Radio waves zap the biomagnetic compass

nodes of social interactions switch on specific
neural circuits at the expense of others8,9. These
circuits underlie stereotyped behaviours, and
coexist in both males and females, whether
they are sexually experienced or not. But only
under specific conditions are they activated.
This is a remarkable example of the modularity
and versatility of mammalian brains. ■
1. Wu, Z., Autry, A. E., Bergan, J. F., Watabe-Uchida, M.
& Dulac, C. G. et al. Nature 509, 325–330 (2014).
2. Lefèvre, C. F., Sharp, J. A. & Nicholas, K. R. Annu. Rev.
Genom. Hum. Genet. 11, 219–238 (2010).
3. Bridges, R. S., DiBiase, R., Loundes, D. D. & Doherty,
P. C. Science 227, 782–784 (1985).
4. Ferris, C. F., Delville, Y., Bonigut, S. & Miller, M. A.
Brain Res. 832, 1–6 (1999).
5. Pusey, A. E. & Packer, C. in Infanticide and Parental
Care (eds Parmigiani, S. & Vom Saal, F. S.) 277–299
(Harwood, 1994).
6. Perrigo, G., Belvin, L. & Vom Saal, F. S. Chronobiol.
Int. 9, 421–433 (1992).
7. Bruce, H. M. Nature 184, 105 (1959).
8. Kimchi, T., Xu, J. & Dulac, C. Nature 448,
1009–1014 (2007).
9. Ferrero, D. M. et al. Nature 502, 368–371 (2013).
Ivan Rodriguez is in the Department of
Genetics and Evolution, University of Geneva,
CH-1205 Geneva, Switzerland.
e-mail: [email protected]
phones), and the strength is about equivalent
to what a bird in flight might experience 5 kilo­
metres away from a 50-kilowatt AM radio
Two results flag this study as particularly
noteworthy, and puzzling. First, the levels of
radio-frequency radiation that affected the
birds’ orientation are substantially below anything previously thought to be biophysically
plausible, and far below levels recognized as
affecting human health. Second, the authors
detect no trace of a sharply enhanced effect
at the Larmor frequency (the natural period
at which single electrons wobble around the
geomagnetic-field direction), which flatly contradicts experiments on the same species performed using a similar protocol6. This failure
Radio waves zap the
biomagnetic compass
Weak radio waves in the medium-wave band are sufficient to disrupt geomagnetic
orientation in migratory birds, according to a particularly well-controlled study.
But the underlying biophysics remains a puzzle. See Letter p.353
agnetobiology has largely been
viewed as a stamping ground for
charlatans since the followers of
physician Franz Anton Mesmer failed to cure
patients using a ‘magnetized’ tree in the eighteenth century. Numerous discoveries have
begun to change that perspective, although
the road has been rocky. For example, early
studies suggesting that migrating animals use
geomagnetic cues for navigation were plagued
by variability, but it is now clear that many
microorganisms and animals use a magnetic
compass for part of their orientation1.
On the fringe of this fringe field were claims
that radio-frequency radiation could have biological effects at levels too weak to act through
the understood mechanisms of tissue heating or shock, but the experiments usually
lacked proper controls and blinding techniques2–4. Now, however, on page 353 of this
issue, Engels et al.5 demonstrate convincingly
that migrating European robins stop using
their magnetic compasses in the presence
of extra­ordinarily weak, radio-frequency
electro­magnetic ‘noise’.
Using rigorous, double-blinded experiments, the authors found that birds housed in
huts screened from background electromagnetic noise were able to use their magnetic
compass to orient themselves appropriately,
but that their orientation was disrupted following the introduction of electromagnetic
noise ranging from 20 kilohertz to 5 megahertz, at intensities similar to that measured
for background anthropogenic noise in the
environment. To put it into perspective, this
is in the medium-wave band used for AM
radio transmissions (not, for example, mobile
1 µm
~500 µm
~1 µm
2 µm
Figure 1 | Biological ‘magnetomonsters’. Several fossil and extant organisms contain highly magnetic
structures. Examples include: a, Magnetobacter bavaricum, a magnetotactic bacterium with nearly
100 times more magnetite in its cells than more typical types; b, Cryptochiton stelleri, a mollusc whose
magnetite-capped radular teeth will stick strongly to a hand magnet; c, a spearhead-shaped magnetite
particle (false-coloured red), prismatic magnetite rods (purple) and typical magnetite-containing
bacterial organelles (magnetosomes; green); d, a bundle of magnetite rods forming ‘wires’. The structures
shown in c and d were extracted from fossilized clay sediments in New Jersey dating to approximately
56 million years ago11. The origins of the spearhead- and rod-shaped objects are not known, but their size
and morphology suggest that they might have belonged to more-complex organisms. Cellular structures
containing enough electrically conducting magnetite could be sensitive to radio-frequency radiation at
levels shown by Engels et al.5 to disrupt birds’ geomagnetic orientation.
