ArticlePDF Available

Abstract and Figures

In ectotherms, the colour of an individual's cuticle may have important thermoregulatory and protective consequences. In cool environments, ectotherms should be darker, to maximize heat gain, and larger, to minimize heat loss. Dark colours should also predominate under high UV-B conditions because melanin offers protection. We test these predictions in ants (Hymenoptera: Formicidae) across space and through time based on a new, spatially and temporally explicit, global-scale combination of assemblage-level and environmental data.
Content may be subject to copyright.
Ant assemblages have darker and larger
members in cold environments
Tom R. Bishop
1,2
*, Mark P. Robertson
2
, Heloise Gibb
3
,
Berndt J. van Rensburg
4,5
, Brigitte Braschler
6,7
, Steven L. Chown
8
,
Stefan H. Foord
9
, Thinandavha C. Munyai
9,10
, Iona Okey
3
,
Pfarelo G. Tshivhandekano
2
, Victoria Werenkraut
11
and
Catherine L. Parr
1,12
1
Department of Earth Ocean and Ecological
Sciences, University of Liverpool, Liverpool,
L69 3GP, UK,
2
Centre for Invasion Biology,
Department of Zoology and Entomology,
University of Pretoria, Pretoria, 0002, South
Africa,
3
Department of Zoology, La Trobe
University, Melbourne, Vic. 3068, Australia,
4
School of Biological Sciences, University of
Queensland, St Lucia, Qld 4072, Australia,
5
Department of Zoology, Centre for Invasion
Biology, University of Johannesburg,
Auckland Park, Johannesburg, 2006, South
Africa,
6
Department of Botany and Zoology,
Centre for Invasion Biology, Stellenbosch
University, Matieland, South Africa,
7
Section
of Conservation Biology, Department of
Environmental Sciences, University of Basel,
Basel, 4056, Switzerland,
8
School of
Biological Sciences, Monash University,
Melbourne, Vic. 3800, Australia,
9
Department of Zoology, Centre for Invasion
Biology, University of Venda, Thohoyandou,
0950, South Africa,
10
School of Life Sciences
College of Agriculture Engineering and
Science, University of KwaZulu-Natal,
Pietermaritzburg, 3209, South Africa,
11
Laboratorio Ecotono, Centro Regional
Universitario Bariloche, Universidad
Nacional del Comahue, INIBIOMA-
CONICET, Bariloche, Rio Negro 8400,
Argentina,
12
School of Animal, Plant and
Environmental Sciences, University of the
Witwatersrand, Wits, 2050, South Africa
*Correspondence: Tom R. Bishop,
Department of Earth, Ocean and Ecological
Sciences, University of Liverpool, Liverpool
L69 3GP, UK.
E-mail: thomasrhys.bishop@gmail.com
This is an open access article under the terms
of the Creative Commons Attribution License,
which permits use, distribution and
reproduction in any medium, provided the
original work is properly cited.
ABSTRACT
Aim In ectotherms, the colour of an individual’s cuticle may have important
thermoregulatory and protective consequences. In cool environments,
ectotherms should be darker, to maximize heat gain, and larger, to minimize
heat loss. Dark colours should also predominate under high UV-B conditions
because melanin offers protection. We test these predictions in ants
(Hymenoptera: Formicidae) across space and through time based on a new,
spatially and temporally explicit, global-scale combination of assemblage-level
and environmental data.
Location Africa, Australia and South America.
Methods We sampled ant assemblages (n5274) along 14 elevational transects
on three continents. Individual assemblages ranged from 250 to 3000 m a.s.l.
(minimum to maximum range in summer temperature of 0.5–35 8C). We used
mixed-effects models to explain variation in assemblage cuticle lightness.
Explanatory variables were average assemblage body size, temperature and UV-
B irradiation. Annual temporal changes in lightness were examined for a subset
of the data.
Results Assemblages with large average body sizes were darker in colour than
those with small body sizes. Assemblages became lighter in colour with
increasing temperature, but darkened again at the highest temperatures when
there were high levels of UV-B. Through time, temperature and body size
explained variation in lightness. Both the spatial and temporal models
explained c. 50% of the variation in lightness.
Main conclusions Our results are consistent with the thermal melanism
hypothesis, and demonstrate the importance of considering body size and UV-
B radiation exposure in explaining the colour of insect cuticle. Crucially, this
finding is at the assemblage level. Consequently, the relative abundances and
identities of ant species that are present in an assemblage can change in
accordance with environmental conditions over elevation, latitude and
relatively short time spans. These findings suggest that there are important
constraints on how ectotherm assemblages may be able to respond to rapidly
changing environmental conditions.
Keywords
Assemblage structure, colour, elevation, latitude, lightness, temperature,
thermal melanism, thermoregulation.
V
C2016 The Authors. Global Ecology and Biogeography
published by John Wiley & Sons Ltd
DOI: 10.1111/geb.12516
http://wileyonlinelibrary.com/journal/geb 1
Global Ecology and Biogeography, (Global Ecology and Biogeography) (2016)
RESEARCH
PAPER
INTRODUCTION
Living organisms displays a huge diversity of colour, which
has captured the imagination of biologists for centuries. Ani-
mals use different patterns and hues of colour to disguise or
advertise themselves (Ruxton et al., 2004), attract mates
(Andersson, 1994) or thermoregulate (Clusella-Trullas et al.,
2007). For ectotherms, which make up the majority of ani-
mal species, thermoregulation is of great importance. Ecto-
therm metabolism is largely dependent on ambient
temperature and, because of this, their performance and geo-
graphical distribution is strongly influenced by temperature
gradients (Buckley et al., 2012). Consequently, the ability to
thermoregulate in response to these gradients is critical for
ectotherm survival (Heinrich, 1996).
Ectotherm cuticle colour affects thermoregulation through
its reflectivity. A dark-coloured or unreflective individual,
with high levels of melanin, will heat up faster and achieve
higher temperature excesses than a light-coloured individual
of the same size and shape (Willmer & Unwin, 1981). The
thermal melanism hypothesis is based on this basic biophysi-
cal principle, predicting that darker individuals should pre-
dominate in low-temperature environments because they will
have a higher fitness (Clusella-Trullas et al., 2007). Higher fit-
ness is a consequence of the longer periods of activity avail-
able to darker individuals as they are able to warm up and
achieve operating temperatures more rapidly (Bogert, 1949;
Clusella-Trullas et al., 2007). Indeed clines in melanism along
temperature gradients have been reported in several taxa (e.g.
butterflies, dragonflies, reptiles, springtails), across a range of
spatial scales and at both intra- and interspecific levels
(Rapoport, 1969; Zeuss et al., 2014). Whilst these effects are
a direct result of melanin pigmentation, the melanogenesis
pathway itself may also influence cold resistance pleiotropi-
cally through its effects on energy homeostasis and metabolic
rates (Ducrest et al., 2008).
A key assumption of the thermal melanism hypothesis is
that individuals have the same size and shape, yet in reality
body size and shape vary greatly within and between species.
This is important, as body size is a critical factor in deter-
mining ectotherm heat budgets. Larger bodies gain and lose
heat more slowly than smaller bodies, but also reach higher
temperature excesses (Stevenson, 1985). This size effect
underpins wide-ranging biogeographical predictions such as
Bergmann’s rule, which states that organisms should be larger
in cold environments (Chown & Gaston, 2010).
The effects of colour and body size on ectotherm thermo-
regulation are expected to interact. Being large in a cold
environment may be advantageous in terms of heat conserva-
tion, but it also means that the animal in question will heat
up relatively slowly. This inverse body mass–heating relation-
ship has been used as an explanation for the apparent lack of
support for Bergmann’s rule in ectotherms (Pincheira-
Donoso, 2010). Melanism increases the rate at which heat is
gained, so may provide a mechanism by which ectotherms
could overcome the limitations of a large body size to
operate more effectively in a cold environment (Clusella-
Trullas et al., 2007). This melanism–body size interaction is
predicted from both theory and experiments (Stevenson,
1985; Shine & Kearney, 2001) and has been shown to operate
across large geographical scales in ectotherms (Schweiger &
Beierkuhnlein, 2015). We therefore expect both body size and
ambient temperature to explain variation in ectotherm colo-
ration – darker forms should be larger and occur more fre-
quently in cold environments.
