of Dowsing

Dowsing could be an important field if the stories of success are the results of
actual dowsing ability. For example, a local landowner had drilled three successive dry
wells at his home site. When the dowser was called in, he located a vein of water and the
fourth well, drilled at the dowser's specified location, provided an adequate source of
2
water.
If the dowsers' claims are true, uses of dowsing for the military should be
developed. The most easily recognized skill is to locate water in remote locations. This
could be for troop support or humanitarian missions. Savings in time and manpower
from drilling dry wells would be greatly beneficial.
In the environmental arena, one could map high water flows in an aquifer, tracing

possible contaminant pathways. One could prepare a preliminary aquifer survey to

identify better ways to implement your remediation plan. Alternately, one may be able to
locate and trace a contaminant plume.

Interviews with Dowsers

Two dowsers were interviewed to help to understand the dowser and his skill.
Each individual has a distinct history of dowsing success reported in local papers and
confirmed by happy clients. They are unique and, though their experiences have little in
common, their processes share many features. Neither individual knows the other.

Mr. Lewis Carl

Mr. Lewis Carl is a Fairborn, Ohio resident who has been dowsing since he

witnessed a man dowsing when he was 12 years old. He started to apply his skill for

2
Story related to author by Mr. Lewis Carl during interview.

9
profit about 25 years ago. He has reported excellent success in locating water for

residential property owners and commercial land developers. By his account only one
well has failed, but due to the well collapsing and not due to water production.

Mr. Carl believes his skill is God-given and that only certain humans possess it at
his capacity. He feels that anyone can dowse with metal rods, but only a certain few can
utilize a tree's branch. He does not believe in metaphysical dowsing. Whenever he

dowses, he grips the V-rod with palms up and claims that he tries to resist any downward
pull. He locates water whenever the rod pulls down to a near vertical position.

Mr. Carl is knowledgeable in the geology of the area but has had no formal
training in geology. His knowledge is based on a need to understand what he is doing.
As such, he knows areas where water is easily found and areas where it is difficult to find.
He classifies water as that which is flowing and that which is not. He considers water
that is flowing as locatable and flowing in veins. He feels he locates the best place for a
well to be drilled based on volume of water which he differentiates by the pull on his V-

rod.
To locate a proposed well site, Mr. Carl usually starts by asking his client where
he wants the well to be located. He provides information to the client about locating the
well with respect to minimum distance from a structure and extra cost associated with

locating the well at extreme distances. He then starts to locate 'veins' in the area the client
has recommended. He walks briskly across the land and marks the direction of the veins

he locates. He attempts to find a point where two such veins cross. This crossing occurs
at different depths. He continues until he has located at least two such proposed well

locations. He marks each with a stake and writes a somewhat cryptic code to indicate the
two veins' depths found in the process described below. Additionally, he initials each

stake.

10
 Next, Mr. Carl estimates the depth of the wells he dowses. He has developed a

device for gau,-;ng depth and reports considerable success. The device, illustrated in Fig
2, is constructed of copper tubing and is 30 inches square. It has spiked tubing on the
bottom to allow the device to be pushed into the ground to be free-standing. It has string
strung at various angles from a vertex at one end. There is a bubble level on the top to
allow Mr. Carl to level the device before using it.
The device is placed ten feet from and perpendicular to the vein Mr. Carl has
supposedly located. He stands next to the device facing the vein and holding his V-rod.
He then crouches down so his hands are near the top of the device and releases his grip on
the rod. The angle the rod dips to is related to a string on the gauge. This gives Mr. Carl
and estimate of the depth to the vein of water. Although original in design, the process
follows traditional wisdom for dowsers, sometimes called Bishop's Rule, who seek depth:
"distance out equals distance down (8:49)."

30"

Bubble Level

30" • • GCopper Tubing

String to Gauge Depth

-- Ground Elevation

10' to Proposed Well Site

Figure 2: Lewis Carl's Depth Gauge

11
 Mr. Jesse Aites
Mr. Jesse Aites is an Oil City, Pennsylvania resident who has been dowsing for

over 40 years in the western Pennsylvania area (Appendix B). He too, believes his skill is
God-given. Mr. Aites has enjoyed similar fame and was featured in a 13-minute
television newscast in 1961 by KDKA Television in Pittsburgh, Pennsylvania.
Mr. Aites adds one twist to the dowsing equation. He was blinded in 1970. His

skill has continued. He dowses with several materials and says he can use as little as one
inch of a twig held between his hands to sense water veins. According to him, as he
approaches a vein, his toes start to tingle, like when they have fallen asleep. This
sensation continues until he is directly over the vein when he becomes nearly "paralyzed"
until he releases the diving rod.
Mr. Aites has also dowsed for oil in the western Pennsylvania hills. He has

currently stopped this type of dowsing because of his religious beliefs.

How Dowsing Works

The biggest question that needs to be answered is how does the dowsing rod
work. Graves believes that the reactions of the rod are due to involuntary muscular

reactions to the questions posed by you to the stick or pendulum (8:11). He provides
detailed lessons on how to hold pendulums, V-rods and L-rods. He suggests that each is
controlled by these muscle movements. To hold a V-rod, he explains that you grab the
stick with the palms facing each other, then you twist your forearms so that the vee points

outward and your palms are either up or down. "Twist your wrists outward slightly; the
rod tip should move upwards sharply..." (8:79) As your body receives the 'answer' or
locates the object, muscles will react and allow the stick to move (1:13).
A study by Irons on the behavior of force pendulums linked the motion of a

dowser's pendulum to the actions of the fingertips as they tightened, to increase the

12
frequency of the pendulum, or loosened, to decrease the frequency. thus lengthening or

shortening the string. (9:113-114)

These processes contradict what the two dowsers who were interviewed said they

did to dowse. Mr. Carl said he holds the stick in his hands with the palms up and resists
the turning of the rod. Using this technique and the device he developed, he can
determine depth of the water source he has located. Even more striking, Mr. Aites says

he only requires about one inch of a twig of a tree held between his hands. As he
approaches a water vein, he feels a sensation in his toes. As he crosses over the source,
he becomes immobilized until he releases the twig.
One of the items both dowsers agree on is that the water must be flowing. Mr.

Carl said he could walk up to a swimming pool and get no reaction. This is an important
point since most literature that contradicts dowsing or literature that describes dowsing
experiments, fails to mention if the water is flowing. One proposed experiment would
have you put water in jars. Randi's experiment did mention that water was flowing
(13:17)

Geology, Groundwater and Dowsing

There are several points of contention between the supporters and deriders of

dowsing. A particular point deals with the aspect of how water is described by dowsers
versus what is technically described by geologists and hydrologists. The dowser typically
refers to water as running in streams or veins (15:31). Their skill is in finding where
these veins are located and typically look for areas where veins cross or run parallel. 3
Technical skeptics refute the dowser's claim by showing that water lies under ground in

most of the world, so it would be difficult not to find water when dowsing. One geologist
said "the dowser [did not possess] any supernatural faculty [and he ascribed the dowser's]
success to a marvelously developed instinct and eye for the country (1:61)."
3
From description by Mr. Carl during initial interview

13
 Technical models of groundwater do not consider the idea of 'veins' (15:31).
Physical laws describe groundwater flow (7:2) and these laws consider groundwater flow
to exist like a field, similar to heat transfer and electric fields (7:11). Lithology, the
"physical makeup, including the mineral composition, grain size, and grain packing of the

sediments or rocks that make up the geological system (7:145)," determines how the
water percolates into the ground. As the percolating water enters the saturated zone it
becomes groundwater (7:2). Groundwater lies in aquifers as determined by the lithology.

The hydrologic cycle recharges aquifers (7:3), and water flows in these aquifers from
areas of higher pressure to areas of lower pressure (7:18).

A flow net describes the pressure in the aquifer. For visualization, the hydrologist
defines lines of equipotential pressure to show water flow (7:5 1). Darcy's Law, an
empirical definition, relates water velocity to the change in pressure with a
proportionality constant, K, defined as the porosity of the soil (7:17). Soils which are
highly porous have high K-values and water flows faster in this soil than in a less porous

soil. Gravels and sands have high porosity while clays and rock have low porosity (7:16).
The velocity determined by Darcy's Law, or Darcy's velocity, is not the true
velocity of the water in the aquifer. This concept represents a macroscopic viewpoint to
simplify the process of groundwater modeling and analysis. Water actually flows around
each of the various soil particles as it moves in the direction of the aquifer. This becomes
the microscopic viewpoint. (7:17) In studying groundwater, the microscopic viewpoint is
dismissed because it is too difficult to measure and model and is, in essence, not a
necessary study.
Herein lies a potential link between the two sides of this issue. The microscopic
view shows that water flows faster in some areas and slower in other areas as the water
moves around the soil particles. The hydrologist sees this variation in velocity occurring

at the grain size level. A dowser says he feels a vein or stream of water. His

14
classification is macroscopic to him. These two events may in fact be identical. Perhaps
reality incorporates both the macroscopic and microscopic viewpoints. The dowser may
be locating a region of localized high velocity water that cannot be accounted for by the
empirical formulas used in the hydrologist's macroscopic world.
A possible example of this phenomenon was shown during the March 1994 RREL

Conference in Cincinnati. A paper was presented that discussed the use of micro-
organisms to plug a water pathway that was created during oil recovery. The oil recovery
process required water to be pumped into the ground to flush the oil out. During usage,
water typically found an alternative pathway, one of lesser resistance. The micro-
organisms were fed into the water and the new pathway and then allowed to grow. When
they grew, they successfully plugged the alternative pathway. (4:95)
This process shows that water will take the route of least resistance in the ground.