2 9 6 | NAT U R E | VO L 5 0 9 | 1 5 M AY 2 0 1 4
© 2014 Macmillan Publishers Limited. All rights reserved
It is unclear whether infanticide provides
such benefits to virgin male mice, given the
fast weaning times and frequent oestrous
cycles that are characteristic of this species.
But aggressive behaviours spontaneously reemerge in male mice 50 days after mating6:
exactly the length of time it takes for their
progeny to be born and weaned. Furthermore,
exposure of a pregnant female to the scent of
an unfamiliar male mouse is sufficient to cause
termination of pregnancy7. As such, mice may
be the champions of infanticide.
A picture is emerging in which regulatory
to replicate that effect perhaps underscores
previously suggested2,3 flaws in the blinding of
earlier studies.
So what might be going on in these birds?
Several other external stimuli that stop animals from responding to geomagnetic cues
have been identified. Early studies of animal
navigation noted that cues from the Sun or
stars would take precedence over magnetic
cues, leading to the idea that magnetism is the
compass of last resort. It was then noticed that
robins would ignore the magnetic field when
the background intensity was shifted 20–30%
outside the normal value1, and that pigeons
raced poorly during geomagnetic storms.
From an evolutionary perspective, ignoring
geomagnetic cues at such times makes sense,
because anomalies in the background field are
often associated with iron deposits or lightning
strikes. Some animals also stop using their
magnetic compass in the presence of red-only
light, but such light is present only at sunrise
and sunset, when the Sun compass is most
Hence, radio-frequency noise might be
just another cue that tells migrating animals
to ignore their magnetic sense, but the puzzle
is why this might have evolved. Surprisingly,
there is a natural source of the radio-frequency
electromagnetic noise identified as disruptive
by Engels and colleagues — that produced by
solar storms. Coronal mass ejection (CME)
events from the Sun slam plasma into Earth’s
magnetosphere every now and then, causing it to ‘sing’ at frequencies from as low as
around 20 kHz up to the MHz range7, some
of which even leaks through Earth’s normally
radio-opaque ionosphere; the lower end of this
range is remarkably close to that identified by
the authors. These CME events generate the
beautiful polar auroras, disrupt our use of the
medium-wave radio band, and sometimes
perturb the background geomagnetic field
at Earth’s surface enough to disturb animal
All known sensory systems in animals are
based on cells specialized to convert the stimulus of interest into a coded stream of action
potentials that are sent to the brain8. If the
effects of radio-frequency radiation are real,
such cells must exist, but the mystery is in the
biophysics. The lack of an enhanced effect at
the Larmor frequency, and the low levels of
radiation concerned, make it unlikely that a
previously proposed mechanism6 for radiosensing, based on light activation of a cellular
protein called cryptochrome, is involved. But
some magnetic effects on animals (such as that
of a short, sharp magnetic pulse1) function
through biological magnetite (Fe3O4) in tissue
— might this also be the radio-wave detector?
If it is, how could such a detection mechanism have arisen? Early animals that had a
simple compass patterned along the lines of
magneto­tactic bacteria would have needed to
survive geomagnetic excursions or reversals
— periods in which Earth’s magnetic field
weakened — and natural selection would have
favoured individuals with higher cellular volumes of magnetite3,9. When the field recovered,
animals would have been left with cells that
have surprisingly large magnetic moments9
(Fig. 1). Such cells might then have evolved to
serve other functions, such as intensity-based
magnetic navigation systems, increasing
the amount of magnetite further. With large
enough volumes of metallically conductive
magnetite in these cells, direct detection of the
small electric and magnetic vectors of radiofrequency radiation might have emerged, as
Engels and colleagues suggest.