In addition to these thermoregulatory effects, colour, and
specifically melanin, has long been linked with a protective
role against harmful ultraviolet-B (UV-B) radiation. UV-B
can cause a range of deleterious direct effects in ectotherms.
These include genetic and embryonic damage, and indirect
effects through changes in host plant morphology and bio-
chemistry (Hodkinson, 2005; Beckmann et al., 2014;
Williamson et al., 2014). Both experiments (Wang et al.,
2008) and correlative studies (Bastide et al., 2014) have pro-
vided evidence that melanistic individuals or species can be
favoured under conditions of high UV-B. Gloger’s rule (Gas-
ton et al., 2008), that endotherms should be darker at low
latitudes, suggests that pigmentation provides protection
against a range of factors including UV-B irradiance. Patterns
in accordance with Gloger’s rule and the influence of UV-B
have been observed in a number of endotherms (Caro, 2005)
and, more recently, in plants (Koski & Ashman, 2015).
The biophysical principles underlying how temperature,
body size and UV-B radiation may affect ectotherm colour
are understood and accepted at the level of the individual or
the species (e.g. Kingsolver, 1995; Ellers & Boggs, 2004). It is
unknown, however, to what extent these effects scale to the
assemblage level and how important they are at broad spatial
and temporal scales. Understanding assemblage-level varia-
tion in colour is important as it can reveal how traits influ-
ence the performance of species in different environments. In
addition, assemblage analyses can generalize across the indi-
vidualistic responses of each species (Millien et al., 2006).
Assemblage-level variation represents changes in the relative
abundances of different species – this reflects which trait val-
ues appear to be successful under a given set of environmen-
tal conditions. In the search for general rules in ecology,
rising above the contingencies of extreme behaviours, physi-
ologies or morphologies of individual species is crucial
(Chown & Gaston, 2015).
Here, we test if temperature, body size and UV-B can
explain variation in cuticle colour of ant (Hymenoptera: For-
micidae) assemblages – specifically, how light or dark the col-
our is. Ants are a diverse, numerically dominant and
ecologically important group of insects (H
olldobler &
Wilson, 1990) with a wide range of body colours (e.g. http://
www.antweb.org). We sampled ant assemblages across repli-
cated elevational gradients on three continents and over mul-
tiple years. This design is novel and powerful, for two
reasons. First, the combined use of assemblage data,
T. R. Bishop et al.
2Global Ecology and Biogeography,V
C2016 The Authors. Global Ecology and Biogeography
published by John Wiley & Sons Ltd
elevational gradients and continental variation provides
broad-ranging yet fine-scale insight across a huge range of
environmental conditions and geographies. This combination
of fine grain and large extent is rarely achieved (Beck et al.,
2012). Second, our use of time-series data provides greater
power to assign mechanistic links between cuticle lightness,
temperature, body size and UV-B than spatial data alone
would.
If cuticle lightness has a thermoregulatory and protective
role then we would expect that average cuticle lightness will
be: (1) positively related to temperature, (2) negatively
related to average body size, and (3) negatively related to
UV-B radiation. We test all three predictions across space at
a global scale, but only the first two through time.
METHODS
Ant assemblage data
Ant assemblage data were compiled from 14 elevational
transects within eight mountain ranges and across three con-
tinents (Table 1). Ant assemblages were sampled using pitfall
traps in almost exactly the same way across all locations. In
South Africa and Lesotho, pitfall traps were arranged into a
10 m 340 m grid. Four grids were placed in each eleva-
tional band with grids being separated by at least 300 m.
Traps were 55 mm in diameter and used a 50% ethylene gly-
col or propylene glycol solution to preserve trapped speci-
mens (Botes et al., 2006; Munyai & Foord, 2012; Bishop
et al., 2014). Sampling grids in Australia were of the same
dimensions, but those within the same elevation were sepa-
rated by at least 100 m. In Argentina, a sampling grid con-
sisted of nine pitfall traps arranged in a 10 m 310 m grid,
with each trap separated from the next by 5 m. A single grid
was used at each elevation. Traps had a diameter of 90 mm
and used a 40% propylene glycol solution to preserve speci-
mens (Werenkraut et al., 2015). Specimens were transferred
into 70–80% ethanol in the laboratory and identified to mor-
phospecies or species level, where possible. Hereafter, all
morphospecies and species are collectively referred to as
species.
All transects were sampled during the austral summer
(November–May). Each transect was sampled during a single
season, except those in the Maloti-Drakensberg, Cederberg
and Soutpansberg regions of South Africa. These transects
were been sampled biannually in two seasons for a number
of years. These long-term data are only used in the analysis
of temporal patterns (see below). For the analysis of spatial
patterns, only a single summer sampling period was used.
For the Maloti-Drakensberg, Cederberg and Soutpansberg a
single year was randomly chosen for the analysis of spatial
patterns. The Argentinian transects were also sampled over 2
years but only data from 2006 are used here (Werenkraut
et al., 2015). Both years showed the same pattern.
In this study, a sampling grid is considered to be an inde-
pendent assemblage of ants. We did not pool replicate
assemblages within elevational bands. Apart from testing for
phylogenetic signal at the genus level, all analyses are per-
formed at the assemblage level. Two hundred and seventy-
four assemblages were available for the main spatial analysis
after some assemblages had been removed because they did
not contain any ants, or environmental data could not be
gathered for them.
Lightness data
The colour of each ant species was classified categorically by
eye using a predetermined set of colours (see Appendix S1 in
the Supporting Information). This method allows for a sim-
ple and standardized assessment of colour without the need
Table 1 Details on the geographical and elevational characteristics of the transects used in this study
Continent Mountain range Transect
Lowest
point
(m a.s.l.)
Highest
point
(m a.s.l.)
No. of
elevations
Assemblages
per elevation
Species
richness References
Africa Maloti-Drakensberg Sani Pass 900 3000 8 4 92 Bishop et al. (2014, 2015)
Soutpansberg North Aspect 800 1700 5 4 129 Munyai & Foord (2012, 2015)
South Aspect 900 1600 5 4–8
Cederberg East Aspect 500 1800 6 4 94 Botes et al. (2006)
West Aspect 250 1900 10 4
Mariepskop Mariepskop 700 1900 5 4 92 Tshivhandekano &
Robertson (unpublished)
Australia Snowy Mountains Back Perisher 400 2000 9 4 109 Gibb et al. (unpublished)
Ben Lomond Plateau,
Tasmania
Stack’s Bluff 400 1400 6 1–3 12 Gibb et al. (unpublished)
MacDonnell Ranges Mt Zeil 600 1400 5 4 49 Gibb et al. (unpublished)
South America Andes, north-west
Patagonia
Bayo 900 1700 9 1 15 Werenkraut et al. (2015)
Chall-Huaco 900 2000 12 1
La Mona 800 1800 11 1
Lopez 800 1800 10 1
Pelado 800 1800 8 1
Gradients in ant colour and body size
Global Ecology and Biogeography, V
C2016 The Authors. Global Ecology and Biogeography
published by John Wiley & Sons Ltd 3
for specialist imaging equipment. The colour of the head,
mesosoma and gaster for six individuals of every species in
the dataset was recorded. We focused only on the colour of
the cuticle and ignored any colouration offered by hairs. The
most common colour across all body parts and individuals
was assigned as the dominant colour for each species. Each
categorical colour was associated with a set of RGB (red,
blue and green) values which were extracted from the origi-
nal colour wheels using the image editing software paint.NET
(v.4.0.3). RGB values were converted into HSV (hue, satura-
tion and value) format using the ‘rgb2hsv’ function in R.
The HSV model is a common cylindrical-coordinate repre-
sentation of colour where hue describes the dominant wave-
length, saturation indicates the amount of hue present in the
colour and the value sets the amount of light in the colour.
