Journal of Comparative Physiology A

, Volume 199, Issue 8, pp 695–701 | Cite as

Cattle on pastures do align along the North–South axis, but the alignment depends on herd density

Original Paper


Alignment is a spontaneous behavioral preference of particular body orientation that may be seen in various vertebrate or invertebrate taxa. Animals often optimize their positions according to diverse directional environmental factors such as wind, stream, slope, sun radiation, etc. Magnetic alignment represents the simplest directional response to the geomagnetic field and a growing body of evidence of animals aligning their body positions according to geomagnetic lines whether at rest or during feedings is accumulating. Recently, with the aid of Google Earth application, evidence of prevailing North–South (N–S) body orientation of cattle on pastures was published (Begall et al. PNAS 105:13451–13455, 2008; Burda et al. PNAS 106:5708–5713, 2009). Nonetheless, a subsequent study from a different laboratory did not confirm this phenomenon (Hert et al. J Comp Physiol A 197:677–682, 2011). The aim of our study was to enlarge the pool of independently gained data on this remarkable animal behavior. By satellite snapshots analysis and using blinded protocol we scored positions of 2,235 individuals in 74 herds. Our results are in line with the original findings of prevailing N–S orientation of grazing cattle. In addition, we found that mutual distances between individual animals within herds (herd density) affect their N–S preference—a new phenomenon giving some insight into biological significance of alignment.


Magnetic alignment Cattle Positions Replication Magnetoreception 


Magnetoreception is a remarkable ability of animals to perceive the magnetic field of the Earth (Wiltschko and Wiltschko 2006). Direction, length and dip of the magnetic vector provide information that may serve animals to choose and keep the right azimuth on their long migrations (Fraser 2010; Johnsen and Lohmann 2005) or to assess the current position with respect to the goal of the journey. Nevertheless, magnetoreception was also found in species which do not undertake long migrations, e.g. fruit fly (Dommer et al. 2008; Gegear et al. 2008), chicken (Wiltschko et al. 2007) or zebrafish (Takebe et al. 2012). Such species seem to utilize their magnetic sense in situations different from the typical compass utilization by humans. The sense for direction of the Earth magnetic field, also termed as the geomagnetic field (GMF), may become evident only as a magnetic alignment—a spontaneous preference of specific body position with respect to North–South (N–S) axis (reviewed recently in Begall et al. 2012). Such a remarkable behavior was shown in a number of species across the animal kingdom, e.g. honeybees (Hsu et al. 2007), cockroaches (Vacha et al. 2010), fruit flies (Dommer et al. 2008; Gegear et al. 2008), zebrafish (Takebe et al. 2012) and carps (Hart et al. 2012), voles (Mather and Baker 1981; Burda et al. 1990; Kimchi and Terkel 2001; Deutschlander et al. 2003; Muheim et al. 2006), bats (Holland et al. 2006; Wang et al. 2007) or foxes (Cerveny et al. 2011).

Although elaborate biophysical models of reception mechanisms have already been put forward (Ritz et al. 2010; Cadiou and McNaughton 2010; Maeda et al. 2012), firmly based experimental evidences of how GMF energy is transduced into nervous signals have not been forthcoming (reviewed, e.g. in Johnsen and Lohmann 2005; Wiltschko and Wiltschko 2005).

Unlike other sensory capacities known to men the adaptive profit of geomagnetic sensitivity may not always be intuitively clear. Magnetic alignment is an example of behavior posing an unresolved problem for its adaptive value. To date, only speculative answers have been suggested (Begall et al. 2012), provided magnetic axis creates a visible pattern superimposed on the visual surroundings (Phillips et al. 2010). It may serve as an independent space frame or metrics giving the animal 3-D coordinates well usable when assessing distance and spatial position. Adjusting the body axis within this magnetic frame may provide the animal with some basic-state perceptual comfort or serve as a range-finder when hunting (Cerveny et al. 2011).