As such, in certain pathways, the water will be flowing faster than in others. As erosion
of the soil particles progresses, a 'vein' could develop.
Even if this highly speculative idea was true, does the dowser provide the best
location for drilling the well? Consideration of this answer is beyond the scope of this
research and needs only be addressed if a verifiable dowsing response exists.

15
 Methodlgy

Almost every major scientific boner-and there have been many of

them-can be traced to a zealous desire to see the world as we
would like it to be rather than as it actually is (15:84).

With this quotation and keeping in mind its wisdom, the chapter that defines the
methodology and introduces the dowsing experiment is presented. There will be a
discussion of some proposed experiments and the problems presented by them. This
provides a foundation for developing our experiment. Next, an experimental
methodology is proposed. This will detail the steps to be followed to conduct the work in

the field. Finally, the statistical aspects of the proposed experiment are discussed. This
includes a recommendation for defining the range of results that lie in the realm of chance
and those that are beyond chance.

Proposed Experiments
Numerous dowsing experiments have been proposed by the technical community.
The best experiment would be to have the dowser go to a fully characterized site, specify

the location and depth of the well, and then drill at that location to verify his work. The
full characterization would include complete geological and hydrological analysis of the
site before letting the dowser visit the area. The experiment itself is difficult to set up and

costly to operate.
Randi's proposed experiment to construct a water circuit under a dirt platform also
requires a large budget (13:17). He constructed a 33 foot x 33 foot mounded area with
three water routes buried underneath it, 20" below the top of the mound.

16
 Carey suggested placing ten ceramic jars, one filled with water, before the dowser

for dowsing. The dowser would then attempt to pick the correct jar. Carey suggests that
two trials would be adequate to determine a successful dowser (3:74). According to Mr.

Carl, the dowser in this study, this method would invalidate his abilities since he says
water has to be moving for him to detect it.

Carey does provide discussion of the aspects of a 'good' experiment. He suggests

the experiment should meet two conditions. First, it shnuld b. something that "the

dowser clearly ought to be able to perform (3:74)." Second, the experiment must
eliminate other methods that may be used by the dowser to locate the water. He mentions
"cheating, coincidence, inadvertent cueing on our part, visual or audio clues" as some of

those methods (3:74).

Martin reported on an experiment that forms the foundation of this work (10:138-
139). His experiment used four plastic hoses covered by a rug in a room of a building.

One hose was randomly selected to carry water and the dowser attempted to locate that
hose. Dowsers could quickly dismiss this experiment because of the site. The structure
would have several 'influences' that would interfere with claimed dowsing skills.

Methodology for the Dowsing Experiment

This experiment will be conducted to meet all essential criteria presented by both
sides of the dowsing argument. To meet the rules of scientific study, the experiment will
follow a systematic approach that can be repeated. To meet the concerns of dowsers, the
experiment will be set up to address, as much as is practical, all possible influences,

accounting for any outside influences that may occur.

As mentioned earlier, the experiment proposed by Martin (10:138-139) provides

the foundation for this experiment. A flow network as shown in Figure 3 will be
constructed with plastic piping. See Figure 4 for the control manifold used for selecting a

17
 30' 1

(ypical) 20' (Typical

( Water Line

i
Screen
~~

Manifold 'A' Manifold 'B'

i Reservoir

Pathway to be
traversed by the
dowser

Figure 3: Schematic of Proposed Network

18
 To Each Line

Ball Valves Gate Valve

From Pump

Figure 4: Manifold 'A'

water route. A pump will be used to keep the water flowing. The network will be

constructed with sufficient distance between lines to insure there can be no claim of line
to line interference. Backflow preventers will be used at Manifold 'B' to keep water from
flowing into an evacuated line. The control manifold area will be screened from the
dowser. The site that the network will be placed on will be selected prior to construction.
The dowser will be asked to locate single path that he feels has no groundwater or
manmade interferences. This pathway will be marked with stakes before constructing the
experiment. If the area is remote, a generator will be used to run the pump. Finally, the
network will be set up after the site has been selected and the dowser will be tested as
soon as possible after construction. Specific test procedures are outlined in Fig 5.

19
 Dowsing Experiment Procedures

1. Select area prior to constructing system by having dowser

reconnoiter area for a location that dowser believes to have no
groundwater influences.
2. Set up network, controls and screen. Check system for leaks.
Insure proper operation and insure screen properly shields
valves.
3. Insure Manifold 'A' operator knows proper method for
determining random order of valve operation, proper recording
procedures and importance of not providing visual clues.
4. Test dowse with stagnant water in lines to confirm site remains
non-influential.
5. Set controls to run water in one line. Allow dowser to cross
network, knowing which line is flowing. Validate process is
working and dowser has no complaints. Resolve any issues.
6. Initiate trials. Record data for which pipe water is running
through and which pipe dowser selects. Plot results on graph
to determine number of trials to execute. Do not inform
dowser of individual results.
7. Continue through trials until the null hypothesis can be
accepted or rejected based on graph mentioned above.

Figure 5: Experiment Test Procedures

20
Establishing Statistical Validity
The purpose of this experiment is to determine whether or not the dowser is

guessing when he selects a pipe in the network. To make this judgment, one can compare
the dowser's performance against the performance of someone who is known to be
guessing. The known guesser would be expected to perform no better than chance. If the
dowser's performance is no better than chance, it would be reasonable to assume he was

guessing. If his performance was markedly better than chance, it would be reasonable to
assume he was not guessing. If the dowser's performance was somewhere in between the
two extremes, the performance would be difficult to classify. Therefore, comparing the
performance against a probability of success provides useful data, but it is not suited to

making such a judgment. An alternative method is required to speak with confidence

about the dowser's claimed skill.
Hypothesis testing provides this alternative. In hypothesis testing, two opposing

hypotheses are proposed and relevant data is collected that will support accepting one of
the hypotheses as the true hypothesis. Associated with each hypothesis will be a measure
of error, called Type I or Type II error. This combination of hypothesis and error

provides a better way of classifying the dowser's performance.

The following sections describe a binomial probability distribution and present an

equation for determining the cumulative odds for successive trials. After establishing
this, there is a discussion of hypothesis testing followed by a discussion of the
hypotheses to be tested. Finally, a method for determining when to accept or reject the

hypotheses is suggested.

Binomial Probability Distribution

The design of the experiment suggests a binomial probability distribution, that is,
only two outcomes for each trial are possible. The outcomes are either the dowser

identifies the proper line or he identifies another line. For these two outcomes, a success

21
will be defined as correctly identifying the water line with water flowing in it and a
failure will be defined as identifying some other line or failure to identify any line. For a
person who is know to be guessing, the probability of a success in this experiment is 1 in
5 or 0.2 for each trial. Using the binomial formula,

P(x) n! p, 0 - p)Y- (I)

x!(n-x)!
where,
x = number of favorable outcomes
n = number of events
p = probability of success

one can determine the probability of success for a series of trials. For example, if the
dowser was to cross the network ten times, then the probability of 10 successes is:
P(10) = 10! 0.2'°(1-0.2)('°-') = 1.024x10-' (2)
10!(10-10)!
Table I presents these probabilities for the first 30 trials. As can be seen, the probability
of the dowser correctly identifying the proper line on every trial diminishes quickly.
Although the probabilities diminish rapidly, one cannot establish regions where
you are certain that the dowser's abilities have been supported or unsupported. For

example, suppose the dowser selects the correct line five times out of 20 trials. He has
beaten the odds, but one cannot say with any confidence that the dowser was not

guessing. Perhaps his next five trials would not be successful. He would then be back at
the experiment's basic odds. Some other means must be used to establish what is "better

than chance."

Hypothesis Testing
Hypothesis testing provides a better means to evaluate what is better than chance.
According to Devore, hypothesis testing is a method "to decide which of two

contradictory claims about [a] parameter is correct (5:283)." It is a process of

22
 Nn 0 Ln N LO I w
L *, ý9 -9 9 9

Inw w w w w '

NR
Nq CR
M

0 9 9 0 0 0 0P -00 N O
4bw ILw ow w w -b NjL

N 0 0 9O 9. 0 ~0 . ~ ~ 8

N 0 000 0 N -0 0

0N 00 0009 004 0 - N% N& 00-

, ow
0 1w toN 2 ninr g 8
N NO

a, N 00000 00 0 00

CL 0 Cw e C4 N
- 0w4 M
w. .X 0 M r-

NN8!2 M 8N
O 8m0 8 ! 0 14 1 .
0 w0 C.) 0 Na 0 0M 0N0 - 0
£ ~ ~ ~ 4 , ,C11 00N4)0N0
- C 0 C)0
0q 6 't88a
0 ON NON 00 C.) 0 0 NO 40
0. C 0 0. 00 0
0~~ 0 CN 00040 M~ 0 -M~ M0N P 00

00 00 0'0
00 N 0
4 0
N 0

U).0 000 0 0 0 0 0 0
V N8 8N 00
'a66o0 0 a'W

000 d 6 NO ~ 6NO06 ON 00- 0 410 - N

0 0 N N 00 NO C) 0040 a r) N OMON 40w
0 nC ; ( 0
,. 0 0 0 C.) - 0 N0
- 0 -w 14 T
. 01) 0. . ON

U)
r,
0666 61600
A
I , - 0 M C
00 41 002 goooo
W0 O 0 0 A. 1 0 C) 0 - - .) - 0 0
0 )0 NO 04 0 ) 04 N 0
0
0000000 0 0 00 - N 0 40 00 N N N6
0
10 0 0 0000 M oo0 o -0:00

o
.00
~ C 4 . N N 0.0 NO C. 00
N N
N N
0
NO
.00
00
N
0 000.
NC)0000 00 006000
0 Mo N 00 N ) 0 N 0 0 0 0 0 N

04 ONr, 00 A 00 w. 0O N M10 N 000 N enN

-iC) N 0 0 0 . . 0 0 000 0
.401x 0 O ' O OON M O , O . 0 4 ON 00 Cw.)C
N
* 0 M M Ln - M o .a,-N .- o. o.
06 66 6 6 6 0o %Nn6 0 68
00
O~ ~ N8000N0 4 0-4 0-
0 C 0
-!2
NI N
0
0--
0
40
0
N
00
N
ý 0
00I)0
I.
10-0
NON
0.
-
UP

N
.
)
0
1
.)NN
00
N
IN
N
0
0
0O 0
0
6.

