Mysterious ways: The riddle of the homing ability in dogs and other vertebrates

Full text

Journal of the Society for Psychical Research [Vol. 79.3, No. 920
140
MYSTERIOUS WAYS: THE RIDDLE OF THE HOMING ABILITY
IN DOGS AND OTHER VERTEBRATES
by MICHAEL NAHM
ABSTRACT
The homing ability of animals has fascinated many laymen and scientists for some
time. Despite considerable efforts by researchers to elucidate its underpinnings,
however, it is still not known exactly how a bird, for example, can determine its
longitudinal position on the globe and find its home from an unknown location to
which it has been displaced. The same seems valid for terrestrial animals such as
dogs. To bring this problem to renewed attention and to stimulate further research
into this topic, this paper introduces two largely unknown sources dealing with the
homing ability in dogs. Both contain remarkable instances of dogs that repeatedly
returned to their homes or keepers from unknown locations. The first source
concerns the work of Edwin H. Richardson with messenger dogs in World War I; the
second source concerns the systematic homing experiments that Bernhard Müller
performed between 1953 and 1962 with 75 dogs. Both authors maintained that
the homing success of the dogs could not be explained in terms of the use of their
usual senses alone. I go on to review currently available explanatory hypotheses
for animal navigation with a focus on bird and mammal homing, and also touch
upon orientation abilities in humans. This review shows that there is no consensus
among leading experts with regard to the most suited model, and many admit that
a conclusive explanation for homing is still lacking. Hence, an additional governing
factor such as ESP might well come into play. Drawing on the training methods for
dogs applied by Richardson, I suggest testing this hypothesis by training dogs to
find their keepers at an unknown location.
INTRODUCTION
The ability of many animal species to find their way home across large
distances, as well as from unknown locations to which they had been displaced,
has intrigued many laymen and scientists for some time. And, although para-
psychology classically centres on experiences of people (Irwin & Watt, 2007,
p.2), this subject has also been discussed by a few members of the parapsychol-
ogical community, and, in fact, seems to have stimulated academic psi research
in animals. In the middle of the 20th century, Joseph B. Rhine (1951) and
Gaither Pratt (1953) were among the early authors who explicitly highlighted
the potential connection between the homing ability of animals and extrasens-
ory perception (ESP), and who pointed to the demand for clarifying studies.
In the meantime, scientists in the mainstream setting have created an
enormous body of knowledge on the biophysical cues involved in animal orient-
ation. Among vertebrates, the most extensive research has been performed with
birds (Akesson, 2003; Bertold, 2012; Newton, 2008; Thorup & Holland, 2009),
but terrestrial animals (Bovet, 1992; Linnell, Aanes, Swenson, Odden & Smith,
1997; Rogers, 1988) and aquatic species (Lohmann, Lohmann & Endres, 2008,
Lohmann, Lohmann, Brothers & Putman, 2013) have also been monitored and
subjected to numerous experiments. As a consequence of these results, it has
seemed likely that psi would most probably not be involved in governing typical
homing behaviour (e.g. Morris, 1977). But interestingly, although several

July 2015] Mysterious Ways: The Riddle of Homing Ability
141
hypotheses have been put forward to explain especially the birds’ homing
success after displacements, there is still no certainty and consensus among
scientists regarding the factors that guide these animals on their journeys (e.g.
Bertold, 2012; Gagliardo, 2013; Newton, 2008; Thorup & Holland, 2009; Wall-
raff, 2005; Wiltschko, 1996). The homing ability of terrestrial animals seems
equally unexplained and perhaps even more puzzling, given their comparably
limited field of vision and relatively slow mode of motion.
Nevertheless, in contrast to researchers from the mainstream setting, para-
psychologists typically have deemed the accounts about animals who found
their owners at an unknown location to be of greatest significance in the
context of the potential involvement of psi in animal navigation, because these
animals would not be able to rely on biophysical cues to find their owners
across considerable distances, some of which were more than several hundred
kilometres (Gaddis & Gaddis, 1970; Rhine, 1951; Rhine & Feather, 1962; Shel-
drake, 2011). Should this reported ability of many animal species, including
birds, dogs and cats, be established experimentally, this might contribute
significantly to elucidating the as yet unknown factors that govern ‘normal’
homing success, especially after displacement. However, with the exception
of work conducted by Rupert Sheldrake (2002, 2011), practical experiments
addressing this specific aspect of animal homing have not been conducted.1
Taking into account the fact that animals appear able in principle to display
psi abilities (for a review, see Dutton & Williams, 2009), it remains possible
that ESP could play a role in governing the mysterious home-finding process of
animals.
In the light of this perspective, the aims of this article are (1) to bring this
possibility to renewed attention, (2) to stimulate further research into this
topic, and (3) to introduce two largely unknown sources concerned with the
homing ability in dogs. The first source reports the work of Edwin H. Richard-
son with messenger dogs in World War I, the second the systematic homing
experiments that Bernhard Müller performed between 1953 and 1962 with 75
different dogs. These sources have not been cited in the literature on homing
and animal psi up to today (e.g. Rhine, 1951; Rhine & Feather, 1962; Gaddis
& Gaddis, 1970; Schwertner, 1984; Nahm, 2007; Sheldrake, 2002, 2011), and
seem absent also from the mainstream literature with the exception of Bovet
(1992), who mentioned Müller’s study in passing. Both Richardson and Müller
support the view of others who have maintained that the dog’s homing success
1 Sheldrake (2002, 2011) has also drawn attention to the phenomenon that some animals, in particular
dogs, seem to be able to know in advance when their owners are coming home. His practical experiments
strongly supported this appraisal. It seems little known that Schwertner (1984) had already conducted
similar experiments. He stated that it would be fairly well known among dog and cat owners that some of
their pets know when their owners return home. Schwertner briefly described experiments he performed
with two dogs that lived in the same house, and with their keepers or unknown persons, respectively. A
keeper or an unknown person would return to the home of the dogs in a car, either in their own car or in
a car unknown to the dogs, while the dogs stayed at home, not being able to see the arriving car. One of
the keepers stayed at home with the dogs. From the possible combinations, the following situation called
most explicitly for an extra-sensory explanation: a beloved keeper was driven home in a car and by a
driver unknown to the dogs. The dogs still reacted in a manner that strongly implied that they anticip-
ated that a loved one would return home. The reactions of one dog were more reliable than the reactions
of the other dog (Schwertner, 1984, pp.59–63).