In research on animal magnetoreception, there is a constant need for experimental verifications of existing hypotheses. The progress of the experimentalists’ work, however, is not easy and fast. To serve as a reliable base for refining theoretical concepts, the collection of primary data should address a number of methodical concerns (Kirschvink et al. 2010). The gold standard for science is independent reproducibility and the situation in magnetoreception research does not appear to be ideal. While successful replication of studies is an exception (Kirschvink and Kirschvink 1991) negative attempts at replication prevail. In our opinion, experimental work focused on replication/reproducibility of published data would prove to be very valuable in the field of magnetobiology.

The report of prevailing N–S body alignment in grazing cattle (Bos primigenius f. taurus) presumably based on magnetoreception (Begall et al. 2008) attracted the attention of media and discussion forums which is rather unusual in the field. The unique features of the report were based upon both simplicity and originality of the data collection methods (evaluation of Google Earth frames) and, especially, the biological phenomenon itself. To date, this original study was subjected to a single independent verification by Hert et al. (2011). This study did not find significantly oriented animals and resulted in a critical assessment of both the N–S alignment as well as magnetoreception as being the most likely explanation of the phenomenon. The primary aim of our work was to provide another independent attempt at this assessment of cattle alignment. In the course of the work, one additional important question was addressed—the relationship between herd density and N–S alignment.


Herd and animal selection

In line with the original work of Begall et al. (2008) the orientations of main body axes were estimated by means of Google Earth satellite pictures. We concentrated on regions with magnetic declination (angle between magnetic and geographic North) close to zero (from −10° to +10°). Cattle herds on pastures were mostly visible from the eye altitude of 1,000 m. Copies of satellite pictures from the altitude between 100 and 1,000 m were stored if the resolution of details was good enough so that determination of cows body axes and differentiation between cows and sheep (smaller and of round shape) were possible. Cows clustered around feeders or drinkers as well as cows following the visible track were not scored.

The herds that we encountered were processed according to criteria set by Begall et al. (2008) in their original study. Some of the criteria we adopted were even stricter. The distance between herd and settlements, communications, forests, feeder/water or water area had to be at least 25 m. A herd was accepted for assessment if it was found on a pasture with an elevation smaller than 5 m over 100 m in any direction. The distance between herd and pasture border or fence had to be at least 15 m.

We also considered the biasing impact of electric power lines reported by Burda et al. (2009), and in our study all herds having <150 m distance to nearest high voltage power-lines were discarded. For common low-voltage power lines the distance from herd was set at 50 m as an acceptable minimum.

There had to be at least ten cows in each group to be considered a herd and only animals within herds were scored. Visually distinct groups separated from other animals were ranked as a herd. One large pasture recognizable according to fences or different color of background could host more than one herd. Solitary and isolated individuals were not scored.

Processing of images

Images of herds were cut out in square shapes. Calves, clustered animals or cows with undetectable main body axis orientation were marked and excluded from the sample. If there were more than 30 % of excluded individuals, the herd was not evaluated. Animals lying down were not differentiated from standing ones. Evaluation was blind: images of herds were randomly rotated by 0°, +90° or −90° and subsequently sent to different persons to be evaluated. Persons scoring the animal positions were not aware of actual orientation of the pictures. Angle between animal axis and geographic N–S axis was found by means of Screen Protractor (Iconico Software). After re-adding +90° or −90° according to original random rotation, statistical evaluation (Rayleigh circular statistic) of final angles was applied.

Herd density

We counted the area of each herd by means of the freely accessible Google Earth application ( and calculated its density as a number of animals per respective area. Herds were ordered according to their density and divided into three groups of approximately same size: low- (24 herds), middle- (26 herds) and high-density herds (24 herds).


Angles of body orientations with geographic North as reference were evaluated by means of the classical method of circular statistics—Rayleigh test. We analyzed both individual positions (individual as a unit) and—due to social relationships biasing individual behavior within a herd—mean vectors of whole herds (herd as a unit). In the former case, all individual positions were pooled together regardless of herds and mean vector direction (μ) and its length (r, a measure of non-random distribution) were calculated by means of first-order Rayleigh test. In the latter case, directions of mean vectors were calculated for each herd and processed by second-order Rayleigh test and respective values (μ, r) were calculated as well. In this procedure, the lengths of the mean vectors of separate herds were not considered (Batschelet 1981) and all herds contributed equally regardless of their parameter r.