) 6 -t 00- N O C4.) 1 0 0 ("1 Ný N 0 N 0 . 00 NM

1 0 0 0 0 0 N- - -I4N N N N. 0~N 1N N 1 N C1

TableN N : dd fo Sucssv Bioma TrNials.. 0

~ 66666 6ooo26
 N I -r

o€ w.

O
N

I••
" "•" w" u

ww h

N M ml c4
CN W

M. c-,(4 e4u

N nN
v! V N ~
(l N!

•"o
cc
co •
V Nin
• - -"
~ N ( N
- c4 c"•

I--

0 09 9 lowih
k- w 7 V
I": WW
""'ý -" - W
R N
c - " w

C4 cq
v!
- - •w!
It*,
N
' 9- • S.•
l ii La a, L w wii
2h N1

Tal 1:Od oUucsieBio ilTil Cn.

= U-24
statistical inference (5:283) used to help one make decisions about a population by
evaluating a sample from that population. In the case of the dowsing experiment, the
parameter is the probability of success. The two contradictory claims are called the null
and alternate hypotheses with the null hypothesis stating the accepted point of view.
Whenever testing a hypothesis, there is a possibility of making errors (5:283).
These errors are caused by the fact that one takes a sample that is supposed to be

representative of the population. There is always a chance that the sample is not
representative and making statistical inference from an unrepresentative sample would
cause one to accept or reject the wrong hypothesis. These errors are called Type I and
Type II. A Type I error is an error caused by rejecting the null hypothesis when it is true.
A Type II error is an error caused by accepting the null hypothesis when the alternate

hypothesis is true (16:16; 5:286). Associated with each type of error is a probability of
making that error. For Type I errors, this probability is named a and for Type II errors, it

is named P3.
Once the hypotheses are stated, the number of samples to take to determine
whether to accept or reject the null hypothesis is decided (16:13). From this point, data is

collected to test the hypotheses and when evaluated will provide information necessary to

make your judgment.

The problem with this approach in an experiment involving humans is the
possibility that you may not complete the required number of samples during the course
of the experiment. The subject could quit, or become sick during the experiment forcing

cancellation and incomplete data collection.

Abraham Wald proposed using sequential analysis in situations where the number
of observations was not determined before the experiment (16:1). Sequential analysis, a
statistical inference method, results in fewer samples being taken than would be required

25
by non-sequential methods (16:1-2). The design of the dowsing experiment and the use

of a human subject suggest that the analysis of this data utilize Wald's methods.

Wald's Equations

As a special application of sequential analysis, Wald explained how to use his

theory in lot selection during a manufacturing process (16:88). The lot is accepted or
rejected based on a series of inspections that determine if each randomly chosen item
from the lot is defective or non-defective. The inspections are recorded on a chart until
the plotted data breaks one of two parallel lines constructed on the chart, or a table that
indicates how many rejects you can tolerate. The area bounded by the parallel lines on
the chart represents a gray area established prior to beginning the inspections by use of
"practical considerations" (16:89) These considerations are discussed below. First, one

needs to study the equations and the effects the parameters have on the acceptance and

rejection lines.
Wald provides his equations in two formats to assist people manually generating
data for his test. His first equations are used to generate acceptance and rejection
numbers. The numbers would have been tabulated before the inspection and used to

determine when to accept or reject the null hypothesis. They are as follows:

P loglPO
log-
log Ilog
2E] - Slog log
log-
am =a P +11? (3)
PO l -po PO I -p0

and
1
log
log 1
rm.rmlog pi - log
cc "1--7--+ p -10I-PI1o-I(4)
109k]o (4)
P0 l -pO PO -pO

26
where:
am = Acceptance Number
rm = Rejection Number
m = Trial number
pO = Lower Bound of Probability

PI = Upper Bound of Probability

cc = Probability of Type I error

03 = Probability of Type II error

Wald's second set of equations defines the slope, the same for both lines, and

intercepts of the ines used to generate the graph ic: data collection. This graph is easier
to use than the table since one can see the only three possibilities required to work with

his theory--accept the null hypothesis, reject the null hypothesis, or continue testing.

The equations for the graph are as follows:

log _

ho = a (5)
1 -PI
logP - log
p0 1i-po

log I_ _
hio= a (6)
lg p0 lg- p0

logppo -Ilog i- po

lo 1 2 (7)

27
where:

h= Intercept for Acceptance Line

h - Intercept for Rejection Line
s = Slope of the lines
Po = Lower Bound of Probability

P1= Upper Bound of Probability

a = Probability of Type I error
03 - Probability of Type II error

Analysis of the Parameters

Equations (3) through (7) rely on four parameters pO ,pl,a, and (3. As stated,
Wald provides no guidance for determining the values of these parameters. His only

discourse on the parameters is a discussion of the effect changing the probability has on
your preference for accepting or rejecting a manufactured lot of material. Per his
discussion, if the observed probability of a defect occurring is greater than the true
probability, your preference is to reject the lot. This preference increases as the observed
probability of failure increases. The reverse is also true. As the observed probability of a

defect occurring decreases below the true probability, your preference is to accept the lot

(16:89).
Two of the parameters, a and (3,are associated with the hypotheses as discussed
earlier. The other two parameters, Po and pl, are associated with the probability of the
event occurring. In the context of the dowsing experiment, this would be the probability

of success for a person who is know to be guessing, p'.

Rather than working with a lot of material, this experiment uses all trials that
could possibly be accomplished in the experiment as the lot. The null hypothesis can be

stated as the dowser's ability is no better than chance. If the experiment tests the dowser's

28
reported ability, then, each attempt to dowse would be considered a sample. Each sample
has an associated true probability of occurring related to the probability of the person
known to be guessing, p, and an actual probability of occurring related to the dowser's
true ability to dowse, p. Therefore, analogous to the previous discussion, if the observed
probability of the dowser's ability is greater than the true probability of the person known
to be guessing, p > p', your preference is to reject the null hypothesis. This preference
increases as the observed probability of the dowser's ability increases. And, as the
observed probability of the dowser's ability decreases below the true probability, p < p',

your preference is to accept the null hypothesis.

As previously discussed, the probability of correctly identifying the line with
flowing water, if the dowser is guessing, is 0.20. This value becomes p'. The values for

PO and pl are established to fit the preferences stated above with p0 <p' and pl > p'.
Thus, define PO <-0. 2 0 andpjl >0.20.
Wald states when the observed probability is close to the true probability you
could accept or reject the null hypothesis without significant consequence of error

(16:89). The key, then, becomes establishing pO and pI where you believe there would
be a consequence in making the wrong decision.

Reviewing the equations and plotting different combinations of the parameters

helps to create an understanding of the effect each parameter has on the resultant test.

Using a graphing program and equations (3) and (4) one can quickly process several
combinations of parameters. Figures 6 and 7 show two combinations of the parameters

with a and P3
held constant.

29
 I Reject Null Hypothesislne I,

r!Ii I 111II K
!I j
14 - -
i•I I !.,PF! Ii ii -Accept Null Hypothesis Uine.

23 4 5 8 7 8 9 1011 121t314 15 1617 18192021 2223 2425 2627 2829 30

Trials
Figure 6: Graphical Test with a• 0.05, J3 0.10, P0 = 0.10, P1 = 0.50

Reject Null IlHypothesis

l, ll LineL-i- I I~t
| t __ ,e
12 i I l " I . I I
Fiue6 Gahcl Test wiha 00 , 0.1, PO 0 0 I05

- t - - - -- l -t 1i - r,'*
-I-k' -V-*:T i t ,

z11 cccept
cept
Null!
Hpt1
T 1sl11L Null Hypothesis Line-I

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 21 30

Trials

Figure 7: Graphical Test with a = 0.05, P= 0.10, P 0 = 0.18, P1 = 0,25

30
 These figures show the effect on testing as the values for PO and P1 change.
When PO and PI are close to the true probability, the gray area is large and a large

number of samples would have to be taken to break either line. When PO and P1 are
farther away from the true probability, the lines are closer and this lessens the number of

samples that must be taken.

Additional analysis of the graphs using other combinations of the parameters and
review of the Wald's equations shows the following relations between the parameters.
First, a and 1 control line separation, with a affecting the "reject null hypothesis" line
and f3 affecting the "accept null hypothesis" line. PO and Pj affect both lines as their
values change and affect the slope and line separation. See Figure 8 for a explanation of

these relationships. With this information the experiment's values for these parameters

can be chosen.