Journal of the Society for Psychical Research [Vol. 79.3, No. 920
142
cannot be fully explained in terms of the use of the ‘normal’ senses alone.
In the first two parts of the present article, I present summaries of these
two sources. In the subsequent parts, I present brief reviews of currently
available explanatory hypotheses for bird and mammal homing, and explain
why — in contrast to the prevailing ‘common’ opinion typically spread via public
media — they are regarded as unsatisfactory even by experts in this field.
The last part before the Concluding Remarks section also includes a short
discussion of the homing ability in humans.
THE WORK OF EDWIN H. RICHARDSON WITH DOGS IN WORLD WAR I
Colonel Edwin Hautonville Richardson spent many years training dogs of
various kinds (Richardson, 1910; Richardson & Richardson, n.d. [presumably
1951]), and established the British War Dog School in Shoeburyness in 1917
(Richardson, n.d. [presumably 1920], 1921). In this school, hundreds of dogs
were trained to act as messenger dogs and sentries close to the front line in
World War I. Here, I focus on the ability of the messenger dogs to return to
their keepers through unknown terrain. Richardson’s interest in the homing
ability of dogs was first raised when his first dog found its way home from the
centre of Brighton, where it had got lost in crowded streets. It was the dog’s
first visit to Brighton, and he was taken there in a carriage on a winding and
“not at all direct” route. Richardson’s house lay “several miles” behind Brighton,
but the dog was seen heading towards the house in the evening, apparently
travelling “over land he had never seen before, and in a totally different way of
travel from that on which he had set out in the carriage that morning” (Richard-
son & Richardson, n.d., p.24). In the following, I cite excerpts from an article
that Richardson (1921) had published in a little-known journal named Psyche:
A Quarterly Review of Psychology. The paragraphs below summarize Richard-
son’s main thoughts and findings regarding the subject in question (pp.52–53):–
The subject of homing instinct is one which leads into many interesting fields
of investigation, and is certainly of peculiar attraction to those whose perception is
awakened to the fact of causation in mind . . . My own actual experiences connected
with this form of intelligence have been almost entirely with dogs, in whom this
instinct is very highly developed. The intensity, however, differs in individual dogs,
and also in various breeds. During the war several hundred dogs were trained as
messengers and dispatch carriers. The training for this work was based on the homing
instinct, but I found it necessary at the training school to study the psychology of each
dog as the bent was much more highly developed in some dogs than in others. Dogs
of wise and affectionate natures were the only ones of any use in the strenuous work
they had to perform in the field, and the great lever by which the homing instinct was
initiated, was that of devotion to the man who was deputed to be the dog’s keeper. A
dog could be taught the duties of messenger service mechanically, but when it came to
maintaining this effort in the field only some very powerful emotion could enable it to
overcome the obstacles in its way. In briefly describing these obstacles, I will explain
that the messenger dogs for the British Army were concentrated in units behind the
line and were dispatched in groups to those parts on the line where particularly
strenuous fighting was expected. They went up in the charge of their keepers, each
man having three dogs. Having arrived at Brigade headquarters the keepers remained
there and the dogs were taken from them by troops occupying the front line . . . They
were frequently taken up to their posts at night, over ground utterly unknown to them
previously, and were released some hours afterwards with their messages. Sometimes

July 2015] Mysterious Ways: The Riddle of Homing Ability
143
they returned by the way they had been taken up, but more often chose a more direct
route straight across the country . . . It will be remembered that this would lead them
over trackless ground, or along trenches and roads crowded with every sort of traffic,
through villages full of troops and every sort of obstruction and temptation. That these
dogs accomplished this work is one of the wonders of the war. How they did it cannot
be fully explained, for the reason that we do not fully understand the influences
which control the animals when under an overpowering desire to return to the place
from whence they came. Suffice it to say that it was the determination to return to
a beloved master, as represented by his keeper, and that as a result of this emotion,
portents and signs indistinguishable to man were waymarks on the journey.
In a preceding book, Richardson described the methods applied in the
training and the use of messenger dogs in more detail. He also dwelt more
extensively on the ‘homing instinct’ of these dogs (Richardson, n.d., pp.163–
182). Again, he maintained that their homing ability could not be explained by
means of the dogs’ usual senses, such as sight, hearing or smell. The sense of
sight could be ruled out because the dogs found their way back to their keepers
in full darkness or thick mist through trackless country unknown to them. In
fact, it seemed as if
trained dogs, and even half-trained ones, work better under these conditions than in
daylight. And this was also borne out in the field. The keepers have related to me that
on certain nights, when the conditions were so bad, the night so dark and thick, the
ground so water-logged and shell-marked, and on certain occasions quite new to the
dogs, that they were fearful that these would prove too much even for their faithful
followers. But the curious point was brought out, that the dogs seem to work much
better than usual, at such times. As one man said: “ It seemed as though ‘Jock’ divined
my fears, and put out an extra effort to show they were needless” [pp. 172f]
According to Richardson, hearing and smell could likewise be ruled out
because the dogs often chose different routes for their return, across unknown
country without trail (p.163). Richardson also related cases from “civilian
canine life” in which dogs were said to have got home across considerable
distances (p.164). Hence, he concluded “that the causative guiding source of
the homing faculty has its origin in the realm of metaphysics, rather than in
external phenomena, and if any explanation is sought, it is here that investig-
ation should be made” (p.174).
When Richardson began the training of dogs, he first used a fixed base,
then a movable base, and he carried the dogs to the place of their release on a
roundabout route. In general, he recommended leaving the dog one or two days
with the keeper at the Brigade headquarters before sending it to the front.
However, dogs were also able to find their keeper when they were removed
from him immediately after their arrival at the headquarters (pp.105–108).
Typical distances that the dogs had to cross were from 3–6 km, but Richardson
also lists examples of the dogs returning to their keepers from distances of 10,
14, 17, 17 and 22 km. With regard to the training and the development of the
homing ability, he reasoned
that observations seem to point to the fact that the intense desire to reach a given
place impels the dog forward; that as it yields to this impulse, that a certain guiding
sense, which is in itself quite independent of any assistance from external phenomena,
comes to its aid, and the sense of direction is, in this very sense — that the dog desires
to be there, and follows this desire, rather than troubling about the aspect of the
surroundings in getting there. The more it becomes accustomed to throw all its effort