Since we did not distinguish between front and rear parts of bodies, all measured angles were axial/bimodal (in the interval 0°–180°) and were processed by doubling the angles method (Batschelet 1981): data are doubled before analysis, thus being converted from bimodal to unimodal. The resulting mean vector is then back-converted, so that the range in the interval was 0°–180° again. Therefore, all data are presented as XX°/XX° + 180°. For illustrative purposes, on circular diagrams, each direction is depicted as a pair of points giving symmetrical pattern.


We evaluated 2,235 cows from 74 herds from various regions of Europe and North America (Table 1); additional data about herd position are given in online resource. Evaluation of body axes of all individuals from all herds (pooled data regardless of herds) gave a significant distribution (first order Rayleigh test, 2,235 individuals, r = 0.21, p < 0.00001) with almost exact N–S direction (μ = 179°/359° with geographic North as reference) (Fig. 1a). Mean vectors of 74 evaluated herds were also clustered significantly (second-order Rayleigh test, r = 0.27, p = 0.00467) close to N–S axis (μ = 5°/185° with geographic North as reference) (Fig. 1b).
Table 1

Numbers of evaluated herds and individuals in particular regions


Czech Republic


The Netherlands




Sum (Σ)

















Fig. 1

N–S orientation of cattle is apparent when both individuals (a), and herds (b) are taken as units. Each pair of points located on opposite sides of the unit circle represents the direction of 12 individuals (first order Rayleigh test, n = 2,235, p < 0.00001, r = 0.21, μ = 179°/359°) (a), or the mean axial vector of one herd (second order Rayleigh test, n = 74, p = 0.00467, r = 0.27, μ = 5°/185°) (b). The inner dashed circles mark the 5 % significance border of the Rayleigh test; the arrows are proportional to the mean vector r, which is a measure of non-random distribution. Arrows exceeding these circles indicate significant axial/bimodal orientation

In the next step, we evaluated herds divided into three categories according to their density. As in the previous case, we evaluated individuals and herds separately. All individuals pooled from low-density herds showed the N–S alignment (p < 0.00001, r = 0.26, μ = 179°/359°) (Fig. 2a) and mean vectors of herds clustered significantly along N–S axis as well (p = 0.004978, r = 0.47, μ = 179°/359°) (Fig. 3a). While individuals from middle-density herds were oriented significantly along N–S direction (p < 0.00001, r = 0.28, μ = 178°/358°) (Fig. 2b) mean vectors of whole herds did not show prevailing orientation (p = 0.124999, r = 0.28, μ = 6°/186°) (Fig. 3b). Evaluation of body axis orientation of individuals from high-density herds were not significantly oriented (r = 0.04, p = 0.445703, μ = 40°/220°) (Fig. 2c), nor did the mean vectors of these herds show significant orientation (r = 0.12, p = 0.710158, μ = 41°/221°) (Fig. 3c).
Fig. 2

N–S orientation of individuals reflects the density of herds. Diagrams of individual orientations revealing the degree of N–S alignment within low density (n = 749, p < 0.00001, r = 0.26, μ = 179°/359°) (a), middle density (n = 853, p < 0.00001, r = 0.28, μ = 178°/358°) (b), and high density (n = 564, p = 0.445703, r = 0.04, μ = 40°/220°) (c), herds. One pair of points represents position of five individuals. The inner dashed circles mark the 5 % significance border of the Rayleigh test; the arrows exceeding these circles indicate significant axial/bimodal orientation

Fig. 3

N–S orientation of herds reflects the density. Diagrams of herds orientations revealing the degree of N–S alignment of low density (n = 24, p = 0.004978, r = 0.47, μ = 179°/359°) (a), middle density (n = 26, p = 0.124999, r = 0.28, μ = 6°/186°) (b), and high density (n = 24, p = 0.710158, r = 0.12, μ = 41°/221°) (c), herds. One pair of points represents mean vector bearing of one herd. The inner dashed circles mark the 5 % significance border of the Rayleigh test; the arrows exceeding these circles indicate significant axial/bimodal orientation