Only lightness (v, or value, in HSV) is analysed here. A
standardized set of 71 photographs from AntWeb (http://ant-
web.org/) was used to assess observer error. Error was low
(Appendix S1), with the standard error of lightness values
estimated from different observers on the same photograph
averaging c. 0.04. The five observers in this study tended to
assign the same lightness value to the same image.
Body size data
The body size of each species was measured as Weber’s
length. This is the distance between the anterodorsal margin
of the pronotum and the posteroventral margin of the pro-
podeum (Brown, 1953). Weber’s length was measured to the
nearest 0.01 mm using ocular micrometers attached to ster-
eomicroscopes. The highest level of magnification that
allowed the entire mesosoma of the specimen to be fitted
under the range of the ocular micrometer was used. Only
minor workers were measured. Six specimens for each species
were measured where possible. Physical specimens were not
available for eight species from the Cederberg transects. For
these species Weber’s length was measured using high-
resolution images from AntWeb (http://www.antweb.org) and
from existing taxonomic publications (Mbanyana &
Robertson, 2008) using the tpsDig2 morphometric software
(http://life.bio.sunysb.edu/morph).
Weber’s length was not available for the ant species from
the MacDonnell Ranges. Instead, it was estimated for these
species using the relationship between head width, head
length and Weber’s length. All three of these traits were avail-
able for the Australian Snowy Mountains and Tasmanian
ants. Only head width and head length were available for the
MacDonnell Ranges ants. Multivariate imputation by chained
equations (MICE; Buuren & Groothuis-Oudshoorn, 2011)
was performed to estimate the missing Weber’s length for
these species (Appendix S2).
Temperature data
Global environmental data
Estimates of air temperature for all of the assemblages from
January to March (the peak of the austral summer) were
extracted from the WorldClim dataset at 30 arcsec resolution
(Hijmans et al., 2005). Levels of UV-B irradiance for all
assemblages were extracted from the glUV dataset
(Beckmann et al., 2014). Mean UV-B irradiances were calcu-
lated using data from January to March.
Data loggers
At all transects in Argentina and at two ranges in southern
Africa (Maloti-Drakensberg and Soutpansberg) data loggers
were used to record daily temperature. In Argentina, a single
HOBO H8 logger (Onset Computer Corporation, MA, USA)
was placed at ground level in the centre of each replicate
block during the sampling months (Werenkraut et al., 2015).
In the two southern African sites Thermocron iButtons
(Semiconductor Corporation, Dallas/Maxim, TX, USA) were
buried 10 mm below ground level at two replicate blocks (of
a possible four) in each elevational band (Munyai & Foord,
2012; Bishop et al., 2014). All temperature data were
inspected for cases where the data loggers had been exposed
to direct sunlight or had clearly malfunctioned. The mean
temperature for each replicate in the sampling month was
calculated. These data logger temperatures were used to vali-
date the temperature estimates from microclim (Kearney
et al., 2014). Furthermore, the data from southern Africa
were used to investigate temporal trends (see below).
Statistical methods
All data manipulation and analyses took place in the R statis-
tical environment (R Core Team, 2014).
Phylogenetic signal
A genus-level, time-calibrated ant phylogeny derived from
Moreau & Bell (2013) was used to estimate the phylogenetic
signal of lightness and body size using Pagel’s k(Pagel, 1999)
and Blomberg’s K(Blomberg et al., 2003). Lightness and
body size traits were averaged at the genus level to test for
signal. A likelihood ratio test was used to assess if there was
a significant departure of these statistics from zero (no phy-
logenetic signal). This was done using the ‘phytools’ package
in R (Revell, 2012). In this study 77.4% of the genera were
present on the phylogeny. Genera missing from the phylog-
eny were omitted from this analysis.
Assemblage-level lightness and body size
Assemblage weighted means (AWM) of lightness and body
size were calculated for each assemblage (n 5274) according
to:
AWM5X
S
i51
pixi
where Sis the number of species in an assemblage, p
i
is the
proportional abundance of each species and x
i
is the trait
value (lightness or body size) of each species.
T. R. Bishop et al.
4Global Ecology and Biogeography,V
C2016 The Authors. Global Ecology and Biogeography
published by John Wiley & Sons Ltd
Data loggers versus WorldClim
The relationship between the mean temperatures collected
through the data loggers and those extracted from World-
Clim was investigated using Type II major axis regression.
This was done with the ‘lmodel2’ package in R (Legendre,
2008). If the 95% confidence intervals of the intercept and
slope encompassed zero and one, respectively, this would
indicate that the WorldClim temperature data accurately
matched that from the data loggers. The significance of the
correlation coefficient was assessed using 999 permutations.
There was a strong correlation between the temperature
values obtained from the data loggers and those extracted
from WorldClim (r50.94, P<0.001; Appendix S4). The
intercept did not differ from zero (95% CI intercept 522.69,
0.03) while the slope differed from one, if only slightly (95%
CI slope 51.11, 1.13). Thus WorldClim temperatures slightly
underestimated the data logger temperatures.
Spatial patterns
Linear mixed models (LMMs) were used to assess how much
variation in assemblage-weighted lightness could be explained
by WorldClim estimates of temperature, the amount of UV-B
radiation and assemblage-weighted mean body size. Model-
ling was done using the ‘lme4’ package in R (Bates et al.,
2014). A term for the temperature–UV-B interaction was also
fitted. As temperature correlates positively with UV-B in our
dataset (r50.81, P<0.001), UV-B was regressed on temper-
ature and the residuals of this relationship were used as the
UV-B variable. All explanatory variables were scaled and
standardized to allow greater interpretability of the regression
coefficients (Schielzeth, 2010). Explanatory variables were
coded as second-order orthogonal polynomials to detect cur-
vature in the relationships between them and assemblage-
weighted lightness. A nested random effects structure of tran-
sect within mountain range within continent was used to
account for the geographical configuration of the study sites.
The response variable of assemblage-weighted lightness was
logit transformed to meet Gaussian assumptions. An infor-
mation theoretic approach was used to assess models with
different combinations of the explanatory variables. Bias cor-
rected Akaike information criterion (AIC
c
) values were used
to compare models. Marginal (due to fixed effects only) and
conditional (due to fixed effects and random effects) R
2
val-
ues were calculated for each model (Barto
n, 2013; Nakagawa
& Schielzeth, 2013). Type III tests using Wald X
2
statistics
were used to assess the significance of the predictors in the
best model. Each of the 274 observations in this analysis was
an independent assemblage of ants.
Common and rare species
Two further spatial analyses took place to disentangle which
species were driving the spatial patterns. For each assemblage,
common species were defined as those making up 90% of
the individuals. The remainder were classed as rare species.
We chose this rule to reflect the extremes of the common–
rare spectrum. Assemblage-weighted lightness and body size
were then recalculated using either only the common species
or only the rare species in each assemblage. Modelling of the
modified assemblage-weighted lightness (and modified
assemblage-weighted body size) took place separately for the
common and the rare species, as described above for the
complete spatial analysis.
Temporal patterns
The Maloti-Drakensberg and Soutpansberg ant assemblages
and temperature data are available for multiple years (7 and
5, respectively). A LMM was used to relate average lightness
to average temperature and body size for each assemblage
across all years. Modelling took place as described for the
spatial analysis but the random effects structure was modified
to take into account temporal pseudoreplication: sampling
grid was nested within transect within mountain range. This
model allows us to detect whether the lightness values of
each assemblage covary according to temporal changes in
temperature and body size. There were 206 observations in
this analysis representing 41 different replicate assemblages
sampled over a number of years (Maloti-Drakensberg519
assemblages over 7 years, Soutpansberg 522 assemblages
over 5 years). There were 243 space/time samples available
but 37 caught no ants, leading to 206 useable observations.
RESULTS
Across all transects 592 ant species were collected (Table 1).
These species spanned the full range of possible lightness val-
ues (0–1). Weber’s length varied from 0.25 to 6.48 mm.
Assemblage-weighted lightness ranged from 0 to 0.9 whilst
assemblage-weighted body size ranged from 0.62 to
2.88 mm.