Parameter Change Slope in Affected Analysis

n.Reject Conduct more testing before

a nH 0 rejecting the null hypothesis

n.c. Accept Conduct more testing before

H0 accepting the null hypothesis

A A A both Quicker acceptance of H. more

PO testing to reject H.

V both
A A Quicker acceptance of H., less
testing to reject H0

Figure 8: Wald's Equations Parameter Analysis

31
Determining the Experiment Parameters
Before the parameters are chosen, the hypotheses must be stated. Wald implicitly
stated his hypothesis: the manufactured material is acceptable. The inspection of pieces

of the lot supports or rejects his hypothesis. In the dowsing experiment, the inspection
becomes observation of the dowser and the data to record becomes his success or failure
on each trial. As such, the hypotheses will be stated as follows:

Ho : The dowser's claimed ability is no better than chance, orp < 0.2.
Ha : The dowser's claimed ability is better than chance, orp > 0.2.
Although Wald suggests choosing values for Po and PI before aL and 5 (16:89),
there is no practical difference when these values are chosen if you understand their
importance. Initially, a could be set equal to P3. In doing this, one assumes that the effect
of making either type of error is the same. Reviewing the errors shows that Type I errors
are more serious, that is, will lead to further study and expense, than Type II errors.
Hence, Type I error probability should be reduced to give more confidence in the results
if the dowser's effort leads to rejecting Ho. Therefore, based on the previous discussion

of Type I and Type II error and knowledge of typical values, set a = 0.05 and 03 = 0.10.
These values give acceptable confidence in the experiment's results and represent

commonly used values.

Establishing PO and P1 proves to be a matter ofjudgment. One can contrive a
suitable combination that creates a small enough gray area to allow for around 30 trials

before a 'chance' operator would be rejected. That is after 30 trials, the 'chance' operator
should have guessed correctly about six times. At this point, the plotted data would break
the 'Accept Ho' line and the experiment would be discontinued. Thirty trials was chosen
at random. With P3established, trial and error provides a value of P1 = 0.40. This value
would require the dowser to perform near this probability before the null hypothesis
would be rejected.

32
 Knowing that a value of Po close to the true probability would make a wider gray

region, you could make the region sufficiently wide to require an excessive number of
trials before rejecting the hypothesis. Doing so would only increase the effort required to
test the hypothesis. Additionally, the stated purpose in using Wald's equations was to
limit the number of samples. Without further guidance, the value for PO was based on a
proportional evaluation of PI. P1 could range from 0.20 to 1.0. That is a range of 0.8.
Twenty-five percent of that range, 0.8 x 25%, is 0.2 and PI =p'+ 0.2 = 0.40. Similarly,
the value of Po is also 25% of the range it could take. The range is from 0 to 0.20, and
0.20 x 25% = 0.05. Po = P' - 0.05 = 0. 15.
With these values determined, Figure 9 shows the chart that will be used in this

experiment. To recap, for the experiment the parameter values have been set at a = 0.05,

PO = 0.15 and PI = 0.40.

S=0.10,

33
.0

Co

o_ Co
7eý 70
M--

40-4

CS,

434
 Results

This chapter presents the results of a field observation used to learn about Mr.
Carl's technique and the field experiment. It is organized chronologically and discusses
an observation at a well site, the selection of the experiment site, setup of the equipment
and conduct of the experiment.

Proposed Water Well Site, State Route 35, Urbana

A site visit was made at this location to witness Mr. Carl's dowsing process on 4
June 1994. The site was a two acre plot that rose from the road to a small plateau area.
The owner wanted the well along the southern boundary of the site towards the front.
The area for the well was sloping at about a 3:1 grade. Mr. Carl located three spots that
he felt had veins of water running through them. He said the first location was not very
"strong." He recommended the second location at which he said had two veins crossing,
one at 60 feet and the other at 80 feet. The other point had the 80 feet deep vein and
another vein running parallel to the first 60 feet vein at 61 feet. See Fig 10.
A follow-up was conducted on 6 June 1994 when the well was being drilled. The
drilling rig operator had difficulty aligning the truck with the stake because of the grade.
He placed the rig over the stake, but when it was leveled the well was spotted about three
feet down gradient from the place marked by Mr. Carl. The location was close to the line
of the 60 foot vein he had spotted. The rig was a mud rotary unit. Drilling time was
about two hours.

35
 61' Deep Vein 64' Deep Ve•n

Proposed Well
Location

80' Deep Vein

IActual
/ I~rjlr-Locatilon
Well

Downhill

Figure 10: State Route 35, Urbana Site

The operator was attempting to locate a gravel lens at 95 feet. He had obtained
his data from the state's geological survey on the site. He uses his experience to tell him
what kind of material he is drilling through. For example, the drilling rig moves
smoothly and steadily through a stiff clay. When the bit hits a rock, the rig reacts by
jumping' about and shaking. When the operator hits gravel, the bit moves rapidly. the rig
shakes and rattles, but the drive chains remain taught.
As the drilling passed 60 feet a rock or very thin layer of gravel was hit. The

operator did not consider stopping at this location because he believed the obstruction
was not a gravel lens that carried water. He continued drilling and located the gravel he

was attempting to find at around 97 feet.

At that depth, he checked the well drillings to confirm that he was in gravel. He
collected a sample and then washed it in a bucket of water. This revealed stones in the

36
cuttings. He was satisfied that he had located the gravel lens and prepared to drive the
well casing. The well casing was screened about eight feet. A screen is an area of the
well casing that lets the water in the aquifer into the well to be pumped out. See Fig 11.

This screen was about three feet from the bottom of the casing. The operator drove in
five 5-inch diameter by 20 foot long sections of well casing. This established a 100 foot

deep well.

Pump and Water Piping

Clay Well Casing

Ar - Sand Packing

Gravel Lens
SScreen

Water Flows From

Clay 7 Gravel, through
Screen and Up Piping

Figure 11: Typical Well Screen

The operator then purged the well to determine if the well was producing water.
The well was purged by high pressure air driven through the bit shaft. Water was then
forced out the well casing. If the well was not producing, the water would stop flowing
out of the top of the casing. A producing well would provide continuous water flowing

37
out of the casing. By the driller's estimate, the well was producing about 10 gallons per
minute.
As far as this site is concerned, it provided little information towards proving or
disproving the dowser's claim. The drilling did not take place at the location the dowser
had marked. By his own accounting, you should drill at his marked spot or you will miss
the 'vein.' Since the driller did not stop to see if there was water at 60 feet, there is no
way to determine if the dowser had located a 'vein' there. Without better data, this site
added only inconclusive information to the dowsing argument.

Selecting the Site

Selecting a site for the experiment proved to be a difficult task. Mr. Carl found
numerous 'veins' throughout three prospective areas. Alternatives were discussed that
required a reduction in the distance between the piping network. After looking at other
sites an area was found, Figures 12 and 13, that Mr. Carl described as having parallel
veins running north to south. They appeared to be regularly spaced. A baseline was
established by placing two stakes 100 feet apart, approximately perpendicular to the
'veins'. A string drawn between the stakes defined a line and along the line stakes marked

each vein.
The stakes ranged from about six to 12 feet apart. A portion of the line was
selected where the stakes appeared to have the widest spacing. With the dGwser's
agreement that the water lines would be placed centered between the stakes and parallel
to the veins.

38
 Possible Sites

TENNIS COURTS

Selected Site

Figure 12: Site Location Plan

39
 Generator-/

's R

-r,4

- ~- "Veins" reported by
Mr. Carl prior to
experiment set-up at
intervals from 9' to 12'

Figure 13: Site Layout Plan

40
Setting Up the Equipment
Appendix C lists the materials used in the experiment. Not all the materials
procured were required because several things changed after the initial planning. First.
the pump used in the experiment was not capable of producing a large volume of water.
During a pump test to determine rates, the pump, rated at 24 gallons per minute (gpm) by
the manufacturer, produced a little over 3 gpm. Upon investigation, the check valves

used to keep water from back flowing through the system caused the reduction in
pressure. When the check valves were bypassed, the rate jumped considerably. This
issue was discussed with the dowser and he felt the lower rate might cause him some
problem. To solve this issue an alternative manifold that eliminated the check valves if

the need arose was provided.

The piping system was laid out the day prior to the scheduled experiment.
Because of the locations of the 'veins', the overall system layout was reduced to

approximately 10 foot spacing rather than the 20 foot spacing initially sought. The stakes
that were placed to mark the veins were left in place and each line was placed at the
midpoint of two stakes. Otherwise, system layout remained as designed.

A four foot high screen was built around Manifold 'A' to allow the manifold
operator to sit behind it and be completely out of the dowser's sight. Tarpaulins were
placed around the frame. As an additional safety measure, cardboard was placed over the
manifold to insure shadows were not cast on the tarps from the valves.

Power for the pump was provided by a portable generator. It was placed
approximately 70 feet from the pump, but close enough to the network to provide

masking noise.
All observers checked the lines to insure there was no signal, such as sound or
vibration, the dowser could pick up from the system. All agreed the system provided no

information that could be used by the dowser to discern the correct pipe.