Journal of the Society for Psychical Research [Vol. 79.3, No. 920
144
into this intuitive prompting, the more it discards any temporary assistance it may be
tempted to use, in the first place, such as noting turns in the road, and other external
aids, and also the more it improves in its way-finding duties. The deduction, in fact,
seems to be plain, that the desire itself brings its own lesson, and a world of intelligence
is opened up to the dog, and to all animals, under stress of this governing force, of
which we human beings are quite unconscious, because we have not yet exercised this
particular mental effort along the same lines as the animals. It will therefore be seen,
that those promptings which have their origin in what we call instinct, are due to
an intelligence quite apart from, and infinitely above, any guidance from the senses.
[p. 179]
Comments on Richardson’s Work
With regard to this claim, it is of interest that, as mentioned, the dogs were
even able to find their keepers at locations where the keepers simply stayed
behind while the dogs were moved onward. In these cases, the dogs had no
opportunity to become acquainted with the place the keeper would remain in.
Such observations highlight the possibility that the homing ability of dogs
might indeed not be restricted to a given geographical location, but that the
keeper himself might be the target of location — irrespective of his whereabouts.
In fact, the ability to find a beloved keeper at an entirely unknown location
after displacement of the dog to another location would be much more difficult
to explain than a dog’s returning to a known location from an unknown place.
The supposition that some animals are in principle capable of finding their
attachment figure at locations entirely unknown to them is supported by
anecdotal reports (Gaddis & Gaddis, 1970; Rhine & Feather, 1962; Sheldrake,
2011). This, however, was not the case with Richardson’s dogs. They had
always spent some time with their keepers at the location where the keepers
remained, even if only a very short time. Thus, suggestive as the reports of
Richardson are, it is unfortunate that he did not conduct the experimentum
crucis that, in case of success, would have removed doubts in this respect: after
sufficient training in the usual way, he could have reversed the task, leaving
the dogs at their base and hiding their keepers at continually increasing
distances. If some at least of the dogs had successfully located their keepers,
this would have had considerable impact on current explanatory models for
goal orientation in animals (see below).2 Moreover, remarkable as Richardson’s
findings are, his descriptions are not as detailed as one would wish them to be
from a critical perspective. It is apparent that his observations regarding the
peculiar homing ability of his war dogs were only a sideline of his comprehen-
sive work with them. Systematic experiments with a scientific background are
described in the next section.
THE WORK OF BERNHARD MÜLLER ON THE HOMING ABILITY OF DOGS
Before I review the work of Swiss researcher Bernhard Müller (1965), I
will briefly sketch a conceptually similar approach that was applied earlier by
2 In fact, the German army had trained messenger dogs not only to return to their base from the front
line, but also to go back again from the base to a second keeper at the front line (Richardson, n d., pp. 74f,
256f). This two-way system, however, relied on familiarity with the terrain and the routes to choose,
and was not exclusively founded on the emotional bond with the dogs’ keepers. Richardson abstained
from training for this two-way approach because it was more difficult and time-consuming to train, and
fraught with several difficulties, especially in times of war.

July 2015] Mysterious Ways: The Riddle of Homing Ability
145
behavioural scientist Bastian Schmid. Apparently unaware of the abilities of
messenger dogs, it seems that he performed the first systematic experiments
to explicitly assess the homing ability of dogs (Schmid, 1932, 1936; for a
summary of Schmid’s work, see Sheldrake, 2011). In these experiments, the
behaviour of three dogs that were displaced to an unknown location 4–5 km
distant as the crow flies was continuously observed by Schmid and several
helpers. Two dogs found home successfully on all trials performed with them
(three and two trials, respectively). One dog was tested in a rural area, the
other in the centre of a major city, namely Munich. The latter’s success is
especially remarkable because being displaced to an unknown location in
4.5 km air-line distance in the centre of a large city with street canyons on
plain ground effectively rules out orientation by sight, and it would likewise
be very difficult to detect salient familiar odours. The third dog failed on three
occasions to find home in a rural area, even when he had full view of a location
very familiar to him. Curiously, the successful dogs didn’t seem to use their
noses on their way home; they didn’t sniff for cues close to the ground, or in the
air. After an initial phase of orientation, they simply trotted in a homeward
direction with raised heads. Tentatively, Schmid attributed their success to
a hitherto unknown sense, an “absolute sense of orientation” (Schmid, 1932,
p.156).3
Whilst Schmid’s exploratory study is regularly cited (e.g. Rhine, 1951;
Sheldrake, 2011), the largely unknown experiments performed by Müller
(1965) represent the most extensive and systematic study on homing dogs ever
performed. Between 1953 and 1962, Müller experimented in Switzerland and
Nepal with 75 dogs. The ideal test series for a single dog consisted of four
runs from the same release site at a distance of 2.5 to 3.0 km from the home
territory, four runs from a different release site at 5 to 7 km distance and
shifted clockwise through an approximate angle of 120°, and four runs from
a distance of 10 to 89 km, shifted another 120°. Hence, a complete test series
consisted of 12 runs that started at three different locations.
Of these 75 dogs, 19 completed the entire series successfully, seven started
successfully but didn’t complete the series for different reasons, and 49 didn’t
find home in the first runs and were not studied further. In total, Müller
recorded 249 successful homing tests. The dogs were carried to the release sites
in closed baskets by a variety of forms of transport, usually via complicated
detours. Experiments were carried out in all weather conditions including
rain, snowfall and fog, by day and night. To document as many details of the
dogs’ behaviour as possible, Müller used stationary and mobile observers along
the potential tracks they might take, and he tried to follow the dogs at a safe
distance. In no case did the dogs follow the route they had been taken to the
release site.
Prior to the tests, Müller determined the dogs’ social status by a number of
selected behaviour characteristics, and classified them into three categories:
3 Another early series of systematic displacement experiments was performed by Francis H. Herrick
(1922). He performed eight tests with a cat that didn’t return after the last trial. Herrick stated that its
homing success could not be attributed to vision, hearing or smell. But because his cat seemed to have
difficulty finding home after it had once been displaced during anaesthesia, he tentatively attributed its
homing ability to the kinesthetic sense.