Evaluations of both individuals and herds showed declining degree of N–S alignment correlating with growing density. We next asked the question: what is the density threshold between oriented and non-oriented herds? Therefore, we re-grouped all herds according to their density (Fig. 4) and highlighted herds oriented significantly (having non-random distribution) (Rayleigh test, p < 0.05) and at the same time having mean vector direction within N–S sectors (μ = 0° ± 45°). Correlation analysis showed significant negative correlation (Spearman r = −0.352, n = 74, p < 0.01) between herd density and a number of herds oriented both significantly and within N–S sectors. The border between oriented and non-oriented herds was estimated in the region of densities between 13 and 15 individuals per 1,000 m2 (Fig. 4). Such a density corresponds to average distance between closest animals 6–8 m (calculated from herds 51–53). In the following step, we dealt with directional homogeneity of herds (mean vector length) separately from N–S orientation (mean vectors directions). While correlation between herd density and a number of herds having mean vector directions within N–S sectors (μ = 0° ± 45°) was significant (Spearman r = −0.261, n = 74, p < 0.05) analysis of relation between herd density and number of herds oriented significantly regardless of direction (Rayleigh test, p < 0.05) showed no correlation (Spearman r = −0.064, n = 74, n.s.).
Fig. 4

Correlation between herd density and N–S alignment. Herds are ordered according to their density. Black bars show herds oriented significantly (Rayleigh test, p < 0.05) and at the same time having mean vector direction within N–S sectors, μ = 0° ± 45°). Significant negative correlation (Spearman r = −0.352, n = 74, p < 0.01) was found between herd density and a number of herds oriented significantly and within N–S sectors. Arrow points to the region where border between N–S oriented and non-oriented herds may be estimated. It corresponds to the density of 13–15 individuals per 1,000 m2


The original finding of N–S alignment of cattle on flat pastures by the group of Begall et al. (2008) was challenged by a critical study from the team of Hert et al. (2011). In the negative replication, the phenomenon of N–S alignment itself is presented as artificial and unreplicable. Our work has focused primarily on N–S alignment verification.

In our study, we present results of evaluation of 2,235 cows in 74 herds obtained from satellite pictures. Our results are in line with the original study (Begall et al. 2008), as well as with subsequent reports from this group (Burda et al. 2009; Begall et al. 2011) showing that cows on flat pastures do prefer N–S alignment. Significant clustering of body axes in N–S direction was apparent if both individuals and herds were considered as units of circular statistic evaluation. It should be noted though that not all spotted pastures with herds as well as all herds and individuals fulfilling the given limits (slope, distance from power lines, no clustering around feeder etc., see “Methods”) were chosen for evaluation.

The question whether an individual or a herd is a proper basic unit for statistical direction assessment was discussed by Hert et al. (2011) and Begall et al. (2011) teams. Our results are in line with Begall et al. (2008) and gave statistically stronger direction preference if evaluation of all individuals pooled from all herds took place. However, taking circular statistics rules into account mutually dependent and independent animals should be separated and second-order statistic working with mean vectors of whole herds as units should be used. Considering well-known social interactions among cows within herds such a rule seems to be well founded and we present both ways of statistical evaluation. However, our results show (see below) that collective behavior characteristic for dense herds is in contradiction with N–S alignment. Such a finding would make the mean vector of whole herd problematic an exclusive unit for this kind of studies.

When looking for and evaluating herds we noticed apparent variability in density of herds. According to preliminary observations diverse distances among animals within herd correlated with the diverse degree of orientation of the whole herd. Hence, we wanted to know whether it is a density of the herd which—as yet unexplored factor—may affect the N–S alignment.