41
Conducting the Experiment
The field experiment was conducted on 16 June 1994. The day was a hot, sunny
day. The temperature was in the high 80's and the relative humidity was over 90%. The
experiment took about two hours to run. The dowser located the line with flowing water

in 10 out of 27 trials. In doing so, the dowser broke the 'Reject Null Hypothesis' line
indicating that Mr. Carl's skill is better than chance.

The experiment was observed by the four member thesis committee, the
researcher and one disinterested person. The disinterested person and one committee
member were used to operate the valves and record data at Manifold 'A'. All other

observers and the dowser remained on the network side of the screen. This produced a

double blind experiment situation.

The procedures were briefed to the dowser before commencing the experiment.
All steps stated in the procedures were followed starting with step three since the first two
steps involved set-up. The dowser was asked before he started if he had any questions or

comments about the proposed experiment. He stated that he had no questions and that

the experiment should be a good test of his skills.

Mr. Carl brought both a V-rod and angle rods to use for dowsing. Initially, he
appeared to have no trouble with the V-rod, but had trouble as the experiment progressed.

He said he was getting a reading from all the pipes and was having trouble discerning
which was the correct pipe. This was contrary to his previous statement that he got no
reading from still water. He was becoming tired from the heat and appeared frustrated at
his inability to easily discern ihe correct pipe. Trials II through 17 had been failures
(though no one but the operators knew this). Because of Mr. Carl's appearance, one of

the observers suggested Mr. Carl take a break to rest and get something to drink. After
this break, he quit using the V-rod and finished the experiment using angle rods. The

dowser's performance improved considerably. He had three straight successes.

42
 At trial 20, Mr. Carl said he knew that the line could not be a particular line
because that line was just selected. At that time the researcher explained to him that each
trial was independent and that any line could be chosen a number of times in a row, based
on the random number tables. Although his performance was not perfect, he did seem to
understand this issue. Trials 18 through 27 included only three failures. See Table 2 and
Figure 14 for the results.
Mr. Carl's technique appeared straightforward. He started at one end of the
network and when queued to go, he quickly walked across the pipes to the other end. On
occasion, he returned to specific lines before stating his answer. He never touched any of
the lines. He normally started at one of the 'veins' that were marked before the
experiment and then paced across the piping. At times, he would stand sideways next to

a line and then turn himself so that the dowsing rod would go over the line to get a
response. While using the angle rods, Mr. Carl placed his thumbs over the rods to keep

them from moving when he crossed one of the 'veins' he had located prior to setting up
the equipment.
Afterwards, Mr. Carl was debriefed about the proceedings. He classified the
experiment as a good experiment. When asked about the random order issue, he stated
that he did not understand the point initially and it had confused him during the early
portions of the experiment. This caused him to guess more often when he could not

easily tell which pipe was running. He could not explain why he was having trouble with
the V-rod. He did remark that the branch was not freshly cut, indicating by his words that

there was some importance to the issue.

The significance of these and others issues requires further analysis. The next

section provides this analysis.

43
 a " 0.05, I -"0.10, P, - 0.15, P1 - 0.40
12 7-
Reject Null Hypothesis i

10

Discussed independence T
Issue with Mr Carl here

S
9
S 6 Continue Testing - -1 . , t 1 I

4 I

2 - " I" !-....,Took long break here

Accept Null Hypothesis

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
Trials

Figure 14: Graph of Experiment Results

Table 2: Experiment Results

Trial Valve Mr. Carl's Trial Valve Mr. Carl's Trial Valve Mr. Carl's
Opened Pick Opened Pick Opened Pick

1 1 4 10 5 5 19 1 1
2 3 2 1 1 5 2 20 2 5
3 3 4 12 1 5 21 5 5
4 3 3 13 3 4 22 2 1
5 5 5 14 2 3 23 3 3
6 1 2 15 3 5 24 1 4
7 2 5 16 5 2 25 2 2
8 1 3 17 2 5 26 5 5
9 2 1 18 2 2 27 1 1

44
 Analysis and Conclusion

The results of this experiment were unexpected. After conducting the initial
research, this researcher believed the experiment would quickly conclude with data
supporting the null hypothesis, that is, the dowser was no better than chance. Mr. Carl's
performance surprised everyone attending except the dowser himself.
This section analyzes the experiment, discusses the implications of these results,
recommends further study and provides concluding remarks for the research.

Analysis of the Experiment

Analysis of the experiment requires review of the experiment with respect to the
process and analysis of the data from the experiment. Four aspects of the experiment
process will be reviewed: appropriateness of the experiment process, failure to explain
random procedures and trial independence to the dowser, implications of taking the break
at Trial 17, and changes in the parameters and how they would effect the results. The
experiment process will be reviewed first.
Experiment Process
All observers agreed before the experiment that the set-up did not provide the
dowser with any inadvertent cueing. One observer instructed the researcher to do a post-
experiment interview with the dowser to be sure to get his after-the-fact feelings on the
adequacy of the experiment. The intent, expressed to the researcher, was to obtain
feedback from the dowser on why he failed. The feedback would provide a basis for
countering any arguments presented by the dowsing community as to the inadequacy of
the experiment. Since the results indicated rejection of the null hypothesis, the post-
experiment interview required only questions on the issue of independence and what had
happened at the time of the break at Trial 17.

45
 The placement of the observers proved fortuitous. The researcher planned to have
the assistant at the control manifold. One of the observers volunteered to sit with the
assistant at the valves to provide oversight. This observer was a known skeptic. By
being there, it can be assumed that the operation of the valves was proper and according

to the experiment protocols.

The remaining observers, the researcher and the dowser remained on the network
side of the screen. Care was taken to insure that no one could discern what was going on
at the valves. Because of this, there could not have been any inadvertent cueing on
anyone's part to clue the dowser into which line to select. These features insured the
experiment was carried out in a double blind fashion.
The set-up provided only three options for the dowser to select the proper line:
guessing, dowsing or trickery. If trickery could be eliminated, then the experiment would
have met all the requirements defined by skeptics like Randi and Vogt & Hyman.

Analysis of the set-up, operation, placement of observers and observation of the dowser
leads this researcher to conclude that no trickery took place. The piping system was
checked for leaks before the start of the experiment. All joints were solid. The pipes
were checked for signs of vibration that may be sensed by the dowser. None of the

observers felt vibrations in the pipes along the dowser's pathway. There was a small
vibration noticeable at the valve manifold, but because of the distance from the dowsing
pathway and the flexibility of the piping, it was dampened before reaching the pathway.

One observer paid particular attention to watching if the dowser touched any of the lines.
At no time was the dowser observed touching a line. The dowser could not hear the
water running because the system was extremely quiet, the flow rate was low enough to
limit noise and the generator provided considerable masking noise. All known avenues
for trickery were eliminated. With dowsing or guessing as the only two options, one can
be confident in the results.

46
 Failure to Explain Random Procedures
By failing to explain the issue of a random procedure to the dowser two issues
surface. First, what was the effect of this misunderstanding on Mr. Carl's effort. In
answering the questions, Mr. Carl expressed that he did not understand the issue of
independence before the experiment started and did not fully understand it even after the

initial explanation was provided at Trial 14. It was not until the second explanation was
provided that he understood the issue. Between the first and twentieth trials, Mr. Carl
faced five occasions (trials 7, 11, 15, 16, 17) where the line that was flowing was the
same line he had previously picked. If these trials were removed from the analysis the
results would not have changed, but the experiment would have concluded earlier
providing the dowser with a better overall performance of 0.45.

The second issue of this aspect deals with a breach of protocol in scientific
process. By talking to the dowser, the researcher interrupted the experiment. It should
have proceeded to its conclusion without the explanation being given. Many would say
this would invalidate the entire experiment. The researcher should have insured the
dowser understood a random experiment. In defense of this breach, there was no intent to
interfere with the results by providing the information. And again, removing the
instances where the dowser was faced with a flowing line identical to the previous line
selected would not change the results. In addition to the five trials mentioned earlier, the
dowser faced only one more incidence of duplicate line selection. If one assumes that
Trial 21 should have been a failure, the dowser still would have broke the "reject null
hypothesis" line if his trend of successes continued. These two possibilities are shown in

Figure 15. Note, however, neither of these alternatives proves adequate. Messing with

the data set only leads to further non-productive speculation on the results.

47
 c" 0.06, p 0.10, Po 0.15, P, OAO
12
Reject Null Hypothesis Original Data

10 t a!1jDa
wth T als 7 ,1llI,
J4 ___._L 16,17 and 21 Removed -- J.-
S8-

C
e3 6 l.. I / "' -. / .,

e
6

--- --- --. Data with Trial 21 -

2i T-,-7 T- i ,
=4-T IAdFailure Calledpa ,

1 Accept Null Hypothesis

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

Trials

Figure 15: Experiment Data Plotted with Removed Trials

The Break at Trial 17

The break that was taken at Trial 17 could not have come at a better time for Mr.

Carl. If he had continued at the pace he had set in the previous seven trials, the

experiment would have ended with the opposite conclusion. An after-the-fact analysis of

the parameters shows that a small change of P1 to 0.43 or P3

to 0.15 would have caused

Mr. Carl to break the "Accept Null Hypothesis" line. When asked about the break, Mr.

Carl stated that he was feeling extremely fatigued prior to the break. Choosing a line was

taking additional effort because all the lines were giving him a reading. During the break

48
Mr. Carl decided he was not getting a clear signal from the V-rod and stopped using it.
His effort improved considerably and during the last ten trials Mr. Carl pick the correct

line 70% of the time.