Journal of the Society for Psychical Research [Vol. 79.3, No. 920
146
dominant alpha-animals, submissive omega-animals, and dogs that belonged
to an intermediate group. He assumed that with increasingly higher rank, the
‘value’ or the importance of the home territory for the dogs would increase,
and would thus result in a greater homing impulse and success. Müller’s
results strongly confirmed this assumption. All 22 dogs classified as alpha-
dogs belonged to the 26 dogs that completed the entire test series successfully
or that started successfully; the remaining four dogs belonged to the upper
midfield. Of the 27 dogs classified as omega-dog, none returned home. Usually,
these dogs would look for human settlements, no matter in which direction,
and try to associate with humans, or in the case of Nepalese street dogs, other
dogs, and stay there. The homing impulse seemed to be absent.
The alpha-dogs, on the other hand, would spend a typical adaption phase in
the surroundings of the opened basket (Müller divided this phase into three
stages), and then leave the release site, determined to go home. On their way,
they would avoid any contact with people. A very characteristic bearing of a
dog on its way home after the initial orientation phase was to hold its head
high and in a peculiarly stiff manner when trotting, its eyes appearing some-
what “veiled ”. Often, and seemingly untypical of the dogs’ normal behaviour,
they would stumble when the soil was uneven, or even collide with low wire
fences. Müller figuratively described this behaviour as if the dogs were “ridden
by an alien force”. In general, orientation by vision seemed to play a negligible
role in their journeys. During the repetition tests on the same release sites, the
dogs displayed all three stages of the agitated initial adaption phase again and
typically commenced their homing phase from a different spot of the release
area. Similarly, even when they reached the same bottleneck such as a ridge
or a pass on each repetition test, they would again take different routes
thereafter. Should they have recognized characteristic topographical features
of these places, they didn’t seem to utilize this knowledge. Rather, they used
new and typically shorter routes on each repetition, which sometimes led
them over different mountains and through different valleys. Hence, the dogs
seemed to have learned how to choose the correct direction in a general sense,
and were able to adjust their routes accordingly.
However, despite the many details that Müller’s experiments with homing
dogs brought forth, he remained at a loss to explain its underpinnings. Müller
postulated a “polarized sense of direction” that would extrapolate the intra-
territorial orientation of dogs also to remote places, but he could not explain
its modus operandi. Moreover, he discarded as insufficient the existing models
that were promoted to explain the homing success especially of birds, and
postulated that if the ability found in bird species to use “absolute reference
systems” analogous to the sun compass were to be detected in dogs, the prob-
lem would again be solved only in part. It would still remain unclear how the
animal could determine its geographical position and the direction to choose at
the release site.
Apparently, this problem is still unsolved. Because one may think that the
dogs’ ability to find home might be related to the same ability known from
birds (especially pigeons), and that both feats might even rely on similar cues
for orientation, I will introduce unsolved enigmas of bird migration research
in the following section of this paper. The current hypotheses about homing

July 2015] Mysterious Ways: The Riddle of Homing Ability
147
birds will be followed by those for mammal homing, and I will also touch upon
aspects of spontaneous spatial orientation in humans.
EXPLANATORY HYPOTHESES FOR BIRD HOMING
In the academic literature on migration, scientists distinguish between
normal orientation and navigation: orientation (or compass orientation) refers
to the ability of animals to align their movement direction to external stimuli
such as the sun or the earth’s magnetic field; navigation (or goal orientation)
refers to the ability of animals to find their way to a defined location from
another, often unknown, location to which they had been displaced, and it
comprises the orientation towards the defined location at the release site
(Berthold, 2012; Thorup & Holland, 2009). The homing success of dogs is an
excellent example of animal navigation. To understand what is currently
known about animal navigation, it is recommended that we look first at
research performed with birds, because the largest series of experiments have
been performed with these animals. As a result, it is now known that perhaps
all birds — migrating and also non-migrating bird species — possess the ability
of navigation. Hence, it should suffice here to cite Peter Berthold, a leading
authority in the field of bird migration research, who summarized the state of
knowledge in the scientific German standard handbook on bird migration as
follows:–
According to the results of thousands of displacement experiments . . . we must
assume theoretically, or as a working hypothesis, the following: after a displacement
to an unknown location, every bird can (1) determine its geographical position at the
release site in relation to the starting point, and (2) return to this starting point —
provided that the bird is healthy and is tested within the scope of its bodily spectrum
of performance . . . Frequently, the spontaneous and rapidly performed initial orienta-
tion at the release site already points towards the target place; and direct observations,
retrieved rings, calculations of flying times, and telemetry studies prove that frequent
essentially direct return flights are made by the shortest way.
[Berthold, 2012, p. 182, my translation]
Many different cues that seem to govern orientation are discussed by
ornithologists, such as a sun and star compass, a geomagnetic compass, wind
directions that are integrated in terms of angles to compass references, atmos-
pheric chemosignals, and visual landscape features. But because these cues
only guide orientation but don’t explain navigation, it is often thought that
birds are somehow able to align these cues with an internal map of the earth’s
surface, of supposedly characteristic features of the earth’s magnetic field, or
even with an olfactory map, and thus find their way home (Akesson, 2003;
Berthold, 2012; Gagliardo, 2013; Newton, 2008; Wallraff, 2005). Yet, it is
difficult to imagine how and why birds, even non-migratory species, should
possess an internal map of the earth’s surface that includes geomagnetic
and/or olfactory features for long-distance homing from locations they have
never been to, and topographic features for shorter distances, as it is often
assumed. Thus the concept of an internal map still needs to be regarded as a
hypothesis that has not so far been well established (Thorup & Holland, 2009;
Berthold, 2012). Consequently, scientists seem at a loss to explain how bird
navigation works. In the words of Berthold:–
It needs to be stated explicitly at the outset that, up to the present, it is entirely