Phylogenetic signal
Lightness was not significantly conserved across the phylog-
eny – closely related genera do not resemble each other more
so than would be expected by chance (Pagel’s k50.32,
P50.06; Blomberg’s K50.59, P50.13). Body size was con-
served, however (Pagel’s k50.81, P50.001; Blomberg’s
K50.86, P50.002). This signal was due to genera in the
Ponerinae subfamily tending to be larger than those in other
subfamilies (Appendix S3). We do not consider this to con-
found the analyses because proportional representation of
Ponerinae in the sampled assemblages does not correlate
strongly with their average body sizes (r520.003, P50.96).
A strong correlation between the proportions of an assem-
blage that are ponerines and average body size would have
indicated that this phylogenetic signal was influencing the
results.
Spatial patterns
The best spatial model was also the most complicated. It
contained the main effects of temperature, residual UV-B,
body size and also included an interaction between
Gradients in ant colour and body size
Global Ecology and Biogeography, V
C2016 The Authors. Global Ecology and Biogeography
published by John Wiley & Sons Ltd 5
temperature and UV-B (Table 2). All variables apart from the
main effect of residual UV-B radiation were significant according
to Type III Wald X
2
tests (Table 3). Assemblage-weighted lightness
declined with increasing assemblage-weighted body size (Fig. 1a).
At low levels of residual UV-B, assemblage-weighted lightness
increased with increasing temperature. At high levels of residual
UV-B the relationship between lightness and temperature was
unimodal – at higher temperatures lightness declined (Fig. 1b).
Species richness did not influence these results given the small
amount of variation in assemblage lightness that species richness
is able to explain (R2
m50.02, R2
c50.38; Appendix S5). The same
results were found when using microclim (Kearney et al., 2014)
temperature data rather than WorldClim data (Appendix S6).
Common and rare species
The best model for common species was exactly the same as
the overall spatial model (which used all species) and also
explained a similar amount of variance (R2
m50.47,
R2
c50.69; Appendix S7). For the rare species, the best model
contained assemblage body size and residual UV-B. Lightness
declined with increasing average body size and formed a U-
shaped relationship with residual UV-B. This model did not
explain much variation (R2
m50.15, R2
c50.47; Appendix S7).
Temporal patterns
The best temporal model included both mean temperature
and body size (Table 2). Lightness showed a negative rela-
tionship with body size (Fig. 2a) and a positive relationship
with data logger-derived temperature (Fig. 2b). Both body
size and temperature were significant according to Type III
Wald X
2
tests (Table 3).
DISCUSSION
Our study shows that broad geographical patterns of cuticle
colour in ants are consistent with a role in thermoregulation
and a UV-B protection, as predicted by experiment and
theory (Stevenson, 1985; Shine & Kearney, 2001; Wang et al.,
Table 2 Comparative and summary statistics for linear mixed models explaining variation in ant assemblage colour across space or
through time.
Model d.f. LL AIC
c
DAIC
c
wAIC
c
R2
mR2
c
Spatial
(BS 1BS
2
)1(T 1T
2
)3(UV 1UV
2
)15 2301.59 635.03 0.00 1.00 0.48 0.62
(BS 1BS
2
)1(T 1T
2
)1(UV 1UV
2
)112315.07 653.14 18.11 0.00 0.38 0.56
(T 1T
2
)3(UV 1UV
2
)132313.37 654.14 19.11 0.00 0.41 0.59
(BS 1BS
2
)1(T 1T
2
)92321.63 661.94 26.90 0.00 0.43 0.61
(BS 1BS
2
)1(UV 1UV
2
)92322.69 664.05 29.02 0.00 0.21 0.62
(T 1T
2
)1(UV 1UV
2
)92325.24 669.17 34.14 0.00 0.28 0.52
(T 1T
2
)72328.97 672.36 37.33 0.00 0.36 0.59
(UV 1UV
2
)72335.04 684.50 49.47 0.00 0.14 0.62
(BS 1BS
2
)72347.08 708.58 73.55 0.00 0.07 0.41
152356.71 723.64 88.61 0.00 0.00 0.44
Temporal
(BS 1BS
2
)1(T 1T
2
)102112.24 245.60 0.00 0.96 0.49 0.74
(BS 1BS
2
)82117.65 252.03 6.43 0.04 0.36 0.69
(T 1T
2
)82145.36 307.46 61.85 0.00 0.11 0.59
162149.12 310.66 65.06 0.00 0.00 0.61
Predictors were all second-order orthogonal polynomials and included average body size (BS 1BS
2
), average summer temperature (T 1T
2
) and
average residual UV-B radiation (UV 1UV
2
). The temperature variables were derived from WorldClim for the spatial models and from data log-
gers in the temporal models. Listed are the degrees of freedom (d.f.), maximum log-likelihood (LL), Akaike’s bias corrected information criterion
(AIC
c
) and its change relative to the top ranked model (DAIC
c
), the model probabilities (wAIC
c
) and the marginal and conditional R
2
s. Marginal
R
2
(R2
m) is the amount of variation explained by the fixed effects, conditional R
2
(R2
c) is that explained by the fixed and random effects.
Table 3 Test statistics (v
2
), degrees of freedom (d.f.) and
P-valuesfromTypeIIIWaldtestson the best spatial and temporal
models (top ranked spatial and temporal models from Table 2).
Spatial v
2
d.f. P
T1T
2
29.77 2 <0.001
UV 1UV
2
3.01 2 0.22
BS 1BS
2
24.81 2 <0.001
(T 1T
2
)3(UV 1UV
2
) 29.43 4 <0.001
Temporal
T1T
2
16.48 2 <0.001
BS 1BS
2
87.85 2 <0.001
Explanatory variables were second-order orthogonal polynomials and
included average body size (BS 1BS
2
), average summer temperature
(T 1T
2
) and average residual UV-B radiation (UV 1UV
2
). The tem-
perature variables were derived from WorldClim for the spatial mod-
els and from data loggers in the temporal models.
T. R. Bishop et al.
6Global Ecology and Biogeography,V
C2016 The Authors. Global Ecology and Biogeography
published by John Wiley & Sons Ltd
2008). Furthermore, the effects that we detected were at the
assemblage level and therefore reflect changes in the relative
abundances of species. Generally, the most abundant species
are those with a cuticle colour that is best suited, in a ther-
moregulatory or protective sense (Stevenson, 1985; Shine &
Kearney, 2001; Wang et al., 2008), to the prevailing environ-
mental conditions. This suggests that assemblage structure
will change as the optimum cuticle lightness changes depend-
ing on the climate. Our temporal data show that this can
happen over a relatively short time-scale through shifts in
species abundance. Such shifts in assemblage structure under
predicted levels of climate change may have cascading effects
on ecosystem functioning and integrity.
Across space, we find that, on average, assemblages have lighter
cuticles in warm environments and darker cuticles where it is
cooler. High UV-B irradiance makes a difference where it is hot,
and is associated with darker cuticles (Fig. 1b). In addition,
assemblage cuticle lightness was negatively correlated with assem-
blage body size (Fig. 1a). We find similar results through time.
Our data show that temporal changes in the assemblage cuticle
lightness were negatively related to body size (Fig. 2a) and posi-
tively related to temperature (Fig. 2b).
