Almost half a century ago, the first firm evidence for magnetoreception (i.e. the ability to sense the geomagnetic field) in the robin was provided, representing the first evidence for magnetoreception in a vertebrate (Wiltschko and Merkel 1966); a quarter century later Burda et al. (Burda et al. 1990) demonstrated magnetoreception for the first time in the laboratory in African mole-rats, the first demonstration in mammals. Over the last 25 years we have revisited the question of magnetoreception in mole-rats and made some spectacular discoveries (Marhold, Wiltschko, and Burda 1997, Nemec et al. 2001, Thalau et al. 2006, Wegner, Begall, and Burda 2006, Burger et al. 2010). Subsequently, magnetoreception has been described also in the blind mole rat Spalax ehrenbergi supersp. (Kimchi and Terkel 2001, Kimchi, Etienne, and Terkel 2004, Marhold et al. 2000), the laboratory mouse Mus domesticus (Muheim et al. 2006), the Siberian hamster Phodopus sungorus (Deutschlander et al. 2003), and in the bats Eptesicus fucus (Holland et al. 2006, Holland et al. 2008) and Nyctalus plancyi (Wang et al. 2007). However, the question if large mammals also sense the geomagnetic field was left unanswered.
How to study magnetoreception in large mammals
Some mammals, such as arctic caribous, North American bisons, African elephants and gnus, and Asian saiga antelopes, and, above all, diverse cetacean species are famous for their long range migrations, which require superb navigational capacities. Based on these capacities and on correlations between navigation failures and local disturbances of the geomagnetic field we may conclude that navigation capacities are based on magnetoreception (Klinowska 1985, Kirschvink, Dizon, and Westphal 1986). Apart from such indirect evidence there are also anecdotal reports on homing abilities of dogs, cats, horses etc. (just remember "Lassie Come Home"). Systematic research on navigational capabilities using homing paradigms is technically difficult. Besides that, there could be plenty of non-navigational reasons why displaced mammals do not return home. Conditioning (training) on magnetic stimuli in large animals is methodically difficult. The animals might switch on the magnetic compass sense only in certain contexts, e.g. when being outdoors, in unfamiliar surroundings, where they cannot rely on other senses, but not under laboratory conditions.
However, looking for spontaneous expressions of magnetosensitivity, such as magnetic alignment is an easy and effective method to scan for magnetoreception in animals. Animals (and also humans) frequently adopt orientation in space which is not random (Begall et al. 2013): e.g. when sun basking they align perpendicular to solar rays (thermic alignment), trout in the brook aligns against the stream (rheo-alignment), a scenting dog aligns against the wind (anemo-alignment), the audience turns towards the speaker (alignment towards an attractor), grazing sheep on a slope wander along the contours (isocline-alignment). Apparently alignment has some advantages - it is comfortable, saves energy, or provides access to food, oxygen or information. And now imagine that there is no slope, no attractor, no wind and the sun is not visible. Will the orientation (body axis, heading) of a resting animal be random or might the animal prefer to adopt some particular compass direction? If we find that animals prefer a particular compass direction under these unconstrained conditions, we may hypothesize that such a non-random orientation represents a case of magnetic alignment, implying magnetoreception. Subsequently, we may test this hypothesis by changing the magnetic conditions and observing whether the orientation also changes. Scanning for cases of magnetic alignment is thus a suitable experimental paradigm for magnetobiological research and might provide new insights into meaning of magnetoreception in animals' everyday life. Why do animals do it? Which information do they get from the magnetic field? Can they save energy (get more comfort) when magnetically aligned? How do they sense the magnetic field?
For over six years we have been following this research program and have reported magnetic alignment in grazing and resting cattle, roe deer and red deer (Begall et al. 2008) and disturbance of this alignment through extremely low frequency magnetic fields (ELF MF) produced by high voltage power lines (Burda et al. 2009). These papers raised high resonance all over the world. Skeptics did not like particularly the idea that ELF MF might influence the behavior (and welfare?) of animals (and human health?). They attacked our results (Hert et al. 2011), but we could show (Begall et al. 2011) that their study was heavily flawed and cleaned of errors it actually fully confirmed our original findings. Our results were subsequently confirmed in an independent study by Slabý et al. (Slaby, Tomanova, and Vacha 2013). In other studies we found magnetic alignment in hunting foxes (Cerveny et al. 2011), carps schooling in circular tubs (Hart et al. 2012), and mallard ducks landing on water (Hart et al. 2013). We concluded that magnetic alignment might help animals to synchronize and coordinate movement in a group, organize cognitive (mental) maps, measure the distance and the slope, and maybe enhance (focus) selective sensory attention. Our model mammals (cattle, deer, fox) were, however, rather "unhandy" and unmanageable for further research. We envisaged that dogs represent ideal model animals for further research, since they are widespread across the world and are highly receptive to learning.