Journal of the Society for Psychical Research [Vol. 79.3, No. 920
148
unknown how determination of their geographical positions and destinations is
performed by birds. This is the last big riddle of animal migration. There exist neither
explanations nor fully satisfying hypotheses. [Berthold, 2012, p.167, my translation]
Many other authors advance similar concerns, although in less drastic
terminology. For example, Wallraff (2005, p.189) states that “even the basic
features of pigeon homing are not yet understood in full detail”. Newton (2008,
p.265) concludes that in addition to compass orientation systems, “birds also
have a map sense that enables them to find familiar places again, often by
direct flight. Exactly how they achieve this remains a mystery, despite our
knowledge of the directional cues involved”, and Thorup and Holland (2009)
assert that it is still unknown how birds determine their geographical position
on the globe, especially with regard to longitude. In fact, tests of the different
possible cues that guide orientation have usually led to inconsistent or contra-
dictory results. At present, the most intensely discussed cues that may govern
bird navigation are geomagnetism and airborne odours. In the sections that
follow, I provide a very brief overview of these two subjects.
Geomagnetism as Potential Factor Guiding Navigation
Regarding geomagnetism, it is established that the homing success and the
straightness of homing routes taken were not impeded in experiments when
birds were fitted with magnets or electric coils that disturbed the perception of
geomagnetic cues (e. g., Bonadonna, Benhamou & Jouventin, 2003; for other
references see Wallraff, 2005). Besides, geomagnetic cues rarely contain three-
dimensional information that would allow a bird to determine its geographical
position at the release site after it was transported there without having access
to any kind of useful cues. Magnetic patterns may also be largely distorted in
areas with geomagnetic anomalies, and geomagnetic parameters at a given
location may change considerably during the timespan of a long-lived bird;
even the positions of the magnetic poles move over the centuries. Furthermore,
the geomagnetic field is prone to disturbances from solar flares (Sheldrake,
2002; Newton, 2008). Consequently, proponents of the olfactory approach stress
that the geomagnetic sense might play only a minor role in bird orientation,
and perhaps none in navigation (e.g. Wallraff, 2005; Gagliardo, 2013).
Volatile Chemosignals as Potential Factors Guiding Navigation
In fact, and surprising as it may be, the experimental evidence in favour of
the olfactory hypothesis appears to be more convincing (Gagliardo, 2013; Papi,
1991; Wallraff, 2005). Nevertheless, it is still regarded with great scepticism
in the ornithological community — some authors don’t even seem to consider
it worthy of explicit discussion (Newton, 2008; Wiltschko & Wiltschko, 1999).
Indeed, conceptual difficulties make it problematic to imagine how exactly
the olfactory system would govern the pigeon’s flight (Berthold, 1996, 2012;
Gould, 2009; Gould & Gould, 2012; Rozhok, 2008; Sheldrake, 2002; Thorup &
Holland, 2009; Wiltschko, 1996). To find home using olfactory cues, pigeons
must detect a gradient of familiar molecules in the atmosphere, or stable ratios
of different volatile compounds. Yet, the air above a given geographical region
is rarely in a stable condition. Sometimes, masses of air displaced in their
entirety, some are divided into different layers, or are mixed with other layers.

July 2015] Mysterious Ways: The Riddle of Homing Ability
149
It remains to be established whether the postulated molecular gradients are
detectable under these conditions over distances of hundreds or thousands of
kilometres, and whether these kinds of distinct atmospheric gradients exist
above the uniform surface of oceans, as suggested by some (e.g. Bonadonna,
Benhamou & Jouventin, 2003), and what the nature of these compounds might
be. Furthermore, especially in certain regions of the earth, the wind direction
changes frequently, and thus will make it difficult to detect and follow a stable
gradient of osmophorous molecules. Müller (1965) stated that the direction of
the wind did not affect the dog’s homing success, and excluded an olfactory
mechanism for dog homing.
Conclusions about Bird Navigation
Given such practical and theoretical difficulties that surround each of the
proposed guiding cues, ornithologists postulate that orientation, and most
likely also navigation, is mediated by a complex redundant system that integ-
rates different cues, which themselves include different sub-systems (Berthold,
1996, 2012; Newton, 2008). Nevertheless, it remains unclear how exactly this
is supposed to work. The unsolved question about how birds can possess an
“internal map” of whatever kind, even for regions unknown to them, adds
additional weight to these difficulties. Hence, ornithologists are still engaged
in controversial debates. Some simply seem to hope that one or more of the
possible methods of navigation would work somehow without going into details
(e.g. Rappole, 2013), and proponents of a certain approach neglect references
to contradictory or problematic findings (Gagliardo, 2013; Wiltschko & Wiltsch-
ko, 1999, 2010). Still, all are in principle unwilling to take potential unorthodox
cues such as extra-sensory perception (ESP) or ‘morphic fields’ in the sense of
Sheldrake (2002, 2011) as a plausible governing factor into account. If ESP
or morphic fields are mentioned at all, they are rejected with utter conviction
(e.g. Wallraff, 2005).
EXPLANATORY HYPOTHESES FOR MAMMAL HOMING
The homing ability of mammals is fascinating in its own right and might
even contribute to a better understanding of the homing ability in birds,
although it is not clear how far cues governing mammal homing are relevant for
bird navigation. In a more practical sense, mammal homing is of importance
in the context of the removal of ‘nuisance’ animals (large carnivores, bears,
or elephants, and in the context of (re-)introducing individuals of endangered
species into a new territory (for reviews, see Rogers, 1988; Linnell et al., 1997).
Some animals that have been displaced in these contexts have successfully
found home across large distances. For example, a white-tailed deer travelled
560 km, a polar bear returned from 470 km away, a bat from 320 km (Rogers,
1988), two cougars found home from distances of 465 and 490 km (Ruth, Logan,
Sweanor, Hornocker & Temple, 1998), and a bear returned home from 389 km
(Landriault, Hall, Hamr, & Mallory, 2006). Among reptiles, a crocodile that
was flown across Cape York in Australia (air-line distance 126 km) followed
the coast line of Cape York for 411 km until it reached its home again, thereby
continuously increasing the distance to its home territory during the first half
of its travel (Read, Grigg, Irwin, Shanahan & Franklin, 2007). Sheldrake (2011)