Having regrouped herds according to their density (Fig. 4), we learned that declining density is proportional to the number of herds oriented significantly (having non-random distribution) and at the same time oriented along with N–S axis. Significant negative correlation between density and N–S alignment was confirmed by Spearman correlation test. Correlation between herd density and a number of herds oriented within N–S sectors regardless their spatial homogeneity (randomness) was significant as well.

Herds falling into the first third (low-density) show significant N–S alignment whether evaluated as pooled individuals (Fig. 2a) or herds (Fig. 3a). Conversely, herds from the densest third did not involve oriented individuals (Fig. 2c), nor were their mean vectors oriented (Fig. 3c). The middle density region could be considered intermediate where analysis of individuals gave significant N–S clustering (Fig. 2b) while analysis of herds did not (Fig. 3b).

On the scale of density (Fig. 4), the border between N–S oriented and non-oriented herds might be sought in the region of densities between 13 and 15 individuals per 1,000 m2. Such a density corresponds with average distance of 6–8 m between closest animals. We reason that while getting over this distance to the nearest partner the animals switch to different modus of behavior with more prominent tendency to N–S position. Alternatively, such a tendency may be steady and natural but relatively weak and nullified or overlapped by diverse social interactions in the vicinity of other members of a herd. The attention paid to others depending on whether they are higher or lower in hierarchy, the division of limited room within a group or simply getting out of the way when moving, are all competing activities potentially masking the N–S alignment. Since no correlation was found between herd density and number of herds oriented significantly regardless of direction we cannot conclude about the impact of social interactions on prevailing direction of the herds. Our study confirms the N–S alignment of cattle as a repeatable phenomenon. Possible reasons for the negative finding of the Hert et al. (2011) have already been extensively discussed in Begall et al. (2011). Our work points out a new and as yet unnoticed impact of density of herd. For succeeding verification or extension studies of alignment phenomenon, it turns out to be important not to concentrate on dense herds only. They may seem to be the most effective in terms of number of scored animals but the N–S alignment may vanish due to animal interactions. For the same reasons, diffuse and scattered herds typical, e.g. for North America should not be omitted.

Another methodical consequence of our work which is important for studies on cattle alignment is the problem of proper unit of orientation assessment—herd vs. individual. In our work, we faced the dubiousness of definition of a herd as a biologically relevant unit. Strict parameters defining a group of animals as a herd according to, e.g. European criterion are inapplicable for scattered herds typical for vast planes of North America where cattle are grazing in much larger distances. Due to this problematic definition of herd, in the latter case it is not possible to ensure the same size and consequently the same weight for statistical analysis of all herds which may range from tens to hundreds of individuals. As mentioned in “Methods”, we discarded groups having <10 animals. For subsequent studies, however, in the light of newly found negative impact of social interactions within herds on N–S alignment, the most appropriate unit would be an individual inside a scattered herd rather than a whole herd.

Masking of the N–S alignment by other kinds of social behavior in dense herds may explain why this phenomenon was escaping the attention of farmers for such a long time. The animals within the loose and scattered herds practically cannot be seen and compared by an observer from the ground. Consequently, it is not surprising that the prevailing N–S orientation of cattle was not discovered prior to accessibility of satellite imaging techniques.

To date, the fundamental question for the biological meaning of this behavior cannot be answered definitively other than with rather speculative hypotheses. Whether the N–S alignment of cattle is truly a reaction to GMF and not to other directional factors like Sun irradiation was discussed by other workers (Begall et al. 2008, 2011; Burda et al. 2009). Our data do not bring any new insights into this matter, nor has it been the goal of our work. Nevertheless, the magnetic alignment of other species is gradually becoming a well-documented phenomenon (reviewed in Begall et al. 2012). Doubts about magnetoreception being the basis for N–S alignment in cattle will probably be definitely dispelled by experiments using artificial and controlled fields. Our replication part of the work is aimed primarily at discussing the validity of the phenomenon of prevailing N–S orientation of cattle on flat pastures. This fundamental phenomenon has been replicated positively. Moreover, herd density as yet unexplored factor has proven important for alignment studies providing some insight into biological significance of alignment.