Our data can be interpreted in light of both of the two
major contrasting ecogeographical rules that describe and
explain colour variation. These are the thermal melanism
hypothesis, or Bogert’s rule (Clusella-Trullas et al., 2007), and
Gloger’s rule (Caro, 2005; Millien et al., 2006). The two rules
differ in their target animal groups and in their principal
underlying mechanisms. The thermal melanism hypothesis is
usually applied to ectotherms and proposes that darker col-
ours should dominate in cold environments (usually high lat-
itudes or elevations) because of the thermoregulatory benefits
of being dark. Gloger’s rule is typically applied to endo-
therms and states that darker colours are found closer to the
equator in warmer environments. This pattern may be caused
by UV-B protection, camouflage or thermoregulatory needs –
white fur can scatter radiation toward the skin for heat gain
whilst dark fur can enhance cooling via evaporation (Caro,
2005; Millien et al., 2006; Koski & Ashman, 2015). Whilst
the majority of our dataset supports the thermal melanism
hypothesis (ants are darker in colder environments) the sig-
nificant interaction of temperature and UV-B in our model-
ling procedure (Fig. 1b, Table 3) suggests that the UV-B
protection mechanism of Gloger’s rule may also be applicable
1.0 1.5 2.0
0.0 0.2 0.4 0.6 0.8 1.0
Assembla
g
e body size (mm)
Assemblage lightness
(a)
15 20 25 30
0.0 0.2 0.4 0.6 0.8 1.0
Data logger ground temperature (°C)
Assemblage lightness
(b)
Figure 2 Plots showing the relationship between mean assemblage lightness and body size (a) and mean data logger-derived summer
temperature (b) through time for the Maloti-Drakensberg and Soutpansberg mountain ranges of southern Africa (n5206). Solid black
lines display the average model predictions. Dashed lines display predictions for each individual assemblage (41 unique assemblages).
R2
m(fixed effects) 50.49, R2
c(fixed and random effects) 50.74.
1.0 1.5 2.0 2.5
0.0 0.2 0.4 0.6 0.8 1.0
Assembla
g
e body size (mm)
Assemblage lightness
(a)
10 15 20 25
0.0 0.2 0.4 0.6 0.8 1.0
WorldClim temperature (°C)
Assemblage lightness
(b)
High UV−B
Low UV−B
Figure 1 Plots showing the
relationship between mean
assemblage lightness and body size
(a) and mean WorldClim-derived
summer temperature (b). Lines
display model predictions. In (b)
the solid line represents predictions
for low levels of UV-B (10th
percentile) and the dashed line
represents predictions for high UV-
B (90th percentile) (n5274). R2
m
(fixed effects) 50.48, R2
c(fixed and
random effects) 50.62.
Gradients in ant colour and body size
Global Ecology and Biogeography, V
C2016 The Authors. Global Ecology and Biogeography
published by John Wiley & Sons Ltd 7
to ant assemblages (e.g. Bastide et al., 2014; Koski & Ash-
man, 2015).
Comparable results to ours have been found using multi-
ple species across large areas. For example, European insects
show positive relationships between cuticle lightness and
temperature (Zeuss et al., 2014), whilst the cuticle lightness
of carabid beetles is negatively related to body size across
Europe (Schweiger & Beierkuhnlein, 2015). Our results are in
agreement with these previous findings, but take them a step
further by using assemblage-level data. This provides infor-
mation on the identities and relative abundances of the spe-
cies (and their cuticle lightness) that were active at the time
of our sampling. As a consequence, the performance of dif-
ferent lightness values in different environments is captured
by our assemblage average. This point is illustrated well in
our temporal analysis. The same point in space shows differ-
ent lightness values under different temperatures – species
with the correct cuticle lightness are able to rapidly take
advantage of altered thermal conditions. The agreement that
we find between the spatial and temporal patterns greatly
strengthens the power that we have to infer a process of
assemblage change mediated by ant physiology than either
pattern would in isolation (White et al., 2010).
By restricting our assemblage data to the most common
species, we find the same patterns in cuticle lightness. This
implies that it is the dominant ant species that are driving
the relationships between cuticle lightness, temperature and
UV-B. This is important as the dominant species are con-
suming most of the energy in the system and can structure
the rest of the assemblage (Parr, 2008). This finding empha-
sizes the importance of the abiotic environment in structur-
ing local assemblages and contrasts with the majority of the
existing literature on ants (e.g. Cerd
aet al., 2013), which has
tended to focus on the importance of biotic factors such as
competition (but see Gibb, 2011). The importance of the
common species in driving these macrophysiological patterns
echoes similar findings in macroecology where it is also the
common species which drive assemblage diversity patterns
(Reddin et al., 2015).
Previous studies on this topic (Zeuss et al., 2014;
Schweiger & Beierkuhnlein, 2015), and in macroecology in
general (Beck et al., 2012), rarely have the right kinds of data
to draw conclusions at the assemblage level. We argue that
understanding the fine spatial and temporal scale of variation
is crucial for appreciating how, and why, organisms respond
to the environment. Most ectotherms do not interact with
each other, or their environment, at the 50-km
2
scale.
Instead, it is the success of individuals at finer grains that
determines population viability and, ultimately, drives ecosys-
tem functioning. It should be noted, however, that despite
the large influence that spatial extent and grain size may
have in determining geographical patterns (Rahbek, 2005),
the relationships between lightness, temperature and body
size in our dataset (grain size of c. 400 m
2
) are consistent
with those studies using a much larger grain size (Zeuss
et al., 2014; Schweiger & Beierkuhnlein, 2015). This
combination of evidence suggests that the thermoregulatory
role of colour in ectotherms may scale consistently to:
(1) influence the success of individuals (e.g. Ellers & Boggs,
2004), (2) shape assemblage structure (this study) and
(3) determine which species are present in the wider regional
pool (e.g. Zeuss et al., 2014).
Although our spatial and temporal models explain a large
amount of the assemblage-level variation in cuticle lightness
in our dataset (c. 50% for fixed effects; Table 2), a consider-
able portion of the variation remains unexplained. There are
likely to be two main sources for this variation. The first is
methodological. Our use of global surfaces (WorldClim and
glUV) in the spatial analysis is likely to have underestimated
the true range of temperatures and UV-B levels encountered
by the sampled ant assemblages. This could lead to assemb-
lages appearing lighter or darker than expected for their esti-
mated temperatures. This is less of an issue for our temporal
analysis as we used data loggers to track temperature. Sec-
ondly, we may be under-appreciating the ability of ants to
thermoregulate without the use of cuticle colour. A range of
other morphological and behavioural mechanisms can play a
role in ant thermoregulation. This has been reported mainly
for extremely hot conditions. For example, Cataglyphis spe-
cies have been recorded to use specialized reflecting hairs
(Shi et al., 2015) to thermoregulate in hot conditions. In
addition, ants have been widely reported to forage at cooler
times of the day to avoid peak temperatures (Fitzpatrick
et al., 2014), which may completely decouple the biophysical
link between their morphological thermoregulatory traits and
the environment. In cold environments, nest architecture and
building materials can keep colonies warm (H
olldobler &
Wilson, 1990), but there is little reporting of individual
worker traits that allow activity to be maintained in the cold.
We assume that these mechanisms are the exception rather
than the rule, but this may not be the case.
In summary, we have shown that the structure of assemb-
lages can be driven by the differential performance of species
based on their thermoregulatory traits. This finding suggests
that ant assemblages will have to shift in ways consistent
with thermoregulatory and protective needs as the climate
changes. Under warmer conditions, ants should become
smaller and lighter coloured. The existing literature largely
agrees with this decrease in body size (Sheridan & Bickford,
2011), but currently suggests that darker melanic individuals
will tend to be favoured under climate change scenarios
(Roulin, 2014). These predicted changes will probably filter
certain kinds of species and alter the functional composition
and outputs of assemblages.
ACKNOWLEDGEMENTS
We thank the DST-NRF Centre of Excellence for Invasion Biol-
ogy, the University of Pretoria, Chantal Ferreira, the Mazda
Wildlife Fund, Ezemvelo KZN Wildlife and the Lesotho Minis-
try of Tourism, Environment and Culture for their various roles
in supporting the Maloti-Drakensberg, Soutpansberg and
T. R. Bishop et al.
8Global Ecology and Biogeography,V
C2016 The Authors. Global Ecology and Biogeography
published by John Wiley & Sons Ltd
Cederberg transects. The Mariepskop transect was finan-
cially supported by the NRF and 19 Helicopter Squadron
at Hoedspruit provided logistical support. T.R.B. was sup-
ported by a NERC studentship. Theo Evans and Chris Jeffs
provided stimulating discussion that much improved the
direction of the manuscript. Alexandre Roulin and an
anonymous referee provided reviews which improved the
final version.