How to study magnetoreception in dogs
Our established team of sensory ecologists from Essen, Germany, and wildlife zoologists from Prague, Czech Republic, asked dog owners (relatives, friends and students) to measure, in analogy with our previous research on cattle and deer, body orientation of feeding and sleeping dogs. We soon realized, however, that body orientation during these activities is influenced, consciously or unconsciously by the master, by house geometry etc. Besides that, the geomagnetic field inside buildings is mostly disturbed, and it would require many cooperators to get sufficient sample sizes. We noticed, however, that dogs defecating and urinating during a walk (actually they are marking their territories) are very selective about where to mark. We can prevent them to mark somewhere but we can hardly make them do the job when and where we wish. Well known, yet enigmatic and unexplained (though there are many ideas) is the circling (some) dogs sometimes perform before urinating or defecating. Every dog defecates on average once and urinates (marks) many times a day; keepers go with their dogs at least three times a day to a walk; if every walk (in the morning, afternoon, and evening) would follow another route, measurements of orientation of particular dogs cannot be considered pseudoreplications, as they would be done always at a different time, at different place, under somewhat different environmental conditions. We standardized the protocol, asked our volunteers to record compass direction of the body axis (heading) of defecating and urinating dogs, which would be unrestrained by walls and fences, i.e. in open field, off the leash, and outdoors (i.e. not at the home backyard or in the familiar garden). We recruited a team of almost 40 observers following 70 dogs of almost 40 breeds over two years, recording about 1,900 defecation and 5,600 urination events. After we closed sampling, we ran circular statistical analyses and were disappointed to see that there was no common preference for any particular compass direction. We analyzed the data with respect to sex, age, breed, and body mass of the dog, time of the day and month but no clear picture emerged. The subsamples, i.e. the numbers of records collected for particular dogs were not even. One dog, a male borzoi named Diadem, was recorded on a regular basis over a long time providing about 2,500 measurements of body orientation. Diadem's data showed to be a gold mine. We noticed that on some days and during some walks Diadem showed perfect compass preference (in his case for Northwest), while during some other walks and on some particular dates his orientation during marking was fully random. It was just by chance that, when analyzing the data, we heard a radio report about geomagnetic storms and the idea was born to check the "space weather" on those days when Diadem was apparently "disoriented". A comparison between "quiet" and "stormy" days confirmed that alignment could be correlated to "magnetic weather". In subsequent weeks and months we spent many hours of leisure time looking for different indices of magnetic weather at time and place of the particular walks and checking which of them correlate best with the "alignment achievement" of the dogs. We finally found that the best predictor of alignment was the speed (rate, slope) of changes in the magnetic declination at the time interval 30 minutes before the dog was urinating or defecating. This single factor had the power to extract a highly significant alignment from a random distribution.
Most volunteers were aware of our previous findings and an unconscious bias in measurements could not be excluded. (Meanwhile we developed smart phone apps to measure the compass direction blindly, i.e., without showing the measured value.) None of us or of the volunteers who measured their dogs knew, however, the "magnetic weather" when taking or analyzing the measurements. Thus, the study was indeed fully blind.
Why do the dogs do it?
The dog has a mental map of its home range or creates such a map whenever being in an unfamiliar region. Our hypothesis is that when the dog marks or defecates, it uses these short stops to store the location ("coordinates") of marking cornerstones (as landmarks) in its memory. Doing so and/or calibrating its magnetic compass is probably easier for it when being aligned with the magnetic field. (When we read a map and use a compass outdoors we also align according to some directions and landmarks). Apparently, the dog needs calm "magnetic weather" to do so.
The publication and its impact
We were quite excited about our findings as this was the first evidence for a) magnetoreception in dogs, b) sensitivity to such small changes of the geomagnetic field in a mammal, c) sensing small changes in polarity (rather than in intensity) of the magnetic field in an animal. Knowing that dogs can sense magnetic fields opens up new possibilities for investigation of principles and mechanisms of magnetoreception. Knowing that animals may sense small oscillations in the geomagnetic field and react with a behavioral change calls for considering this factor when studying and analyzing many aspects of animal behavior.
Of course, we wanted to publish our results in a top multidisciplinary journal. Yet all the journals returned the manuscript immediately - as being of only minor interest for their general readerships. We assume that the editors were afraid to become embarrassed by publishing a study dealing with something as trivial as defecation. Maybe they have misunderstood its message. We are very grateful to Frontiers in Zoology, a top journal in zoology, for letting the manuscript to be peer reviewed. After passing the very quick but also very thorough review process, the paper was published during Christmas holidays in 2013. Still before our universities published a press release and before we made any advertisement for it, a report about it was published in the Journal of Improbable Research. Within one week after publication, the paper achieved an altmetric score of 1,715 and was accessed about 740,000 times. (The score rose to 2,854 and the number of accesses to almost 880,000 three months later.) We were contacted by many journals and newspapers, internet sites, radio and TV news, professionals and interested laymen. Many published reports were based on tertiary sources and most of the "reporters" have never read the original article, although it is freely accessible in the internet. The message of our article was reduced to the statement that dogs prefer to face northwards when defecating. (Something we never claimed.) Many applauded while others berated. We were often rebuked for wasting tax money for nonsense research. People claiming this have perhaps never deliberated on possible costs of such studies. This research on dogs was done in our leisure time - we walked with our dogs in the same way as millions of people do all over the world - but we took also compass, paper and pencil, and in the evening when others were watching TV, surfing in the web or going in for their hobbies, we analyzed our results. So - it was our hobby and we had fun doing it. We have experienced immense reactions upon our research already before, when we described magnetic alignment in cattle. Our studies dealing with other animals aroused much less interest. Based on our experience we were not surprised that our article on dogs made quite a splash. Many dog owners assume they "know" everything about dogs and are qualified to comment. Very few people have pet deer, foxes, mole-rats or shrews, so our work on these animals did not challenge them personally.