Our results are in line with the original findings (Begall et al. 2008) of prevailing N–S orientation of grazing cattle. Moreover, the results of our study suggest that interactions among animals within dense herds inhibit the tendency to align along the N–S axis. Consequently, alignment could be understood as a manifestation of individual and solitary behavior in situations with limited social interactions with other animals. Whether such an alignment behavior is truly based on a magnetic compass sense is an enigma that remains to be tested and solved.



We wish to thank Natraj Krishnan and two anonymous reviewers for critical reading of the manuscript and valuable comments.

Supplementary material

359_2013_827_MOESM1_ESM.pdf (111 kb)
Supplementary material 1 (PDF 111 kb)


  1. Batschelet E (1981) Circular statistic in biology. Academic Press, LondonGoogle Scholar
  2. Begall S, Cerveny J, Neef J, Vojtech O, Burda H (2008) Magnetic alignment in grazing and resting cattle and deer. PNAS 105:13451–13455PubMedCrossRefGoogle Scholar
  3. Begall S, Burda H, Cerveny J, Gerter O, Neef-Weisse J, Nemec P (2011) Further support for the alignment of cattle along magnetic field lines: reply to Hert et al. J Comp Physiol A Neuroethol Sens Neural Behav Physiol 197:1127–1133 discussion 1135–1126PubMedCrossRefGoogle Scholar
  4. Begall S, Malkemper EP, Cerveny J, Nemec P, Burda H (2012) Magnetic alignment in mammals and other animals. Mamm Biol. doi:10.1016/j.mambio.2012.05.005 Google Scholar
  5. Burda H, Marhold S, Westenberger T, Wiltschko W, Wiltschko R (1990) Magnetic compass orientation in the subterranean rodent Cryptomys hottentotus (Bathyergidae, Rodentia). Experientia 46:528–530PubMedCrossRefGoogle Scholar
  6. Burda H, Begall S, Cerveny J, Neef J, Nemec P (2009) Extremely low-frequency electromagnetic fields disrupt magnetic alignment of ruminants. PNAS 106:5708–5713PubMedCrossRefGoogle Scholar
  7. Cadiou H, McNaughton PA (2010) Avian magnetite-based magnetoreception: a physiologist’s perspective. J R Soc Interface 7:S193-S205Google Scholar
  8. Cerveny J, Begall S, Koubek P, Novakova P, Burda H (2011) Directional preference may enhance hunting accuracy in foraging foxes. Biol Lett 7:355–357PubMedCrossRefGoogle Scholar
  9. Deutschlander ME, Freake MJ, Borland SC, Phillips JB, Madden RC, Anderson LE, Wilson BW (2003) Learned magnetic compass orientation by the Siberian hamster, Phodopus sungorus. Anim Behav 65:779–786CrossRefGoogle Scholar
  10. Dommer DH, Gazzolo PJ, Painter MS, Phillips JB (2008) Magnetic compass orientation by larval Drosophila melanogaster. J Insect Physiol 54:719–726PubMedCrossRefGoogle Scholar
  11. Fraser JP (2010) Maps and Compasses. In: Breed MD, Moore J (eds) Encyclopedia of Animal Behavior, vol 2. Academic press, Oxford, pp 375–380CrossRefGoogle Scholar
  12. Gegear RJ, Casselman A, Waddell S, Reppert SM (2008) Cryptochrome mediates light-dependent magnetosensitivity in Drosophila. Nature 454:1014–1018PubMedCrossRefGoogle Scholar
  13. Hart V, Kusta T, Nemec P, Blahova V, Jezek M, Novakova P, Begall S, Cerveny J, Hanzal V, Malkemper EP, Stípek K, Vole C, Burda H (2012) Magnetic alignment in carps: evidence from the Czech Christmas fish market. PLoS ONE 7(12):e51100. doi:10.1371/journal.pone
  14. Hert J, Jelinek L, Pekarek L, Pavlicek A (2011) No alignment of cattle along geomagnetic field lines found. J Comp Physiol A 197:677–682CrossRefGoogle Scholar