REFERENCES
Andersson, M.B. (1994) Sexual selection. Princeton University
Press, Princeton, NJ.
Barto
n, K. (2013) MuMIn: multi-model inference. R package
version 1.15.6. http://cran.r-project.org/package=MuMIn.
Bastide, H., Yassin, A., Johanning, E.J. & Pool, J.E. (2014)
Pigmentation in Drosophila melanogaster reaches its maxi-
mum in Ethiopia and correlates most strongly with ultra-
violet radiation in sub-Saharan Africa. BMC Evolutionary
Biology,14, 179.
Bates, D., Maechler, M., Bolker, B., Walker, S., Christensen,
R.H.B., Singmann, H. & Dai, B. (2014) lme4: linear mixed-
effects models using Eigen and S4. R package version 1.1-6.
http://cran.r-project.org/package=lme4.
Beck, J., Ballesteros-Mejia, L., Buchmann, C.M., Dengler, J.,
Fritz, S.A., Gruber, B., Hof, C., Jansen, F., Knapp, S &
Kreft, H. (2012) What’s on the horizon for macroecology?
Ecography,35, 673–683.
Beckmann, M., V
aclav
ık, T., Manceur, A.M.,
Sprtov
a, L.,
Wehrden, H., Welk, E. & Cord, A.F. (2014) glUV: a global
UV-B radiation data set for macroecological studies. Meth-
ods in Ecology and Evolution,5, 372–383.
Bishop, T.R., Robertson, M.P., van Rensburg, B.J. & Parr,
C.L. (2014) Elevation–diversity patterns through space and
time: ant communities of the Maloti-Drakensberg Moun-
tains of southern Africa. Journal of Biogeography,41, 2256–
2268.
Bishop, T.R., Robertson, M.P., van Rensburg, B.J. & Parr,
C.L. (2015) Contrasting species and functional beta diver-
sity in montane ant assemblages. Journal of Biogeography,
42, 1776–1786.
Blomberg, S.P., Garland, T., Jr, Ives, A.R. & Crespi, B. (2003)
Testing for phylogenetic signal in comparative data: behav-
ioral traits are more labile. Evolution,57, 717–745.
Bogert, C.M. (1949) Thermoregulation in reptiles, a factor in
evolution. Evolution,3, 195–211.
Botes, A., McGeoch, M.A., Robertson, H.G., van Niekerk, A.,
Davids, H.P. & Chown, S.L. (2006) Ants, altitude and
change in the Northern Cape Floristic Region. Journal of
Biogeography,33, 71–90.
Brown, W.L. (1953) Revisionary studies in the ant tribe
Dacetini. American Midland Naturalist,50, 1–137.
Buckley, L.B., Hurlbert, A.H. & Jetz, W. (2012) Broad-scale
ecological implications of ectothermy and endothermy in
changing environments. Global Ecology and Biogeography,
21, 873–885.
Buuren, S. & Groothuis-Oudshoorn, K. (2011) mice: multi-
variate imputation by chained equations in R. Journal of
Statistical Software,45, 1–67.
Caro, T. (2005) The adaptive significance of coloration in
mammals. BioScience,55, 125–136.
Cerd
a, X., Arnan, X. & Retana, J. (2013) Is competition a
significant hallmark of ant (Hymenoptera: Formicidae)
ecology. Myrmecological News,18, 131–147.
Chown, S.L. & Gaston, K.J. (2010) Body size variation in
insects: a macroecological perspective. Biological Reviews,
85, 139–169.
Chown, S.L. & Gaston, K.J. (2015) Macrophysiology – pro-
gress and prospects. Functional Ecology,30, 330–344.
Clusella-Trullas, S., van Wyk, J.H. & Spotila, J.R. (2007)
Thermal melanism in ectotherms. Journal of Thermal Biol-
ogy,32, 235–245.
Ducrest, A.L., Keller, L. & Roulin, A. (2008) Pleiotropy in
the melanocortin system, coloration and behavioural syn-
dromes. Trends in Ecology and Evolution,23, 502–510.
Ellers, J. & Boggs, C.L. (2004) Functional ecological implica-
tions of intraspecific differences in wing melanization in
Colias butterflies. Biological Journal of the Linnean Society,
82, 79–87.
Fitzpatrick, G., Lanan, M.C. & Bronstein, J.L. (2014) Thermal
tolerance affects mutualist attendance in an ant–plant pro-
tection mutualism. Oecologia,176, 129–138.
Gaston, K.J., Chown, S.L. & Evans, K.L. (2008) Ecogeograph-
ical rules: elements of a synthesis. Journal of Biogeography,
35, 483–500.
Gibb, H. (2011) Experimental evidence for mediation of
competition by habitat succession. Ecology,92, 1871–1878.
Heinrich, B. (1996) The thermal warriors: strategies of insect
survival. Harvard University Press, Cambridge, MA.
Hijmans, R.J., Cameron, S.E., Parra, J.L., Jones, P.G. & Jarvis,
A. (2005) Very high resolution interpolated climate surfa-
ces for global land areas. International Journal of Climatol-
ogy,25, 1965–1978.
Hodkinson, I.D. (2005) Terrestrial insects along elevation
gradients: species and community responses to altitude.
Biological Reviews,80, 489–513.
H
olldobler, B. & Wilson, E.O. (1990) The ants. Springer-Ver-
lag, Berlin.
Kearney, M.R., Isaac, A.P. & Porter, W.P. (2014) microclim:
global estimates of hourly microclimate based on long-
term monthly climate averages. Scientific Data,1, 140006.
Kingsolver, J.G. (1995) Fitness consequences of seasonal poly-
phenism in western white butterflies. Evolution,49, 942–
954.
Koski, M.H. & Ashman, T.L. (2015) Floral pigmentation pat-
terns provide an example of Gloger’s rule in plants. Nature
Plants,1,5.
Legendre, P. (2008) lmodel2: model II regression. R package
version 1.7-2. http://cran.r-project.org/package=lmodel2.
Mbanyana, N. & Robertson, H.G. (2008) Review of the ant
genus Nesomyrmex (Hymenoptera: Formicidae: Myrmici-
nae) in southern Africa. African Natural History,4, 35–55.
Gradients in ant colour and body size
Global Ecology and Biogeography, V
C2016 The Authors. Global Ecology and Biogeography
published by John Wiley & Sons Ltd 9
Millien, V., Kathleen Lyons, S., Olson, L., Smith, F.A.,
Wilson, A.B. & Yom-Tov, Y. (2006) Ecotypic variation in
the context of global climate change: revisiting the rules.
Ecology Letters,9, 853–869.
Moreau, C.S. & Bell, C.D. (2013) Testing the museum versus
cradle tropical biological diversity hypothesis: phylogeny,
diversification, and ancestral biogeographic range evolution
of the ants. Evolution,67, 2240–2257.
Munyai, T.C. & Foord, S.H. (2012) Ants on a mountain: spa-
tial, environmental and habitat associations along an altitu-
dinal transect in a centre of endemism. Journal of Insect
Conservation,16, 677–695.
Munyai, T.C. & Foord, S.H. (2015) Temporal patterns of ant
diversity across a mountain with climatically contrasting
aspects in the Tropics of Africa. PLoS One,10, e0122035.
Nakagawa, S. & Schielzeth, H. (2013) A general and simple
method for obtaining R
2
from generalized linear mixed-
effects models. Methods in Ecology and Evolution,4, 133–
142.
Pagel, M. (1999) Inferring the historical patterns of biological
evolution. Nature,401, 877–884.
Parr, C.L. (2008) Dominant ants can control assemblage spe-
cies richness in a South African savanna. Journal of Animal
Ecology,77, 1191–1198.
Pincheira-Donoso, D. (2010) The balance between predic-
tions and evidence and the search for universal macroeco-
logical patterns: taking Bergmann’s rule back to its
endothermic origin. Theory in Biosciences,129, 247–253.
Rahbek, C. (2005) The role of spatial scale and the percep-
tion of large-scale species-richness patterns. Ecology Letters,
8, 224–239.
