Skip to main content


Open Access

Dogs are sensitive to small variations of the Earth’s magnetic field

  • Vlastimil Hart1,
  • Petra Nováková1,
  • Erich Pascal Malkemper2,
  • Sabine Begall2,
  • Vladimír Hanzal1,
  • Miloš Ježek1,
  • Tomáš Kušta1,
  • Veronika Němcová1,
  • Jana Adámková1,
  • Kateřina Benediktová1,
  • Jaroslav Červený1 and
  • Hynek Burda1, 2Email author
Contributed equally
Frontiers in Zoology201310:80

DOI: 10.1186/1742-9994-10-80

Received: 25 November 2013

Accepted: 23 December 2013

Published: 27 December 2013

Back to article

Archived Comments

  1. Comment on ''Dogs are sensitive to small variations of the Earth's magnetic field''

    17 March 2014

    Ludek Pekarek, National Institute of Public Health

    In the commented paper [1], Hart et al. claims to provide proof (based on a statistical evaluation of more than 7000 observations of dogs) of:

    •  axis alignment of a dog's body with geomagnetic field lines during defecation or urination,
    • loss of this ability if the fluctuations of the Earth's magnetic field declination exceed a rate of 0.02 arc minutes per minute of time (in [1] written in percentage terms as 2 %, without repeating the dimension).

    In this comment we show that neither the north-south alignment of the dogs nor the disturbance of their anticipated magnetic sensing has been proved in [1]. The major argument leading to this conclusion lies in the miniscule value of the declination change rate that was in [1] considered sufficient for disrupting the assumed magnetic sense. It should be pointed out here that any movement of the dog that has a rotation component relative to ground will change the angle between its body axis and the position of magnetic north. In the reference frame rotating with the dog, this would necessarily be perceived as changing the angle of magnetic declination. Even the rotation by just one arc degree per second (one revolution per 6 minutes, i.e. much less than a dog routinely performs) will by orders of magnitude exceed the rate of change at the largest ever encountered fluctuation of magnetic declination. Furthermore, even if the dog were to stay motionless, imperceptible but unavoidable movements of its body still would create fluctuations of the angle between its body axis and the north-south direction at a rate larger than a geomagnetic storm. To give a particular example, if the backside of the dog would not move, while its nape (say, 0.5 m apart) was moving in the horizontal plane with velocity as small as 1 millimeter per a second of time, the rate of change of the angle of declination observed from the dog’s reference frame would approximately be equal to 40000 % in the units used in [1] (compare to 2 % claimed in [1] as disrupting).

    Apart from the fundamental overlook mentioned above, no details are given in [1] about the procedure for assigning the quiet regions of declination to the observed cases of north-south orientation (about 20 % of the observations). In addition, it was scarcely possible to keep the double blind method during this crucial step of evaluation, and strong bias could not be avoided. The uncertainty concerning the appropriateness of the categorization is therefore high. The double blind procedure also could not have been applied during the primary observation, as the dog observers knew what outcome was expected from the whole experiment. In brief, the double blind method was not fulfilled at several steps in the experiment and its evaluation.

    In conclusion, the idea of the assumed magnetic sense of dogs being disturbed by fluctuations of magnetic declination, quoted as the main discovery achieved in [1], must be rejected. In particular, the suggested far-reaching consequences challenging biologists and physicians “to seriously reconsider effects magnetic storms might pose on an organism” or “When extrapolated upon other animals and other experiments and observations on animal magnetoreception, this (i. e. the claimed disturbing effect of fluctuations of Earth's magnetic field) might explain the non-replicabilty of many findings and high scatter in others”, must be regarded as unsubstantiated.

    [1] Hart et al.: Dogs are sensitive to small variations of the earth’s magnetic field. Frontiers in Zoology 10, 80 (2013)


    Lukas Jelinek (CTU FEE Prague), Ludek Pekarek (SZU Prague)

    Competing interests

    Authors of this comment are not aware of any competing interests with respect to the authors of the commented paper.

  2. Reply to L. Jelínek and L. Pekárek's "Comment on Dogs are sensitive to small variations of the Earth's magnetic field''

    21 March 2014

    Hynek Burda, Faculty of Biology, University Duisburg-Essen

    In their comment L. Jelínek and L. Pekárek express doubts on the validity of the data published in Hart et al. (2013).

    The scepticism of the authors is based on their misinterpretation of our study, their oversimplified purely mechanistic concept of mechanisms of animal magnetoreception, and ignoring some basic principles of sensory integration.