Changes in the Parameters

As stated earlier, small changes in PI and P3
would have caused the dowser to
break the "accept null hypothesis" line at Trial 17. Figure 16 shows these changes

graphically.

a 0.05, P - 0.10, P0 -0.15, P, "0.40

12
Reject Null Hypothesis ' IOriginal Data - I

10 ! -- "Accept" Line with Pu 0. 15

s 8I "Accept" Line with P1 0.43-,- 1

e 6 Original "Accept" Line--.,, ,

001

o4.... -! - Accept Null Hypothesis

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 192021 22 23242526272829 30
Trials

Figure 16: Experiment Data Plotted with Changed Parameters

The values for these parameters could be scrutinized for the near failure at Trial
17. For the value of P1 , the seemingly small change may lead one to suggest a higher
value should have been chosen. However, remember that the experiment was designed to

49
see if the dowser was better than chance. The value of P 1 defines a breakout point where
you would reject the null hypothesis. It would be unreasonable to increase the value.

The test already requires the dowser to perform nearly twice as good as a known guesser
would perform. To require the dowser to perform to a higher standard for this experiment

is not necessary.
Changes to the value of [3 may also be suggested. Recalling that [3 is the

probability of accepting the null hypothesis when the alternate hypothesis is true,
changing the value would only increase your chance of accepting the wrong hypothesis.

Although the dowser's claim would have been refuted. one could not be as confident in
the results. Additionally, it could be argued that increasing the value of 03 while
maintaining or reducing the value of a is stacking the experiment against the dowser.
Doing so would destroy any objectivity the experiment has maintained.

Implications of the Results

The results of this experiment will be controversial. Prior to this experiment,
there has been no empirical data to support the notion of dowsing. If one believes this
data and the previous conclusions then the data directly supports the notion. The
statistics indicate, with a 95% confidence level, that Mr. Carl can dowse for flowing
water in a network system and better than one could who operates by chance. Discussion

of the implications of the results will provide a basis for further study into the experiment

and its results.

The experiment did not prove that a dowsing sense exists. As a preliminary
study, caution must be used before making such a claim. The experiment does provide
support for the notion of dowsing. Mr. Carl appears to have a unique skill. However,

there was no data collected that indicated that his skill was any better than the skill of a
trained geologist, hydrogeologist or well driller at locating the best place for a water well.

50
 These results suggest much more work needs to be done regarding this

phenomenon. Additional experiments following the procedures established herein should

be conducted on Mr. Carl and other reputable dowsers. Revisions to the experiment may
be warranted and may provide different results. Refinement and agreement on different
values for parameters used in Wald's equations should occur. These items need to occur
before this experiment can be classified as a repeatable experiment that provides
consistent results.
Mr. Carl's claim to be getting readings from the stagnant water lines was contrary
to an earlier claim that he got no reading from still water. Prior to the experiment, the
dowser had commented on this as a possible problem. He had explained to the researcher
that he tried to locate water in a hose at his home. He said he was getting a reading from
the hose even after it had been drained of water. This apparent inconsistency was
interpreted by the researcher, before the experiment, to be the basis for a future
"explanation" as to why the dowser could not do better than chance in the experiment.

Since the dowser did better than chance, no one will know if this was to be the basis for a

failure excuse.

Conclusion
Mr. Lewis Carl has shown this researcher that he has a unique skill. During the
course of the study, both points of view regarding dowsing were accepted as plausible at
some time or the other. Vogt and Hyman's Water Witching, U S. A. seemed to close the

discussion on the subject. Their conclusions are compelling and anyone who reads their
work should come to the same conclusions.
Meeting and watching Mr. Carl allows one to witness a man who truly believes in
his ability, has set very specific limits on its application and refuses to believe many of
the claims other dowsers report to have accomplished. Several things that Mr. Carl does
while dowsing don't fit into many of the definitions that have been proposed. For

51
example, during one of the times Mr. Carl was dowsing, the researcher videotaped his
hands to see if there was any movement of the hands that would cause the v-rod to turn.
Though there was no formal evaluation other than slow motion observation, this
researcher could not detect any motion of the hands other than would occur while he
walked. On another occasion, Mr. Carl stated that the rod make a noise while it turned in

his hands. The noise, similar to one made while gripping a finger in the palm of one hand
and then twisting the finger, could be clearly heard, but not explained.
hlere are many aspects of dowsing that were not addressed by this experiment
and these results do not apply to them. Additionally, many people claim to have a
dowsing sense but are not reputable. Beware of anyone who wields this work as proof of
their skill. As in any scientific experiment, repeatability, under the scrutiny of the
scientific community, is the determinant of valid results. Unless this validation occurs,
these findings will remain experimental data.
These things combined with the results of the experiment suggest that something
has occurred that cannot be explained. Additional research as suggested earlier is
required to substantiati, that a dowsing sense could exist.

52
 .00
00 IT
o% ou 00 m 00
00 =%'00 0%

CA 12
-1

-5-

0 -0

0) 0

CC4

4)4_ __ _ _ __ _ _ _

4,) 40. 4t.

00

0 0

0.0 4.
0 0 0 0 0* .0c

0 4.

U)> 440

ON0.

530
00 0
oo a, 00c0 t-
ON t- 1-
coNN0%0

0%~o c0% % %
ONc 0 00 % 0

0 _0

c o .a 0

4 .S 0. .
> .2.02 ~
0U

V 00 00 0 0

2 cc

.2a _ 0

on 0.

1~~ 0

00 0.

CA-
U ~ ccvi

>0 .250
8- @0 00 i0
0%~~ =0o %
12 0% 0 % -j0
- - -

(Au
m.

8 2 2 ~V 0

0 00

040 0

.0 V

L. bo ca

0 I0

ON a,- <71a

4.o

i
> >0

u~

im1
i. u u u U 81

(A U50
 Qut oo 00 00 Nt- r- en t 00 t

a 4

0 0l c

in wi ( .20Cn

>.0,

- a.

be*a 0

00
2 On
NO2

00) 0to>

a r0a. 0> .

40

.~ 00 00 0

00

0 0 4

w 00

U-.

C-- 56 0
0ON 0 00
oo
N-
r-
%0
'IT
n
0%0
ON0%
Ch0 rN
r-
Q %0
0%0 % 0%

0 ~8 o
- 0
. 00 g -
a-ro _ ~ cr.

00

2 0
C ~ 0 0U)a
0
Ja V 0AC

> 0-

U)U

0..
*0 0be

aCO 'A V b

0r 04
to A

0 U;

*1.1 *057
rQ t- oo co 0%
r 00 00 m% 00 00 0% 0

U0 4)1 00

L)6 . 0M
I-U 0.~0

ol 0

00

o0 0 cc

z z u~u c. .
'A.
00

to IV. 00

2 02
to cc c

0.2 r. c 0 0*2

00 s
0 7B
.2 Gn c
04) ir & !
0.0
cc 0
s) v - I2 M

0 2o a 4 -4)

05 0. C0
.-
2 .

2: ~ > ý0 000

is - S 0l- . r. m)

S.--

ai

enU
E0
cc .C

-~ -958
00 c0 t- ~ 00" 00 % I-N N C'

U)3U

-6 0
o~ 0 4i0

00

G02 4)

C4

00

CuF

5a..59
 0q
D 0% C400S ell r- e
00 00 f- 0%@00 0
GO0% 00 ON 0
ON 0%l ON 0%0 0 0% ON0 ON @00

i0 0

14~§ . ~~
< ~. 0 o Efu

be j 00b

CA CLu

U2

60
ol 0
00 Wfl
00
%n
-1
00 vi ~ %0
cc 0
0cc
c-ea ell

clI-

! -
Z U C. -1

0 0

00 cc

L.
LV 00

a *0 -D
V 0-1.
00. 0) bo

A ) 4.

0*

.00. . .0. 4

0 00 00
.
4)

4) ."61
m%
en4
00
-
4.
'.0
4
00 0- I(
I
00
C14
Co
00

00 *0

4- 0 0 z

-ý - C 0

Er 2 OD.2~

0 4)0 z -
.~4 0! - O

W o ~ 0000

.04 L 0

0 V
U0 'Ab2

A - int 0 ON

0 0
> V
to0 0 0A >.
.2 0

e'~ 4)
.2 0 . .20

000

062
 - 0 C'l %n %nV~ 0% t-0 a- 0 M01 0%, 4
t,- 0h og M0 NO M1 0%0C~ 00
00 ~~~~ % 0% I% 0
!! 2.1%% %%_ 0

CCA

4?c

U 00
to o

4?r A4 0

or- vi

- 0

V) CA0cn -4?

63 2
 t-0 00ý -ý 00 00 00
ION N 0 00

0%C4 %0 0%%0 %0

0 4)

<k 0 00
.s4.0 .0W-

440

00

> 0 0;
C3 0

Ss 0 M

ID b
too
>0

0 0 9t

a- *0S th0IIw0
W .>

000

00

.000 0 ~ 64
00
00 0
00.

CA

00 U4

A .2

I..

00

40.0
.2

AA
Ro Z
M-0 ..

