Brief comm 20/1/00 mx

© 2000 Macmillan Magazines Ltd

---

loidea as a sister-group to the Papilionoidea
and Hesperoidea, it is possible that Vogel’s
organ is a degenerate ‘bat detector’. Our
discovery may help to bring to light the evo-
lutionary origin of this group of butterflies.
Jayne E. Yack*, James H. Fullard†
*Department of Biology and College of Natural
Sciences, Carleton University, 
Ottawa, Ontario K1S 5B6, Canada
†Department of Zoology and Erindale College,
University of Toronto, Mississauga, 
Ontario L5L 1C6, Canada
e-mail: [e-mail]
1. Fullard, J. H. in Comparative Hearing: Insects (eds Hoy, R. R.,
Popper, A. N. & Fay, R. R.) 279–326 (Springer, New York, 1998).
2. Conner, W. E. J. Exp. Biol. 202, 1711–1723 (1999).
3. Silberglied, R. E. in The Biology of Butterflies (eds Vane-Wright,
R. I. & Ackery, P. R.) 207–223 (Academic, London, 1984).
4. Yack, J. E. & Fullard, J. H. Ann. Entomol. Soc. Am. 86, 677–682
(1993).
5. Ribaric, D. & Gogala, M. Acta Entomol. Slov. 4, 5–12 (1996).
6. Scoble, M. J. Bull. Br. Mus. Nat. Hist. (Entomol.) 53, 251–286
(1986).
7. Scoble, M. J. & Aiello, A. J. Nat. Hist. 24, 159–164 (1990).
8. Jacobs, K., Otte, B. & Lakes-Harlan, R. J. Exp. Zool. 283,
270–285 (1999).
9. Hoy, R. R., Nolen, T. & Brodfuehrer, P. J. Exp. Biol. 146,
287–306 (1989).
10. Kristensen, N. P. & Skalski, A. W. in Handbook of Zoology,
Arthropoda: Insecta, Part 35, Vol. 1, Lepidoptera: Moths and
Butterflies (ed. Kristensen, N. P.) 7–25 (de Gruyter, Berlin, 1998).
11. Novacek, M. J. Nature 315, 140–141 (1985).
12. Vogel, R. Z. Wiss. Zool. 100, 210–244 (1912).
13. Otero, L. D. Bol. Entomol. Venez. 5, 128–138 (1990).
decagonal symmetry for the entire cluster
(Fig. 1a) which enables the decagons to
have identical subtiles, as suggested by the
Gummelt coverage model, ensuring that the
overlap rules required by this model are
satisfied. 
However, it is hard to see how a struc-
ture with such extensive broken symmetry
can be energetically stable. Figure 1b shows
a typical Z-contrast image of clusters that
do not have strong broken symmetry. In a
Z-contrast image, the intensity is directly
correlated with the mean-square atomic
number (Z), so that transition metal (TM)
columns are seen with much higher inten-
sity than Al columns.
The superimposition in Fig. 1c shows
that the structure of the decagon proposed
by Steinhardt et al. does not match our Z-
contrast image in significant ways: the four
proposed TM columns causing the broken
symmetry are definitely absent from our
image. In addition, our image reveals the
presence of ten closely spaced TM column
pairs in the outermost ring of the cluster, as
indicated by double blue arrows in Fig. 1b.
These are only single TM columns in the
model of Steinhardt et al. 
Turning to the central ring, our image
clearly shows its underlying ten-fold sym-
metry. It is seen as a ring, with an intensity
varying between that of an Al column (red)
and a TM column (yellow). This is incon-
sistent with Steinhardt et al.’s triangular
arrangement. This intensity pattern shows
that there are ten closely spaced atomic
columns around the central ring with a
composition intermediate between that of
an Al column and a TM column. However,
there are many clusters with broken sym-
metry in the central ring. 
Figure 2 is a typical Z-contrast image of
such a cluster where the intensity in the
central ring shows broken decagonal sym-
metry. The intensity distribution shows that
266
NATURE
|
VOL 403
|
20 JANUARY 1999
|
www.nature.com
principal obtectomeran groups originated
around the late Palaeocene epoch (~60 mil-
lion years ago)
10
, by which time microchi-
ropteran bats had evolved echolocation
11
.
We conclude that predation by bats imposed
a great selection pressure on the evolution of
ears in Lepidoptera.
Butterflies are the largest and most
diverse group of diurnal Lepidoptera, and
the selection pressures generally proposed
for their diurnality include various physio-
logical and ecological factors, but not selec-
tion pressure by bats. Given the significant
impact of bats on other obtectomeran taxa,
we suggest that diurnality in non-hedylid
butterflies was also an anti-bat strategy,
promoted by selection for individuals that
avoided bats by appearing during the day.
The butterfly, in effect, was therefore
‘invented’ by the bat.
Is the earless condition of other (non-
hedylid) butterflies primitive or secondarily
derived? Consider the Vogel’s organ, a
forewing structure of unknown function that
is distributed sporadically with varying
degrees of development among certain
Papilionoidea
12,13
. Our comparative anatomi-
cal studies show that the hedylid ear and
Vogel’s organ are homologous structures.
Given the current placement of the Hedy-