     The major argument for mistrust of the critics "lies in the miniscule value of the declination change rate that [the authors] considered sufficient for disrupting the assumed magnetic sense." We agree that the changes are very small, yet the fact that something is unlikely (or better to say that a theoretician finds it unlikely) does not mean that the phenomenon does not exist. We described a phenomenon, which we recorded objectively and fully blindly (see below). We are aware that we provide only correlative evidence but not an explanation. However, the factor “rate of change in magnetic declination” is the best predictor of the observed dog alignment. No other single factor (wind, sun position, etc. etc.) had the power to turn a random distribution into a highly significant alignment.

    We assume, however, that the misunderstanding is based on the fact that we compared subsamples recorded during periods of 0% declination change (let us call it as percent for now, though we actually mean rate of changes) with those sampled during periods of 0.1-2.0%, 2.1%-4.0% etc. This does, however, not mean that we claim that dogs can distinguish a change of only 0.1%. Maybe we should have used more figurative classifications: the dog shows a different behaviour in a calm magnetic field (0% declination change) than in a slightly unquiet (appr. 1%: 0.1-2.0%), a disturbed (appr. 3%: 2.1-4.0%), or turbulent (>4%) geomagnetic field. Such classifications are common for wind strengths: calm, light air, light breeze, gentle breeze etc. They require quantified ranges (in knots, m/s, km/h etc.), and their limits are administratively determined. If we had written that a dog can distinguish e.g. light air from a light breeze, probably no one would object; if on the other side we had just given ranges such as calm (<1.0) vs. light air (1.1-3.0) (knots), someone could object that we were not able to measure the wind with an accuracy of 0.1 knots and that the dog cannot recognize it. We are afraid, that this is what happened here.

    In fact, the sensitivity of animals to natural geomagnetic fluctuations was reported also for other animals in other contexts. We cite such studies in our paper and would like to extend the list here for another paper on the influence of small geomagnetic variations on pigeon homing, which unfortunately, came to our attention only recently: Schiffner and Wiltschko (2011).

    The authors of the comment base their theoretical calculations of rotation components on a very naïve assumption of a dog behaving (or its magnetic compass sense functioning) like a simple technical magnetic needle. No hypothesis of magnetoreception functioning in this manner is taken seriously by any researcher in the field of biomagnetism today. We would like to point out again that we found the best correlations with the index quantifying the rate of changes of the declination. This does not mean that dogs measure changes in declination. This means only that this index correlates well with some factor characterizing fluctuation of the magnetic field and that the dog shows a behavioural reaction upon such fluctuations. Furthermore, Jelínek and Pekárek state that “…any movement of the dog that has a rotation component relative to ground will change the angle between its body axis and the position of magnetic north”. Here, the authors ignore the fact that the dogs (like humans) are able to calculate their own movements with the vestibular organ and use sensory feedback mechanisms to extract external movements from self-created noise. It’s a fundamental prerequisite of every sensory system.

    Jelínek and Pekárek claim that we do not give details about the procedure for assigning the quiet regions of declination. This is simply not true. The procedure is described on page 10-11 and it is illustrated in Fig. 4.

    Also fully unsubstantiated is the claim that the study was not blind because the dog observers knew what outcome was expected from the whole experiment. We write on page 5: "The study was truly blind. Although the observers were acquainted with our previous studies on magnetic alignment in animals and could have consciously or unconsciously biased the results, no one, not even the coordinators of the study, hypothesized that expression of alignment could have been affected by the geomagnetic situation, and particularly by such subtle changes of the magnetic declination. The idea leading to the discovery of the correlation emerged after sampling was closed and the first statistical analyses (with rather negative results, cf. Figure 1) had been performed. Also, the acquisition of data on magnetic declination was carried out without knowledge of heading values on the respective time and date." The dog observers in the field could by no means know about the magnetic conditions during the time of sampling and the data analysers who quantified the magnetic conditions later and attributed them to the sampling periods were unaware of the measured dog-directions during those respective times. Thus, there was no way of pushing the data into a “desired direction”, the procedure was blind.

    The conclusion of the critics is thus unsubstantiated and the result of misunderstanding, a mechanistically simplified idea of animal magnetoreception, and ignoring general principles of sensory integration.



    Hart et al. (2013): Dogs are sensitive to small variations of the earth’s magnetic field. Frontiers in Zoology, 10: 80.

    Schiffner I, Wiltschko R. (2011): Temporal fluctuations of the geomagnetic field affect pigeons’ entire homing flight. J. Comp. Physiol. A, 197: 765–772.



    Hynek Burda, Sabine Begall, Erich Pascal Malkemper, Vlastimil Hart, Petra Nováková

    Competing interests

    Authors of the original paper and this reply are not aware of any competing interests with respect to the authors of the comment

Authors’ Affiliations

Department of Game Management and Wildlife Biology, Faculty of Forestry and Wood Sciences, Czech University of Life Sciences
Department of General Zoology, Faculty of Biology, University of Duisburg-Essen