00 05
 Appendix B

Interview with Jessie Aites

Gaisford: Mr. Aites, my mother-in-law tells me you are a dowser. Can you tell me
about your skill?
Aites: I've been doing it for 50 years. I got the first nickel to take in. I don't charge
no how. So, now I've been doing it for, I'm 72, I've been doing it since I was
21 and I've never missed.
Gaisford: What do you dowse for?
Aites: Water, oil, you name it.
Gaisford: Have you been used a lot around here for oil?
Aites: I had to quit doing it for oil. I used to do it an awful lot for oil but I got
probably seven or eight thousand water wells. Never missed. I tell where its
at, how deep it is, how much its making, what color it is, everything.
Gaisford: Can I go back and get some biographical data? How old are you sir?
Aites: I'll be 72 in August.
Gaisford: And you said you were 21 when you started, so that should help me pretty
good.
Aites: 21 when I started.
Gaisford: How did you get involved with dowsing?
Aites: Well.. you just want me to go back to when I got started?
Gaisford: Yes, what got you interested and how you figured out you had a gift.
Aites: Well, the way I figured out. I didn't figure it out. I was told I had it. My
wife's grandfather, he was a water dowser for years. He lived to be 96 and he
was still doing it when he died. But one day, it was shortly after we got

66
 married. That was 50 years ago. Anyhow, he was at her brother-in-law's
finding water out in Ten Mile Bottom, about ten miles from here. And I went

out. And Wayne Millright went out and he was finding water for her brother-
in-law. He found the water and told them how deep it was. He said to me, he
said, he handed me the limb and said, "Here, see if it works for you." I took a
hold of that limb and I started across about where he was and I fell down.
Here it paralyzed my body. And so I've been doing it since that, to this day.
Gaisford: What's your primary technique? What branch do you use, willow or any kind

of branch?
Aites: Well, no not any kind. Willow works, but I like peach the best because I got
all them peach trees around here to trim. But I use wild cherry.
Gaisford: Do you tape the ends together or do you find a special branch?

Aites: I use Y's, and I don't have to. The only reason you use a Y is so the other
person can see it. I can use one inch of a peach branch and do the same thing I

do with a limb. You probably know what a limb is.

Gaisford: Yes sir.
Aites: Okay, I use that. That's just for you to see. I'll take one inch of it and hold it
between my two hands and I can tell you exactly where that water is by
holding that one limb. You don't see anything.

Gaisford: You can feel it rotating.

Aites: Its in my body.
Gaisford: What do you think does it?
Aites: Its a gift from God. No kidding around. That's what does it.
Gaisford: Have you ever tried to talk to anyone else who dowses and understand what

they have that you might have?

67
Aites: Oh yes. Well, they say the same thing. Its a gift from God. There's no other
way. I have two sons and I'd love to have one of them be able to do it. If I

could pass it on any way. But there's no way you can. I've been doing this for
years now and I only ran into actually about four who could do it. (Simon, his

seeing eye dog, approaches and Mr. Aites tells him to sit.) No, I only ran into
about four that could do it. But the way it works for them, it works for me.
There's no getting around it, it works. And I can find oil the same way. It
works the branch different but I can tell oil, lead, or water anytime.
Gaisford: Does the water have to be flowing for you to detect it?
Aites: Naturally, water is flowing all the time. If its not, its no good.
Gaisford: Well, the man I'm working with said that he could approach a swimming pool
and he could not tell the pool was next to him.
Aites: Oh no, I can't tell if its a swimming pool.

Gaisford: Cause its not moving water?

Aites: It won't pick it up. For a matter of fact, right here in the Allegheny River, I
found an underground vein under it. The top water is moving. I can't tell that

but I can find the water vein underneath it.

Gaisford: How do you determine depth?
Aites: Well, I have about three ways to do it. The way I go over it. As many
times...each time I walk over the vein of water the limb goes down. And I
keep doing that. Count one, two, three, four whatever it is. And when it quits
going down, I take three times whatever it is. Three times three. Three three's
are nine. It's nine feet deep. If its thirty, its three time it, you know. That's
the way I determine. And it works. I never miss I missed at one time by two
feet right down here at my neighbors. Oh, about ten or fifteen years ago. I
went down and told him his water was 167 feet. He come up about a wreek

68
 later, knocked on my door and says "you missed." And I said, "Maybe I can't
get them all. How much did I miss it?" He said, "Well, I never got water until

I was 169 feet." I missed it by two feet. (Chuckles) Now, whether the driller
measured it at the top of his pipe or down to where he found the water, but I
measure from where I stand, from my feet down. And he was actually put out

because I missed it by two feet. So now, I have a thing. I tell them. I go out
and find them water and I say its "thirty feet," thirty feet, one way or the other.
(Chuckles) So, that's the way...cause...you know when a driller sometimes
puts a pipe in the ground, it sticks out two feet. He measures that two feet
there. He's doing it for the money, by the foot. So I really don't know how
this man down there measured it.
Gaisford: Do you have any well drillers that use you as a consultant?

Aites: Oh yes. Well, I have the Harry Brothers. I don't know if you know them or
not. There down there out of Franklin. They have drilled quite a few that I
have found. As a matter of fact, they used to go to people and they would
want a well drilled and they would say, "Well, to save you some money, why

don't you go and get a hold of Aites. And then we'll know exactly how much
its going to cost before we start." There's different ones. Over in Clarion
there's the Tiger Brothers. They call on me every once in a while. Different

ones.
Gaisford: The reason I'm asking that question is because I'm doing some research and
perhaps the well drillers have documented what you told them in their drilling
logs. If they have kept something, I can get that and use that as backup for
what you have told me. Not that I doubt your word, it helps to have some

written...

69
Aites: Like I said, I don't charge a nickel for this. I never did and I don't think I ever
missed. I can show you papers here, newspaper clippings with drillers that go
and drill it and they already know before they drill how far its going to be
because I told them. That's right in the paper. Then they go ahead and drill its
in.. .I don't know if you ever heard of Don Riggs on television or not.

Gaisford: No, its not familiar.

Aites: He made a movie. Its a 22-minute movie. Made it for KDKA Pittsburgh.
They went out here a couple of miles, about five miles from here, I guess.
And we determined how deep it was going to be before the news. KDKA,

came out to take the movie. Went through all that. And it was on the air, it
had to be, like you said...like...go and question someone else about it. And it
had to be that way for television because, boy, you'd be surprised the people
who wrote into there and questioned them. So a... this one a.. .boy this one

fella I used to go so many times. I think he's dead now. He's a water driller. I
can't think of his name now its so many years ago. Anyhow I had a good
many drillers come get me and didn't even want the water. But they would
come and get me and have me go and measure it. And they would drill it to

make sure it was going to be there before they would charge people for what I
was telling them. (pause) Continental Can, I found water for them. I used to
work for Continental Can. Our water bill down there was running us three
hundred, three hundred and eighty-two dollars a month. A water bill for in the
plant, water and sewer. Pat Cohen, the manager, he come out and said, "Jess
if we have to pay this water bill every month. Its $380 now a month and its

going up. We can't hardly afford that. Can you find us a vein of water?" I
told him "I don't know why not." He told me to bring a limb down to the

shop. I took a limb down about a week later and I went outside the building

70
 and I found them a vein of water, 57 feet deep. I told them, I said "Right here

is the spot. Its 57 and a half' 57 feet I told them. And okay, they called a
driller in and they come down and he said, "I want this..." Pat Cohen said, "I
want this well drilled right here." "Why do you want it drilled here?" He
says, "Jess Aites come down here and found the water vein. Its 57 feet deep."

And he says, "No I won't drill it for you. I won't have anything to do with it."
Pat says, "Why? You drill water?" And he says, "Yeah, but if Aites had
anything to do with it, I won't come down." So Pat called me in and talked to
me. I said, "Let's let him decide." I said, "Okay, I'll bring my limb down." I

said, "I'd find you another vein of water." So I went down in the parking lot
and I walked around, found a vein of water there. I told Pat, "Here's one here I
measured. Its 57 feet deep." It's probably 100 feet from the other one. So he

called then, Millat, the well driller down. And he said, "Okay, lets go out in
the parking lot." They went out in the parking lot and they walked around a
little while. Pat Cohen took him to this spot I had found. It was marked on
the ground. And Pat's moving his foot around and Millat said, "Okay let's
work around here some place." Pat Cohen put his foot right on the spot we
had marked. He said, "Let's drill here." Millat says, "Okay, that sounds
good." (chuckles) So he come down and drilled a nine inch hole. He went to
57 and 1/2 feet deep. Got all kinds of water. I guaranteed him 40 gallon a
minute. He drilled it 57 and 1/2 feet and their flow was over 50 gallon.

(chuckles) There using that today and I found that in a...oh boy...'73. I think
it was in '73. And they are still pumping that and using all kinds of water.
The bill went from $180 a month down to $40 a month. (chuckles)
Gaisford: That's good.
Aites: Yeah, I think they probably have that right in their record down there.

71
Gaisford: How long have you been blind?
Aites: I can't see daylight. Went blind July the 5th, 1973.
Gaisford: So, has your ability changed any since you were blinded?
Aites: No, no, not a bit. For a matter of fact, its a gift of God. I don't know any
more about it now then when I could see. No, I've been blindfolded by a good
many people. I'd walk around there and find them a spot. They'd say, "Can we
blindfold you?" They'd put something over my head and I'd walk around and

find the same spot. You know, you can't change the water.
Gaisford: You talk about veins of water. In studying geology and hydrology, the study

of the earth and water in the ground, we typically say aquifers run in several
layers and not in small, thin veins of water. They are very large pools or
continuous areas that have water in them. How would you relate a vein to that
kind of analysis?
Aites: Well, most of them, around here, may be the size of my finger, the vein of
water. That's a lot of water. The biggest one I ever found was up in Henry's
Bend and I'd say it was maybe a foot wide. I don't know how big around that
was, I couldn't tell. I know it was, let's say like a creek on top of the ground

around a foot wide. That's the widest one I ever found, but most of them are

the size of a pencil or my thumb. As a matter of fact, did you ever see how

they did it when they drill.