  15. Holland RA, Thorup K, Vonhof M, Cochran WW, Wikelski M (2006) Bat orientation using Earth’s magnetic field. Nature 444:653–702CrossRefGoogle Scholar
  16. Hsu C-Y, Ko F-Y, Li C-W, Fann K, Lue J-T (2007) Magnetoreception system in honeybees (Apis mellifera). PLoS ONE 2:e395PubMedCrossRefGoogle Scholar
  17. Johnsen S, Lohmann KJ (2005) The physics and neurobiology of magnetoreception. Nat Rev Neurosci (advanced online publication) 6:703–712CrossRefGoogle Scholar
  18. Kimchi T, Terkel J (2001) Magnetic compass orientation in the blind mole rat Spalax ehrenbergi. J Exp Biol 204:751–758PubMedGoogle Scholar
  19. Kirschvink JL, Kirschvink AK (1991) Is geomagnetic sensitivity real? Replication of the Walker–Bitterman magnetic conditioning experiment in honey bees. Am Zool 31:169–185Google Scholar
  20. Kirschvink JL, Winklhofer M, Walker MM (2010) Biophysics of magnetic orientation: strengthening the interface between theory and experimental design. J R Soc Interface 7:179–191CrossRefGoogle Scholar
  21. Mather JG, Baker RR (1981) Magnetic sense of direction in wood mice for route-based navigation. Nature 291:152–155CrossRefGoogle Scholar
  22. Maeda K, Robinson AJ, Henbest KB, Hogben HJ, Biskup T, Ahmad M, Schleicher E, Weber S, Timmel CR, Hore PJ (2012) Magnetically sensitive light-induced reactions in cryptochrome are consistent with its proposed role as a magnetoreceptor. PNAS 109:4774–4779Google Scholar
  23. Muheim R, Edgar NM, Sloan KA, Phillips JB (2006) Magnetic compass orientation in C57BL/6J mice. Learn Behav 34:366–373PubMedCrossRefGoogle Scholar
  24. Phillips JB, Muheim R, Jorge PE (2010) A behavioral perspective on the biophysics of the light-dependent magnetic compass: a link between directional and spatial perception? J Exp Biol 213:3247–3255PubMedCrossRefGoogle Scholar
  25. Ritz T, Ahmad M, Mouritsen H, Wiltschko R, Wiltschko W (2010) Photoreceptor-based magnetoreception: optimal design of receptor molecules, cells, and neuronal processing. J R Soc Interface 7:S135–S146PubMedCrossRefGoogle Scholar
  26. Takebe A, Furutani T, Wada T, Koinuma M, Kubo Y, Okano K, Okano T (2012) Zebrafish respond to the geomagnetic field by bimodal and group-dependent orientation. Sci Rep 2:727–732PubMedCrossRefGoogle Scholar
  27. Vacha M, Kvicalova M, Puzova M (2010) American cockroach prefers four cardinal geomagnetic compass positions at rest. Behaviour 147:425–440CrossRefGoogle Scholar
  28. Wang Y, Pan Y, Parsons S, Walker MM, Zhang S (2007) Bats respond to polarity of a magnetic field. Proc R Soc B 274:2901–2905PubMedCrossRefGoogle Scholar
  29. Wiltschko W, Wiltschko R (2005) Magnetic orientation and magnetoreception on birds and other animals. J Comp Physiol A 191:69–675CrossRefGoogle Scholar
  30. Wiltschko R, Wiltschko W (2006) Magnetoreception. BioEssays 28:157–168PubMedCrossRefGoogle Scholar
  31. Wiltschko W, Freire R, Munro U, Ritz T, Rogers L, Thalau P, Wiltschko R (2007) The magnetic compass of domestic chickens, Gallus Gallus. J Exp Biol 210:2300–2310PubMedCrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2013

Authors and Affiliations

  1. 1.Department of Animal Physiology and Immunology, Faculty of Science, Institute of Experimental BiologyMasaryk UniversityBrnoCzech Republic

Personalised recommendations