Journal of the Society for Psychical Research [Vol. 79.3, No. 920
150
lists several anecdotal reports in which dogs and cats, but also horses, sheep
and a pig, found their way home after being displaced to distant locations
unknown to them. Yet experimental research that has addressed potential
underpinnings of the homing abilities in terrestrial mammals in the wild is
clearly under-developed.
Research into Mammal Navigation
Many experiments have, however, been performed to assess the initial
orientation of mammals after their release at an unknown location. Sometimes
the animals, often small rodents, were released from a cage in the field, but
these experiments are difficult to interpret because the agitated animals might
look for shelter in the first place instead of identifying their homeward direction
(Bovet, 1992). In other experiments, animals were put into symmetrical
labyrinths or orientation arenas in a laboratory. A classical test series was
performed by Precht and Lindenlaub (1954) with several dozen cats. They
used a labyrinth with a radial symmetrical base area, and transported it
to different locations unknown to these animals. The cats themselves were
transported in sacks that prevented them from seeing their surroundings. By
observing through which of 24 exits the cats were leaving the labyrinth, the
authors intended to determine the initial orientation of the displaced animals.
In two experimental series, the cats left the labyrinth in the direction of their
home with a statistically highly significant frequency — at least, if the displace-
ment distances were not greater than 5 km. At a distance of 12 km, the results
were distributed more or less evenly. Lindenlaub (1955, 1960) repeated these
experiments successfully with mice. They even preferred their home direction
when the labyrinth was placed under a Faraday cage, thus excluding the
possibility that they made use of electromagnetic cues (Lindenlaub, 1960).
Bovet (1971) pointed out that the best results had been obtained by Lindenlaub
and by himself, and also by another experimenter (Fisler, 1967) when all
visual clues were removed from the mice. Karlsson’s (1984) results obtained
with bank voles support this finding, but only with adult and sub-adult
individuals that already possessed a home range. These observations seem in
line with Richardson’s assertion that his dogs often found their keepers better
on black nights or in mist than in daylight with several stimuli affecting their
usual senses, or after explicit training to disregard these stimuli. Karlsson’s
observations also seem convergent with Müller’s finding that the homing
ability was most pronounced in alpha-animals for whom the possession of a
home territory seemed most important.
Explanations for Mammal Navigation
In general, it is thought that mammal navigation functions along similar
lines as bird navigation, with a geomagnetic sense playing the most important
role. In fact, there is a growing body of evidence that shows that mammals are
able to perceive geomagnetic cues (Begall, Burda & Malkemper, 2014; Begall,
Malkemper, Červený, Nĕmec, & Burda, 2013). Also, dogs seem to be sensitive
to the earth’s magnetic field, even to small variations of it (Hart et al., 2013).
But again, it remains unclear to what extent this (subconscious? ) ability to
perceive geomagnetic stimuli might help them to navigate home, especially

July 2015] Mysterious Ways: The Riddle of Homing Ability
151
if animals such as dogs change their alignment and thus their orientation,
according to small geomagnetic alterations and variations, as reported in Hart
et al. (2013). Obviously, a cue that would allow for proper navigation should
be as stable as possible to avoid the choice of wrong directions because of cue
fluctuations.
As for humans, most attempts to find a geomagnetic sense for orientation
were unsuccessful, and the few positive results need to be regarded with
caution (Begall et al., 2013, 2014; Bovet, 1992; Finney, 1995).4 However,
anthropologist Ben Rudolph Finney (1995) described two instances of persons
who successfully determined directions in small sailing canoes when they were
sailing without technical aid across the open sea. When no cues for orientation
were available, one of them concentrated intensely and slowly rotated his body
until he felt that a certain direction felt different, correctly assuming that this
was South — a typical example for orientation. The other, a Hawaiian named
Nainoa, Finney’s navigator with whom he sailed thousands of nautical miles
across the Pacific Ocean, was apparently even able to perform proper navigation.
Once, Nainoa relaxed after a prolonged and desperate struggle to find a visual
cue in a black night on the ocean with shifting winds. Suddenly, he felt a
strange calmness, and “knew” where the moon was. Nainoa described it as if
“there was something that allowed me to understand where the direction was
without seeing it. And it was almost like I would just give up fighting to try to
find something with my eyes” (Finney, 1995, p.503). On later voyages, Nainoa
has allegedly had similar experiences, and he started to cultivate them. Finney
reports that he strived “to keep his mind blank and avoids straining to detect
direction from the wind, swells, or glimpses of celestial bodies through the
overcast. Then, when he gets a directional feeling, he tries to accept it without
question, for ‘if you doubt your feelings you are lost’ ” (p.503). Sometimes,
Nainoa would even see a mental image of the location and the distance of
islands he was approaching. Finney speculated that these individuals were
able to apply correctly a usually unconscious sense of direction that would be
based on geomagnetism. But again, I find it hard to imagine how one can detect
the position of the moon or an island out of sight with a sense of direction that
relies on magnetism.
CONCLUDING REMARKS ON THE HOMING ABILITY OF DOGS AND OTHER
VERTEBRATES
Although many hypotheses have been put forward to explain bird and
mammal navigation, and even though some are obviously relevant for animal
orientation, there is at present no conclusive theory for navigation, and also
no consensus among scientists with regard to the most suitable model. This is
especially valid for the most intriguing cases, such as Müller’s dogs or Nainoa.
Apart from that, what may be learned from the literature reviewed in this
article is the following :–
4 It seems that most humans are not even able to sense a strong magnetic field. This has already been
tested frequently in the context of Animal Magnetism and the researches on the odic force postulated by
Karl von Reichenbach (for a review, see Nahm, 2012). Only some ‘sensitives’ were found to be able to
detect the presence of a strong magnet in the dark.