Rapoport, E. (1969) Gloger’s rule and pigmentation of Col-
lembola. Evolution, 622–626.
R Core Team (2014) R: a language and environment for Sta-
tistical Computing. R Foundation for Statistical Computing,
Vienna, Austria.
Reddin, C., Bothwell, J. & Lennon, J. (2015) Between-taxon
matching of common and rare species richness patterns.
Global Ecology and Biogeography,24, 1476–1486.
Revell, L.J. (2012) phytools: an R package for phylogenetic
comparative biology (and other things). Methods in Ecology
and Evolution,3, 217–223.
Roulin, A. (2014) Melanin-based colour polymorphism
responding to climate change. Global Change Biology,20,
3344–3350.
Ruxton, G.D., Sherratt, T.N. & Speed, M.P. (2004) Avoiding
attack. Oxford University Press, Oxford.
Schielzeth, H. (2010) Simple means to improve the interpret-
ability of regression coefficients. Methods in Ecology and
Evolution,1, 103–113.
Schweiger, A.H. & Beierkuhnlein, C. (2015) Size dependency
in colour patterns of Western Palearctic carabids. Ecogra-
phy, doi: 10.1111/ecog.01570.
Sheridan, J.A. & Bickford, D. (2011) Shrinking body size as
an ecological response to climate change. Nature Climate
Change,1, 401–406.
Shi, N.N., Tsai, C.C., Camino, F., Bernard, G.D., Yu, N. &
Wehner, R. (2015) Keeping cool: enhanced optical reflec-
tion and heat dissipation in silver ants. Science,349, 298–
301.
Shine, R. & Kearney, M. (2001) Field studies of reptile
thermoregulation: how well do physical models
predict operative temperatures? Functional Ecology,15,
282–288.
Stevenson, R. (1985) The relative importance of behavioral
and physiological adjustments controlling body tempera-
ture in terrestrial ectotherms. The American Naturalist,
126, 362–386.
Wang, Z., Liu, R., Wang, A., Du, L. & Deng, X. (2008) Pho-
totoxic effect of UVR on wild type, ebony and yellow
mutants of Drosophila melanogaster: life span, fertility,
courtship and biochemical aspects. Science in China Series
C: Life Sciences,51, 885–893.
Werenkraut, V., Fergnani, P.N. & Ruggiero, A. (2015) Ants at
the edge: a sharp forest-steppe boundary influences the
taxonomic and functional organization of ant species
assemblages along elevational gradients in northwestern
Patagonia (Argentina). Biodiversity and Conservation,24,
287–308.
White, E.P., Ernest, S.K.M., Adler, P.B., Hurlbert, A.H. &
Lyons, S.K. (2010) Integrating spatial and temporal
approaches to understanding species richness. Philosophical
Transactions of the Royal Society B: Biological Sciences,365,
3633–3643.
Williamson, C.E., Zepp, R.G., Lucas, R.M., Madronich, S.,
Austin, A.T., Ballar
e, C.L., Norval, M., Sulzberger, B.,
Bais, A.F. & McKenzie, R.L. (2014) Solar ultraviolet radia-
tion in a changing climate. Nature Climate Change,4,
434–441.
Willmer, P. & Unwin, D. (1981) Field analyses of insect heat
budgets: reflectance, size and heating rates. Oecologia,50,
250–255.
Zeuss, D., Brandl, R., Br
andle, M., Rahbek, C. & Brunzel, S.
(2014) Global warming favours light-coloured insects in
Europe. Nature Communications,5, 3874. [24866819]
SUPPORTING INFORMATION
Additional supporting information may be found in the
online version of this article at the publisher’s web-site:
Appendix S1 Colour assignment.
Appendix S2 Multivariate imputation using chained
equations (MICE) of MacDonnell Ranges body size.
Appendix S3 Phylogenetic signal.
Appendix S4 Relationship between temperature values
obtained from data loggers in the Maloti-Drakensberg,
Soutpansberg and Patagonian Andes and those extracted
from WorldClim.
Appendix S5 Species richness effects.
Appendix S6 Modelling of lightness across space using
microclim temperature data.
Appendix S7 Modelling of lightness across space for
common and rare species.
T. R. Bishop et al.
10 Global Ecology and Biogeography,V
C2016 The Authors. Global Ecology and Biogeography
published by John Wiley & Sons Ltd
Figure S1.1 Colour wheels used to categorize the colour of
ant species across the three continents.
Figure S1.2 Histograms showing the standard errors of
lightness values estimated for the head, mesosoma and gaster
by the five observers used in this study on a set of 71
photographs of ants from http://antweb.org/.
Figure S2.1 Plots showing relationship between
morphological traits for Australian ants from the Snowy
Mountains, MacDonnell Ranges and Ben Lomond Plateau,
Tasmania.
Figure S3.1 Plot showing the distribution of average genus
body sizes across the ant subfamilies present in this study.
Figure S4.1 Relationship between temperature values
obtained from data loggers and those extracted from
WorldClim.
Figure S5.1 Stacked density plot showing the distribution of
lightness values for each mountain range.
Figure S5.2 Plot showing the relationship between
assemblage lightness and species richness.
Figure S6.1 Relationship between temperature values
obtained from data loggers and those extracted from
microclim.
Figure S6.2 Plots showing the relationship between
assemblage lightness and body size and average microclim-
derived summer temperature.
Figure S7.1 Plots showing the relationship between
assemblage lightness and body size and average WorldClim-
derived summer temperature.
Figure S7.2 Plots showing the relationship between
assemblage lightness and body size and residual summer
UV-B.
Table S6.1 Comparative and summary statistics for linear
mixed models explaining variation in ant assemblage colour
across space.
Table S6.2 Test statistics (v
2
), degrees of freedom and
P-values from Type III Wald tests on the best spatial model
(top ranked from Table S6.1).
Table S7.1 Comparative and summary statistics for linear
mixed models explaining variation in ant assemblage colour
across space for either common or rare species.
Table S7.2 Te s t s t a t i s t i c s ( v
2
),degreesoffreedomand
P-values from Type III Wald tests on the best spatial models for
common and rare species subsets (top ranked from Table S7.1).
BIOSKETCH
Tom R. Bishop is interested in using morphology and
physiology to understand the distribution of biological
diversity, particularly that of the ants.
Editor: Daniel Pincheira-Donoso
Gradients in ant colour and body size
Global Ecology and Biogeography, V
C2016 The Authors. Global Ecology and Biogeography
published by John Wiley & Sons Ltd 11
... In general, larger bodies gain and lose heat more slowly than smaller bodies, and darker colored organisms heat up more rapidly than paler colored ones (Willmer & Unwin, 1981). These relationships appear to play out at the community level: ant species with darker (Bishop et al., 2016) and larger (Bishop et al., 2016;Gibb et al., 2017) workers are more abundant in colder environments across the globe. Consequently, as the climate warms, we may expect to see species with smaller and paler workers coming to dominate ant communities as their phenotypes are more suited to existing in warm conditions. ...
... In general, larger bodies gain and lose heat more slowly than smaller bodies, and darker colored organisms heat up more rapidly than paler colored ones (Willmer & Unwin, 1981). These relationships appear to play out at the community level: ant species with darker (Bishop et al., 2016) and larger (Bishop et al., 2016;Gibb et al., 2017) workers are more abundant in colder environments across the globe. Consequently, as the climate warms, we may expect to see species with smaller and paler workers coming to dominate ant communities as their phenotypes are more suited to existing in warm conditions. ...
... As ectotherms, the effects of UVB interact with temperature. With lowmoderate UVB, there is a benefit to being darker in cool locations and paler in hot locations, however, in hot locations with very high UVB, ants tend to have darker pigmentation as protection (Bishop et al., 2016;Law et al., 2020). We might, therefore, expect that where UVB is especially high, assemblage composition will change as darker colored ant species will do better. ...