Do the authors’ findings have implications
for humans? It seems that geomagnetic sensitivity dates back to an early ancestor of animals, and it is clearly present in many extant
mammalian species. Human tissues also
contain biological magnetite10. Many people
claim to be bothered by radio transmissions,
and some have even moved to live in radiofrequency ‘quiet zones’ around radio telescopes. Modern-day charlatans will undoubtedly seize on this study as an argument for
banning the use of mobile phones, despite the
different frequency bands involved. However,
if the effect reported by the authors stands
the acid test of reproducibility, we might consider gradually abandoning our use of this
portion of the electromagnetic spectrum and
implementing engineering approaches to
minimize incidental low-frequency noise, to
help migratory birds find their way. ■
Joseph L. Kirschvink is in the Division of
Geological and Planetary Sciences, California
Institute of Technology, Pasadena,
California 91125, USA, and at the EarthLife Science Institute, Tokyo Institute of
Technology, Japan.
e-mail: [email protected]
1. Wiltschko, W. & Wiltschko, R. J. Comp. Physiol. A
191, 675–693 (2005).
2. Kirschvink, J. L. Bioelectromagnetics 13, 401–411
3. Kirschvink, J. L., Winklhofer, M. & Walker, M. M.
J. R. Soc. Interface 7, S179–S191 (2010).
4. Kobayashi, A. K., Kirschvink, J. L. & Nesson, M. H.
Nature 374, 123 (1995).
5. Engels, S. et al. Nature 509, 353–356 (2014).
6. Ritz, T. et al. Biophys. J. 96, 3451–3457 (2009).
7. LaBelle, J. & Treumann, R. A. Space Sci. Rev. 101,
295–440 (2002).
8. Block, S. M. in Sensory Transduction (eds Corey, D. P.
& Roper, S. D.) Ch. 1, 1–17 (Rockefeller Univ. Press,
9. Eder, S. H. K. et al. Proc. Natl Acad. Sci. USA 109,
12022–12027 (2012).
10.Kirschvink, J. L., Kobayashi-Kirschvink, A. &
Woodford, B. J. Proc. Natl Acad. Sci. USA 89,
7683–7687 (1992).
11.Schumann, D. et al. Proc. Natl Acad. Sci. USA 105,
17648–17653 (2008).
This article was published online on 7 May 2014.
Geology and climate
drive diversification
Data from the Galapagos Islands exemplify how geology and climate can interact
to cause episodes of isolation and fusion of the biota across a landscape. Different
scales of such cycles dictate varying mechanisms of species generation.
riting in the Journal of Biogeography,
Ali and Aitchison1 examine geolo­
gical and climatic events over the
past 700,000 years, namely island ontogeny and
shifting sea levels, and their effects on biodiversity in the Galapagos Islands. The authors propose a process that can be considered a general
evolutionary mechanism: that the dynamics
of isolation caused by geological and climatological processes plays a fundamental part
in shaping diversity. Whether these processes
promote or constrain species diversification, however, depends on the spatial (global,
regional or local) and temporal (multimillion,
multi­millennial or multi­decadal) scales and
periodi­city of isolation and coalescence.
Geological events have long been known to
mould and shape biodiversity. A breakthrough
in understanding the underlying mechanisms
came with the recognition that ancient splitting of landmasses resulted in shared diversity.
The concept of vicariance biogeography — the
separation of a group of organisms by a geographical barrier — provided the means for
rigorous hypothesis testing in a hitherto largely
descriptive field. This established that vicariance resulting directly from geological events
can cause diversification, such that geological
history will be clearly reflected in the resulting biotic assemblages. The isolation created
by ancient geological events is fundamental.
Yet, what is given is frequently taken away —
separate land masses can become connected
and biotic assemblages reunited to various
degrees. For example, the Great American
1 5 M AY 2 0 1 4 | VO L 5 0 9 | NAT U R E | 2 9 7
© 2014 Macmillan Publishers Limited. All rights reserved