Journal of the Society for Psychical Research [Vol. 79.3, No. 920
152
• Dogs, but also other terrestrial vertebrate species and humans, seem to
have access to an as yet unexplained ability to locate their home.
• Not all animals of a given species display the impulse and/or the ability for
homing. One factor that determines whether animals make use of their
homing ability appears to be the social ranking of the individual and the
emotional bond with its home.
• This ability appears to work most reliably under conditions when no
distinctive external cues are available, and possibly after entering a
special state of mind. Finally, homing dogs seem to assume a characteristic
posture, trotting home with raised heads and without sniffing, once they
have successfully orientated themselves towards their goals.
• The ‘sense of direction’ can apparently be trained in individuals receptive
to it.
As mentioned already, there are numerous anecdotes in which an animal
found a beloved person, rather than a definite location, from another location
(Gaddis & Gaddis, 1970; Rhine & Feather, 1962; Sheldrake, 2011). Given the
truth of these anecdotes, they would fit easily into the deductions just presen-
ted, and they would clearly demonstrate that the usual hypotheses to explain
navigation need to be supplemented by an important factor akin to ESP.
Indeed, Richardson also thought along these lines; in his last book (Richardson
& Richardson, n.d.), he stressed again that the homing ability of dogs could not
be explained with the use of the normal senses, and likened it to clairvoyance
(p.164). In addition, he included several anecdotes that in his opinion demonst-
rated telepathic interactions between dogs and their keepers, and other dogs.5
The hypothesis concerning the proposed additional factor aiding navigation
can be tested experimentally. A promising approach would consist of training
animals to find a beloved person, or a beloved mobile home. Sheldrake (2002)
has already attempted to train pigeons to locate displaced pigeon lofts. These
attempts failed for various reasons, and the implementation of the required
training and tests was time consuming and far from being trivial. Training
dogs along the lines described by Richardson (n.d., 1921) might in principle
be cheaper and easier, although many dog owners might be reluctant to offer
their dogs for experiments. Hence, the final stages of these tests especially
should be performed over sufficiently large and safe terrain. Moreover, it is not
necessary any more to employ teams of observers to follow the dogs on their
5 Some of the reports Richardson cited related to crisis experiences, and they were taken from the
Journal of the Society for Psychical Research, the Revue Spirite, the Annales des Sciences Psychiques, and
Light. Apparently, he showed interest in psychical research. Richardson also mentioned in passing that
he once lived in a haunted house, one wing being disturbed by “unaccountable noises and happenings”.
The dog he lived with at that time was afraid of this part of the house and never went there (Richardson
& Richardson, n.d., p.23). Moreover, he reported an example of a mother who was searching for the spot
where her son died on the battlefield (Richardson, 1921, p.54). She was accompanied by two officers with
a map of the site. While the officers searched the location with the help of their maps, she waited, and
had the distinct feeling that they were searching in the wrong places. After several fruitless attempts,
they followed the mother to where she would have gone to from the start, and she was right. She told
Richardson that she “just knew” where the place was. Readers familiar with the literature on psychical
research and crisis experiences in particular might not be surprised by such accounts, but their possible
relation to animal navigation is intriguing.

July 2015] Mysterious Ways: The Riddle of Homing Ability
153
routes, as it was the case in the experiments of Schmid (1932) and Müller
(1965). A convenient way to monitor the whereabouts of the dogs constantly
consists in fitting them with tracking devices such as are used in telemetry
studies. The signals emitted would enable a dog to be retrieved if it got lost,
and allow their routes to be studied in far more detail than previously possible.
Considering the apparent impasse of current research into animal navigation,
and the enormous efforts, financial budgets, technical equipment, and intricate
experimental designs that have already been employed in researching animal
homing over the last century, such experiments appear comparatively cheap
and realizable. They might result in intriguing findings that would complement
and advance the presently available hypotheses on animal navigation in
important respects.
Ida-Kerkovius-Strasse 9
79100 Freiburg i.Br., GERMANY
REFERENCES
Akesson, S. (2003) Avian long-distance navigation: experiments with birds. In Berthold,
P., Gwinner, E. and Sonnenschein, E. (eds.) Avian Migration, 471–492. Berlin:
Springer.
Begall, S., Burda, H. and Malkemper, E. P. (2014) Magnetoreception in mammals.
Advances in the Study of Behavior 46, 45–88.
Begall, S., Malkemper E. P., Červený, J., Nĕmec, P. and Burda, H. (2013) Magnetic
alignment in mammals and other animals. Mammalian Biology 78, 10–20.
Berthold, P. (1996) Control of Bird Migration. London: Chapman & Hall.
Berthold, P. (2012) Vogelzug (7th edition) Darmstadt: Wissenschaftliche Buchgesellschaft.
Bonadonna, F., Benhamou, S. and Jouventin, P. (2003) Orientation in ‘featureless’
environments: the extreme case of pelagic birds. In Berthold, P., Gwinner, E. and
Sonnenschein, E. (eds.) Avian Migration, 367–377. Berlin: Springer.
Bovet, J. (1971) Initial orientation of deer mice (Peromyscus maniculatus) released on snow
in homing experiments. Zeitschrift für Tierpsychologie 28, 211–216.
Bovet, J. (1992) Mammals. In Papi, F. (ed.) Animal Homing, 321–361. London: Chapman
& Hall.
Dutton, D. and Williams, C. (2009) Clever beasts and faithful pets: a critical review of
animal psi research. JP 73, 43–70.
Finney, B. (1995) A role for magnetoreception in human navigation ? Current Anthropology
36, 500–506.
Fisler, G. F. (1967) An experimental analysis of orientation to the homesite in two rodent
species. Canadian Journal of Zoology 45, 261–268.
Gaddis, V. and Gaddis, M. (1970) The Strange World of Animals and Pets. New York, NY:
Cowles.
Gagliardo, A. (2013) Forty years of olfactory navigation in birds. Journal of Experimental
Biology 216, 2165–2171.
Gould, J. L. and Gould, C. G. (2012) Nature’s Compass: The Mystery of Animal
Navigation. Princeton, NJ: Princeton University Press.
Gould, J. L. (2009) Animal navigation: a wake-up call for homing. Current Biology 19,
R338–R339.
Hart, V., Nováková, P., Malkemper, E. P., Begall, S., Hanzal, V., Ježek, M., Kušta, T.,
Nĕmcová, V., Adámková, J., Benediktová, K., Červený, J. and Burda, H. (2013) Dogs are
sensitive to small variations of the Earth’s magnetic field. Frontiers of Zoology 10, 80.
DOI: 10.1186/1742-9994-10-80
Herrick, F. H. (1922) Homing powers of the cat. Scientific Monthly 14, 525–539.

