First published online October 5, 2006
Journal of Experimental Biology 209, 4140-4153 (2006)
Published by The Company of Biologists 2006
doi: 10.1242/jeb.02483
The variable colours of the fiddler crab Uca vomeris and their relation to background and predation
Jan M. Hemmi1,2,
Justin Marshall3,
Waltraud Pix2,
Misha Vorobyev1,3 and
Jochen Zeil1,2,*
1 ARC Centre of Excellence in Vision Science, Australian National
University, Canberra ACT 2601, Australia
2 Centre for Visual Sciences, Research School of Biological Sciences,
Australian National University, Canberra ACT 2601, Australia
3 Vision Touch and Hearing Research Centre, School of Biomedical Sciences,
University of Queensland, Brisbane, QLD 4072, Australia
View larger version (73K):
[in a new window]
|
Fig. 1. Set-up for, and analysis of, dummy bird predation experiments. (A)
Experimental set-up. An experimental and a control female with their burrows
about 80 cm apart were separated by a barrier 20 cm high and 1.2–2.4 m
long, which prevented the control crab from seeing the bird dummy. The dummy
consisted of a 3 cm black styrofoam sphere that was moved directly over the
burrow of the test crab every 2–3 min, 20 cm above ground along a track
parallel or perpendicular to the barrier. The dummy was suspended from a
fishing line which ran through three pegs as shown in the diagram, thus
allowing an operator sitting about 4 m away to move it across the area. An
area of about 225–400 cm2 around each of the crabs' burrows
was continuously monitored by video throughout the activity period of the
crabs. (B) Data analysis. Individual frames were extracted from the video
sequences and images were histogram matched within and across days to one
standard image; subsequently, a region of interest was extracted in the blue
channel of the RGB image for pixel values larger than 200 (of 0–255
values in an 8-bit image). The pixels captured in this procedure are shown in
red in the bottom right image. For further details see Materials and
methods.
|
|
View larger version (138K):
[in a new window]
|
Fig. 2. The carapace colours of male (A) and female Uca vomeris (B) from
Bowling Green Bay, Queensland, Australia. Note the individual patterns of blue
and white patches and the fact that individuals first appear to develop
colouration on the ventral part of the posterior carapace, before more dorsal
regions of the carapace become colourful (A).
|
|
View larger version (93K):
[in a new window]
|
Fig. 3. The spectral signatures of U. vomeris carapace colours. (A)
Photographs of male and female crabs, shortly after capture and after spectral
reflectance measurements were taken using a fibre optics spectrograph (Ocean
Optics USB2000) from two lateral and one central location on the blue-white
carapace areas. Reflectance spectra for the same crabs are shown in (B), blue
for males and purple for females.
|
|
View larger version (79K):
[in a new window]
|
Fig. 4. In situ reflectance measurements of brightly coloured crabs
recorded using a spectrographic imager (CASI) (A) and a UV-sensitive camera
(B). Imager and camera viewed a fiddler crab colony from a height of about 1 m
at an angle of 45° with an angular resolution of 0.025°. Example scans
are shown on top for the spectrographic imager and below for the UV-sensitive
camera. (A) Average reflectance spectra with their standard deviation (shaded)
for three female crabs numbered 1–3 in the spectrographic imager scan:
blue: blue-white patches on carapace; black: mudflat background. Inset
pictures show, for crab 1, how areas of interest were selected (left: original
image, right: regions of interest with blue carapace marked blue and mud flat
background marked black). For details see Materials and methods. (B) UV images
together with their 2D density functions.
|
|
View larger version (91K):
[in a new window]
|
Fig. 5. Rapid, stress-related changes of carapace colours in Uca vomeris.
The reflectance measurements in (A) and (B) were made about 1 min before the
colour photographs were taken at different times after capturing a female and
a male crab. Measurements were taken using a fibre optics spectrograph (Ocean
Optics USB2000) from three points on the posterior carapace. The brilliant
blue colours change to a dull, dark blue within 15–20 min of
capture.
|
|
View larger version (52K):
[in a new window]
|
Fig. 6. (A) The average spectral reflectance of crabs in three neighbouring
colonies of Uca vomeris and the presence (B) and behaviour (C) of
predatory birds at the same three sites. At three sites, called `North',
`Central' and `South' (see inset map on top), we measured the spectral
reflectance of carapace colours for 10–21 crabs at each site and
subsequently recorded bird activity over 3 consecutive days. (A) The
reflectance spectra of crabs from the three sites [means (red) ± s.e.m.
(dark blue)]. The black bars below the right panel indicate the range of the
spectrum over which the respective reflectance distributions do not overlap at
the level of means ± 2 s.e.m. (light blue areas). Three measurements
per crab were averaged, before averaging across crabs. (B) The total number of
birds observed at the three sites over 3 consecutive days. (C) The total
number of high-flying (High), low-flying (Low), and ground-hunting birds
(Ground) at the three sites, summed over 3 days of observation. Compared to
the `North' and `Central' colonies, the `South' colony, with the most
colourful crabs, experiences very few predatory birds (right panels).
|
|
View larger version (31K):
[in a new window]
|
Fig. 7. Modelling luminance and chromatic contrast of crab colours for crab and
bird visual systems. (A) The spectral sensitivities used for modelling (for
references see Materials and methods). (B) The chromatic distance of carapace
colours to the mudflat background in the three focal colonies North, Central
and South (see Fig. 6), as box
and whisker plots showing the median, the 75% quartiles, the maximum and the
minimum values. (C) The chromatic distance to the white point (top) and
luminance signals (centre and bottom) of carapace colours and mud for crab and
bird photoreceptors. For details, see Materials and methods.
|
|
View larger version (84K):
[in a new window]
|
Fig. 8. In situ colour changes of an individual crab exposed to several
days of dummy bird predation. (A–C) Pairs of video images of a female
U. vomeris at the beginning (left), in the middle (centre) and at the
end (right) of her entire activity period. (A) First day of dummy bird
treatment. (B) Third day of dummy bird treatment. (C) Day 4, on which
treatment was discontinued. Note colour changes throughout activity period and
as a consequence of dummy bird predation.
|
|
View larger version (30K):
[in a new window]
|
Fig. 9. Quantifying the colour changes of U. vomeris females in
situ as a consequence of alternating exposure to dummy bird predation.
(A,B) Two female U. vomeris were selected for each experiment (see
Materials and methods and text for details). Circles are 30 min averages of
numbers of pixels on the carapace of crabs with values higher than 200 (red,
test animals subjected to dummy predation; blue, controls). Lines are
regression lines through the data points. Stars mark the times of burrow
closure. Note that crabs exposed to dummy predation tend become dull at a
faster rate than control animals and can lose their bright colouration all
together over a matter of days. See text for a full description of
experiments.
|
|
View larger version (24K):
[in a new window]
|
Fig. 10. Colour changes of an individual U. vomeris female subjected to 3
days of dummy bird predation. Small symbols connected by thin lines show
individual measurements (open circles, number of pixel values between 150 and
200; filled circles, number of pixel values between 201 and 255). Large red
symbols connected by thick lines show 30 min means of numbers of pixels with
values between 150 and 200 (open symbols) and between 201 and 255 (filled
symbols). Inset pictures show the appearance of the crab at the given times of
day and the carapace areas exhibiting pixel values between 150 and 200 (green)
and between 201 and 255 (red). Arrowheads above the x-axis mark the
times when the female closed her burrow.
|
|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati 
Twitter What's this?
© The Company of Biologists Ltd 2006