Gaisford: No. I know how they drill, but I haven't watched them.
Aites: No. I mean to see the flow of water actually flowing down there.

Gaisford: No.
Aites: If you are around where they are drilling a well, I don't care if its 90', 100', 50'.
Sun shines and you get a mirror, put it in your hand. You hold that mirror up
to the sun and you can see that vein of water running down there. That's

72
 (chuckles) that's hard to believe but it works. Yeah, I had one that one fellow.
I can't even think of his name. He told me, and I could see at the time, and he
got a mirror. He held it up there in the sun and you could see the water. The

one I happened to see was about the size of my finger. Come right out into
that hole and going right down the hole they drilled.
Gaisford: What is the deepest you have found water at?
Aites: Oh boy, I'd say over 200 feet. I haven't kept track. I know it was over 200'. I
measured an oil well up here one time, 1100'. (Chuckles) I didn't see it, but
the driller who owned it, he said, "You're exactly right. I had that on the

books. That's 1300'." (pause) Found one. The guy told me to come out. He
wanted water. Okay, I went out and found him water and he lived about
100...the house was about 100' off his driveway. I found him water and I told
him, "Oh boy, your water is not quite six feet deep." "Oh no," he said, " I
can't have water that...I've lived here all these years and carried water." I said,

"I know if you dig down there about three feet you're going to get water."
"No, no, I can't." And there was a bacihoe up and I says, "Why don't you get
"thatfella with that backhoe, come down here and find out." So, after I left, he
went and seen that fella and they went down the same evening. And that

backhoe went down there. He got down just about three feet and there was
moisture. One more scoop and there was water, just flowing. Another one

right up here, up the road here, people hauled water for years. Fifteen years
they had lived there and they hauled water. I went up, he called me up and
told me he can't get water. I said, "Let's look." And I found it. I took him

right down in his basement. I found water. I said, "You got water right under

your basement floor." I think it was nine feet. "Oh, no," he said. "Jess, I've
been hauling water all the time I've lived here. I know there's not water here

73
 'cause so-and-so didn't have water." "Okay," I said, "You asked me to come

up, I told you." And they went down in the basement and they got that water.
Nine feet. All them years they didn't have water, yet it was there. That's the
reason I say its a gift from God. I can't put water where it isn't. I can tell them
where its at if there is any. And when I go to places like that and they say,

"Fifteen years I lived here and so-and-so couldn't get water." Its there. I can
tell you where I moved, one fella was digging. Glenn Perry was digging a
well in his front yard. He was down 35 feet. Gang of men digging it. They
had me come out to find water and I found it in his back yard at 21'. "We can't

have water, we're 3o some feet." "Well, you asked me to come out." They
moved from the front of the house out back of the house. They dug it out
there. Got water exactly at 21 feet. So, I know it works. Like I said, I never
charge a nickel for it.

Gaisford: Sir, I thank you for your time. If there is anything else I can think of to ask
you is it alright to give you a call?

Aites: Oh, yes. You can call me anytime.

74
 Appendix C

Experiment Materials List

LtmQmanmy Meaur
Adapter, 3/4" F x 1/2 S 2 Ea
Adapters, 3/4 M x 1/2 S 10 Ea
Brackets, Equipment Support 1 Pair
Cement, PVC, 4 oz 2 Ea
Cleaner, PVC, 8 oz 1 Ea
Couplings, 1/2", S-S 17 Ea
Elbow, 90, 1/2", S-S 11 Ea
Elbow, 45, 1/2", S-S 16 Ea
Pipe, Steel, 3/4" x 3" 2 Ea
Plywood, 3/4" x 24" x 30" 1 Ea
Power Cord, 12-3 10 Ft
Pump, Bell and Gossett Model SLC-30 1 Ea
Tape, Teflon 1 Roll
Tee, 1/2", S-S-S 8 Ea
Threaded Rod, 3/8" x 4" 2 Ea
Tubing, PVC, 1/2" 540 Feet
Valve, Ball, 1/2", S-S 5 Ea
Valve, Check, 3/4", F-F 5 Ea
Valve, Gate, 1/2", Brass 1 Ea
S - Smooth F - Female Thread M - Male Thread

75
 Bibliography

1. Barrett, Sir William and Theodore Besterman. The Divining Rod. Toronto, Canada:
Coles Publishing Company, Ltd., 1979.

2. Bird, Christopher. The DiviningHand: The 500-Year-Old Mystery of Dowsing.

New York, NY: E.P. Dutton, 1979.

3. Carey, Stephen S. A Beginner'sGuide to the Scientific Method. Belmont, CA:

Wadsworth, Inc., 1994.

4. Costerton, J. William. "Isolation of Pollutants Using a Biobarrier Technology," 20th

Annual RREL Research Symposium Abstract Proceedings. EPA/600/R-94/0 11, pp 95-
101. Contract Number 68-C2-0148, Work Assigment 2-7. Ft. Washington, PA, Marcil
1994.

5. Devore, Jay L. ProbabilityandStatisticsfor Engineeringand the Sciences.

Belmont, CA. Wadsworth, Inc., 1991.

6. Eastwood, Norman B. "Some Observations on Dowsing and the Human Magnetic

Sense," The Lancet: pp 676-677 (19 Sep 87).

7. Freeze, R. Allan and John A. Cherry. Groundwater. Englewood Cliffs, NJ:

Prentice-Hall, Inc., 1979.

8. Graves, Tom. The Dowser's Workbook: Understandingand Using the Power of

Dowsing. New York, NY: Sterling Publishing Co., Ltd., 1989.

9. Irons, F. E. "Concerning the Non-Linear Behavior of the Force Spherical Pendulum

Including the Dowsing Pendulum," EuropeanJournalof Physics, Vol 11: pp 107-115
(1990).

10. Martin, Michael. "A New Controlled Dowsing Experiment," The Skeptical Inquirer,
Vol 8, No 2: pp 138-140 (Winter 1983-84).

11. McCarthy, Paul. "Dowsing with Magnetite," Omni, Vol 10, No 5: p 88 (Feb 1988).

12. Randi, James. "The Great $110,000 Dowsing Challenge," The Skeptical Inquirer,
Vol 8, No. 4: pp 329-333 (Summer 1984).

13. Randi, James. "A Controlled Test of Dowsing Abilities," The Skeptical Inquirer,
Vol 4. No. 1: pp 16-20 (Fall 1979).

76
14. U. S. Geological Survey. Water Dowsing. Report 621. Washington: Government
Printing Office, 1977.

15. Vogt, Evon Z. and Ray Hyman. Water Witching U.S.A. Chicago, IL: The
University of Chicago Press, 1979.

16. Wald, Abraham. SequentialAnalysis. New York, NY: John Wiley and Sons, 1947.

77
 Vita

David Ira Gaisford was born in Pasadena, Texas in 1959. His father was an Air
Force pilot allowing David to live in several communities during his childhood. He
attended college at the University of Texas at Austin and graduated in 1983 with a

bachelor of science degree in Architectural Engineering.

David entered active duty as an Air Force Civil Engineering officer in October
1983. He served at Holloman AFB, New Mexico, Howard AFB, Panama, Myrtle Beach

AFB, South Carolina, and Wright-Patterson AFB, Ohio.

During Operation Desert Shield/Storm, David worked as the Chief of Engineering

for the 354th Tactical Fighter Wing (Deployed) at King Fahd International Airport,
Kingdom of Saudi Arabia. His efforts were paramount to creating an operational air base

to house over 5000 troops and 100 aircraft. These efforts were the basis for him receiving
the Bronze Star during that war.
In 1992, David was sent to the Air Force Institute of Technology (AFIT) to earn
his master's degree. Air Force leadership, however, decided that it didn't need non-rated,
experienced, reserve officers and choose to release him from active duty.
In January 1993, David started work as an Environmental Engineer for Robins
AFB, Georgia. He immediately made plans to return to AFIT to finish the master's
degree program he had started. He was selected to continue his education with the 94S

Class and returned to Wright-Patterson AFB in January 1994.

David is a registered professional engineer in the state of Texas.

Permanent Address: David Gaisford

C/O W. H. Gaisford
R. D. # 1 Box 318
Cooperstown, PA 16317

78
 September 1994 Master's Thesis

EXPERIMENT IN WATER DOWSING

David I. Gaisford, GS-12

Air Force Institute of Technology, WPAFB OH. 45433-6583 AFIT/GEE/ENS/94S-0I

Approved for public release; distribution unlimited

Dowsing is a folklore process used to locate an unknown, such as the best location for a water
well, by the use of a hand-held device. The process is commonly know as water witching, divining,
dowsing or radiesthesia. The practice continues despite the lack of a proven scientific basis.
This research develops an experiment to test the claims of a dowser. Specific procedures are
established and statistical theory is applied to determine if one man can identify which of five water
lines has flowing water in it better than a chance operator could achieve. The statistical analysis uses
Abraham Wald's sequential analysis procedures for establishing when to accept a hypothesis in a
binomial situation. The dowser's performance proved to be better than chance. Further research is
recommended.

Dowsing. Water Witching, Sequential Analysis, Binomial 88

Experiment

Unclassified Unclassified Unclassified UL