Journal of the Society for Psychical Research [Vol. 79.3, No. 920
154
Irwin, H. J. and Watt, C. A. (2007) An Introduction to Parapsychology (5th edition).
Jefferson, NC, and London: McFarland.
Karlsson, A. F. (1984) Age-differential homing tendencies in displaced bank voles,
Clethrionomys glareolus. Animal Behaviour 32, 515–519.
Landriault, L. J., Hall, M. N., Hamr, J. and Mallory, F. F. (2006) Long-range homing by an
adult female Black Bear, Ursus americanus. The Canadian Field Naturalist 120, 57–60.
Lindenlaub, E. (1955) Über das Heimfindevermögen der Säugetiere. II. Versuche an
Mäusen. Zeitschrift für Tierpsychologie 12, 452–458.
Lindenlaub, E. (1960) Neue Befunde über die Anfangsorientierung von Mäusen. Zeitschrift
für Tierpsychologie 17, 555–558.
Linnell, J. D. C., Aanes, R., Swenson, J. E., Odden, J. and Smith, M. E. (1997) Translocation
of carnivores as a method for managing problem animals: a review. Biodiversity and
Conservation 6, 1245–1257.
Lohmann, K. J., Lohmann, C. F. M., Brothers, J. R. and Putman, N. F. (2013) Natal
homing and imprinting in sea turtles. In Wyneken, J., Lohmann, K. J. and Musick, J. A.
(eds.) The Biology of Sea Turtles (Vol.3), 59–78. Boca Raton, FL: CRC Press.
Lohmann, K. J., Lohmann, C. F. M. and Endres, C. S. (2008) The sensory ecology of ocean
navigation. Journal of Experimental Biology 211, 1719–1728.
Morris, R. L. (1977) Parapsychology, biology and anpsi. In Wolman, B. B. (ed.) Handbook
of Parapsychology, 687–715. Jefferson, NC: McFarland.
Müller, B. (1965) Experimentelle Untersuchungen über das Heimfinden beim Hund. Inaug-
ural-Dissertation. Basel: Philosophisch-Naturwissenschaftliche Fakultät der Universität
Basel.
Nahm, M. (2007) Evolution und Parapsychologie. Norderstedt: Books on Demand.
Nahm, M. (2012) The Sorcerer of Cobenzl and his legacy: the life of Baron Karl Ludwig von
Reichenbach, his work and its aftermath. Journal of Scientific Exploration 26, 381–407.
Papi, F. (1991) Olfactory navigation. In Berthold, P. (ed.) Orientation in Birds, 52–82.
Basel: Birkhäuser.
Pratt, J. G. (1953) The homing problem in pigeons. JP 17, 34–60.
Precht, H. and Lindenlaub, E. (1954) Über das Heimfindevermögen bei Tieren. I.
Versuche an Katzen. Zeitschrift für Tierpsychologie 11, 485–494.
Rappole, J. H. (2013) The Avian Migrant. The Biology of Bird Migration. New York, NY:
Columbia University Press.
Read, M. A., Grigg, G. C., Irwin, S. R., Shanahan, D. and Franklin, C. E. (2007) Satellite
tracking reveals long distance coastal travel and homing by translocated Estuarine Croc-
odiles, Crocodylus porosus. PLoS ONE 2 (9), e949. DOI: 10.1371/journal.pone.0000949
Rhine, J. B. (1951) The present outlook on the question of psi in animals. JP 15, 230–251.
Rhine, J. B. and Feather, S. R. (1962) The study of cases of ‘psi-trailing’ in animals. JP
26, 1–22.
Richardson, E. H. (1910) War, Police, and Watch Dogs. Edinburgh and London: Blackwood.
Richardson, E. H. (1921) The homing instinct in dogs. Psyche 2 (new series), 52–56.
Richardson, E. H. (n.d.) British War Dogs: Their Training and Psychology. London:
Skeffington. [Presumably published in 1920]
Richardson, E. H. and Richardson, B. (n.d.) Fifty Years with Dogs. London : Hutchinson.
[Presumably published in 1951]
Rogers, L. L. (1988) Homing tendencies of large mammals: a review. In Nielsen, L. and
Brown, R. D. (eds.) Translocation of Wild Animals, 76–91. Milwaukee: Wisconsin
Humane Society, and Kingsville Texas: Caesar Kleberg Wildlife Institute.
Rozhok, A. (2008) Orientation and Navigation in Vertebrates. Berlin: Springer.
Ruth, T. K., Logan, T. K., Sweanor, L. L., Hornocker, M. G. and Temple, L. J. (1998)
Evaluating cougar translocation in New Mexico. Journal of Wildlife Management 62,
1264–1275.

July 2015] Mysterious Ways: The Riddle of Homing Ability
155
Schmid, B. (1932) Zeitschrift für Hundeforschung 2, 133–156.
Schmid, B. (1936) Begegnungen mit Tieren. Munich: Knorr & Hirth.
Schwertner, P. (1984) Psi in der Tierwelt. Hannover: Landbuch.
Sheldrake, R. (2002) Seven Experiments That Could Change the World (2nd edition)
Rochester, VT: Park Street Press.
Sheldrake, R. (2011) Dogs That Know When Their Owners are Coming Home (2nd edition).
New York, NY: Three Rivers Press.
Thorup, K. and Holland, R. A. (2009) The bird GPS: long-range navigation in migrants.
Journal of Experimental Biology 212, 3597–3604.
Wallraff, H. G. (2005) Avian Navigation: Pigeon Homing as a Paradigm. Berlin: Springer.
Wiltschko, R. (1996) The function of olfactory input in pigeon orientation: does it provide
navigational information or play another role? Journal of Experimental Biology 199,
113–119.
Wiltschko, R. and Wiltschko, W. (1999) Das Orientierungssystem der Vögel. II.
Heimfinden und Navigation. Journal für Ornithologie 140, 129–164.
Wiltschko, R. and Wiltschko, W. (2010) Avian magnetic compass: its functional properties
and physical basis. Current Zoology 56, 265–276.