A study that makes people laugh and then they start to think
The Journal of Improbable Research which started this hype is dedicated to research that makes people laugh and then think. We believe that this will be also the fate of our paper. Indeed, finding that dogs can sense and recognize small fluctuations of the geomagnetic field provides us with lots of stuff to think seriously about.
Begall, S., H. Burda, J. Cerveny, O. Gerter, J. Neef-Weisse, and P. Nemec. 2011. "Further support for the alignment of cattle along magnetic field lines: reply to Hert et al." J Comp Physiol A no. 197 (12):1127-1133. doi: 10.1007/s00359-011-0674-1.
Begall, S., J. Cerveny, J. Neef, O. Vojtech, and H. Burda. 2008. "Magnetic alignment in grazing and resting cattle and deer." Proc Natl Acad Sci U S A no. 105 (36):13451-5. doi: 10.1073/pnas.0803650105.
Burda, H., S. Begall, J. Cerveny, J. Neef, and P. Nemec. 2009. "Extremely low-frequency electromagnetic fields disrupt magnetic alignment of ruminants." Proc Natl Acad Sci U S A no. 106 (14):5708-13. doi: 10.1073/pnas.0811194106.
Burda, H., S. Marhold, T. Westenberger, W. Wiltschko, and R. Wiltschko. 1990. "Magnetic compass orientation in the subterranean rodent Cryptomys hottentotus (Bathyergidae, Rodentia)." Experientia (46):528-530.
Burger, T., M. Lucova, R. Moritz, H. OelschlŠger, R. Druga, H. Burda, W. Wiltschko, R. Wiltschko, and P. Nemec. 2010. "Changing and shielded magnetic fields suppress c-Fos expression in the navigation circuit: input from the magnetosensory system contributes to the internal representation of space in a subterranean rodent." Journal of The Royal Society Interface no. 7 (50):1275-1292. doi: 10.1098/rsif.2009.0551.
Cerveny, J., S. Begall, P. Koubek, P. Novakova, and H. Burda. 2011. "Directional preference may enhance hunting accuracy in foraging foxes." Biology Letters no. 7 (3):355-357. doi: 10.1098/rsbl.2010.1145.
Deutschlander, M., M. Freake, C. Borland, J.B. Phillips, R.C. Madden, L. Anderson, and B. Wilson. 2003. "Learned magnetic compass orientation by the Siberian hamster, Phodopus sungorus." Animal Behaviour no. 65 (4):779-786. doi: 10.1006/anbe.2003.2111.
Hart, V., T. Kusta, P. Nemec, V. Blahova, M. Jezek, P. Novakova, S. Begall, J. Cerveny, V. Hanzal, E. P. Malkemper, K. Stipek, C. Vole, and H. Burda. 2012. "Magnetic alignment in carps: evidence from the Czech christmas fish market." PLoS One no. 7 (12):e51100. doi: 10.1371/journal.pone.0051100.
Hart, V., E. P. Malkemper, T. Kusta, S. Begall, P. Novakova, V. Hanzal, L. Pleskac, M. Jezek, R. Policht, V. Husinec, J. Cerveny, and H. Burda. 2013. "Directional compass preference for landing in water birds." Front Zool no. 10 (1):38. doi: 10.1186/1742-9994-10-38.
Hert, J., L. Jelinek, L. Pekarek, and A. Pavlicek. 2011. "No alignment of cattle along geomagnetic field lines found." Journal of Comparative Physiology A no. 197 (6):677-682. doi: 10.1007/s00359-011-0628-7.
Nemec, P., J. Altmann, S. Marhold, H. Burda, and H. Oelschlaeger. 2001. "Neuroanatomy of magnetoreception: The superior colliculus involved in magnetic orientation in a mammal." Science no. 294 (5541):366-368. doi: 10.1126/science.1063351.
Slaby, P., K. Tomanova, and M. Vacha. 2013. "Cattle on pastures do align along the North-South axis, but the alignment depends on herd density." Journal of Comparative Physiology A no. 199 (8):695-701. doi: 10.1007/s00359-013-0827-5.
Thalau, P., T. Ritz, H. Burda, R. E. Wegner, and R. Wiltschko. 2006. "The magnetic compass mechanisms of birds and rodents are based on different physical principles." J R Soc Interface no. 3 (9):583-7. doi: 10.1098/rsif.2006.0130.
This article and its reviews are distributed under the terms of the Creative Commons Attribution 4.0 International License, which permits unrestricted use, distribution, and redistribution in any medium, provided that the original author and source are credited.