Article
Full-text available
Ants (Hymenoptera: Formicidae) are one of the most dominant terrestrial organisms worldwide. They are hugely abundant, both in terms of sheer numbers and biomass, on every continent except Antarctica and are deeply embedded within a diversity of ecological networks and processes. Ants are also eusocial and colonial organisms-their lifecycle is built on the labor of sterile worker ants who support a small number of reproductive individuals. Given the climatic changes that our planet faces, we need to understand how various important taxonomic groups will respond; this includes the ants. In this review, we synthesize the available literature to tackle this question. The answer is complicated. The ant literature has focused on temperature, and we broadly understand the ways in which thermal changes may affect ant colonies, populations, and communities. In general, we expect that species living in the Tropics, and in thermally variable microhabitats, such as the canopy and leaf litter environments, will be negatively impacted by rising temperatures. Species living in the temperate zones and those able to thermally buffer their nests in the soil or behaviorally avoid higher temperatures, however, are likely to be unaffected or may even benefit from a changed climate. How ants will respond to changes to other abiotic drivers associated with climate change is largely unknown, as is the detail on how altered ant populations and communities will ramify through their wider ecological networks. We discuss how eusociality may allow ants to adapt to, or tolerate, climate change in ways that solitary organisms cannot and we identify key geographic and phylogenetic hotspots of climate vulnerability and resistance. We finish by emphasizing the key research questions that we need to address moving forward so that we may fully appreciate how this critical insect group will respond to the ongoing climate crisis.
... First, as darker species absorb more solar radiation, they will be more prevalent in cooler environments (Clusella-Trullas et al., 2007). In contrast, another hypothesis states that with increased solar radiation in hotter environments, darker species will be more prevalent, as they can mitigate the deleterious effects of UV-B radiation (Bishop et al., 2016). Similarly, and finally, as darker individuals also have less permeable cuticles, they are predicted to be dominant in drier (and thus hotter) areas (Law et al., 2020). ...
... Indeed, a recent study of ants in Borneo showed that species are darker in the canopy (Law et al., 2020). Different hypotheses are thought to be of varying importance in different systems, however (Bishop et al., 2016;Law et al., 2020;Stanbrook et al., 2021). Similarly, the relationship between body size and land-use change is complex, with different associations depending on taxa, region and study (e.g. ...
... Based on previous studies, we hypothesised that species with more positive associations with temperature would exhibit higher CT max and lower pilosity (Peters et al., 2016). We expected that dung beetle body size and cuticle lightness would mediate species responses to temperature in our study landscape, but were unsure as to the directionality of the relationship due to competing hypotheses in the relevant literature (body size- Gardner et al., 2008;Fuzessy et al., 2021; cuticle lightness- Bishop et al., 2016;Law et al., 2020;Stanbrook et al., 2021). ...
Article
Full-text available
Temperature change is an often‐assumed, but rarely‐tested, mechanism by which sensitive species may decline in forest landscapes following habitat degradation, fragmentation and destruction. Traits mediate how species respond to environmental change, with physiological, morphological and behavioural traits key to determining the response of ectotherms to temperature. We collected data on traits linked to thermal sensitivity (critical thermal maxima, body size, cuticle lightness and pilosity) for 46 dung beetle species (Scarabaeinae) in a forest‐oil‐palm mosaic in Malaysian Borneo. By combining these data with a large‐scale community sampling campaign (>59,000 individuals sampled from >600 traps) and an airborne‐LiDAR‐derived thermal map, we investigated how traits mediate species‐ and community‐level responses to temperature. Using hierarchical models, we found that critical thermal maxima predicted how species respond to maximum temperatures. These results were mirrored in community‐level analyses alongside similar patterns in other thermal traits. Increased body size and decreased pilosity were associated with higher temperatures, while cuticle lightness showed a complex relationship with temperature across the disturbance gradient. Our findings highlight the potential mechanisms causing the decline of forest specialists in human‐modified landscapes that result in changes to community patterns and processes.
... Melanin provides UV protection, and insects that live in cooler environments tend to have darker cuticle to aid in heat absorption, while insects that live in warmer climates tend to be lighter (TRULLAS & al. 2007, BISHOP & al. 2016. Rarely, ants display structural colors that appear iridescent or metallic, which are produced by nanostructures that interfere with light reflection . ...
... Ants with larger colonies also tend to produce workers with lower nitrogen levels, which provides further support that workers with thinner cuticle are cheaper to produce (DAVIDSON 2005). With respect to color, BISHOP & al. (2016) found that darker-colored ants were more likely to be found in cold environments, suggesting that dark colors aid in heat absorption. In a tropical rainforest, Law & al. (2019) found that ant species living in the canopy or understory were two times darker than ants living on the ground or in subterranean strata, which does not match predictions related to temperature and instead supports a role of increased melanin for UV protection and desiccation resistance. ...
Article
My thesis covers the intricacies of ant microsculpture diversity, classification, evolution and function. To do this, the thesis is organized into two chapters. The first chapter is a review of ant cuticle and microsculpture diversity as well as including analyses examining the evolution and lability of microsculpture traits. I then review the literature for functional hypotheses related to ant microsculpture. The second chapter explores the relationship of microsculpture and important morphological, physiological, and ecological traits to evaluate support for functional hypotheses.
... Several hypotheses have been proposed to explain the variation in body melanization in animals, with a special focus on ectotherms. The Thermal Melanism Hypothesis suggests that darker species are more commonly found in cooler habitats (Clusella-Trullas et al. 2007;Bishop et al. 2016;Pinkert et al. 2017;Pinkert and Zeuss 2018;Clusella-Trullas and Nielsen 2020), as body melanization is an important feature for thermoregulation, with darker-coloured animals absorbing heat at faster rates (Willmer and Unwin 1981). The Photoprotection Hypothesis posits that species from habitats with higher UV-B incidence are darker, as melanin protects cells against mutational effects by absorbing most of the incident radiation (Mosse and Lyakh 1994;Wang et al. 2008;Ulbing et al. 2019). ...
... Despite these behavioural idiosyncrasies, studies on ant melanization have never considered this facet of ant biology, focusing instead on a single worker subcaste (e.g. Bishop et al. 2016;Schofield et al. 2016;Law et al. 2020). Therefore, these previous studies have overlooked important information from taxa with high morphological variation between colony members. ...
Article
Full-text available
One of the main aspects associated with the diversity in animal colour is the variation in melanization levels. In ectotherms, melanism can be advantageous in aiding thermoregulation through heat absorption. Darker bodies may also serve as a shield from harmful UV-B radiation. Melanism may also confer protection against parasites and predators through improving immunity responses and camouflage in regions with high precipitation, with complex and shaded vegetations and greater diversity of pathogens and parasites. We studied melanism evolution in the globally distributed ant genus Pheidole under the pressures of temperature, UV-B radiation and precipitation, while considering the effects of body size and nest habit, traits that are commonly overlooked. More importantly, we account for worker caste polymorphism, which is marked by distinct roles and behaviours. We revealed for the first time distinct evolutionary trajectories for each worker subcaste. As expected, major workers from species inhabiting locations with lower temperatures and higher precipitation tend to be more melanised. Curiously, we show a slight trend where minor workers of larger species also tend to have darker bodies when inhabiting regions with higher precipitation. Lastly, we did not find evidence for the effects of UV-B radiation and nest habit in the lightness variation of workers. Our paper explores the evolution of ant melanization considering a marked ant worker polymorphism and a wide range of ecological factors. We discuss our findings under the light of the Thermal Melanism Hypothesis, the Photoprotection Hypothesis and the Gloger’s Rule.
... In addition, we recorded differences between species in other traits that might affect predation due to interaction with the daily variable environment on the steppe, with very hot middays and cold nights. These traits include colour (surrogate for thermal melanism and environmental stress) since a darker ant tolerates higher UV radiation and heats faster allowing an earlier foraging onset, and cuticle sculpturing and pilosity (proxies of thermal tolerance) as higher texture allows better temperature regulation (Bishop et al., 2016;Law et al., 2020). These traits were categorized as suggested by Parr et al. (2017), using the head/thorax/gaster combination for colour codification and counting the number of hairs for pilosity. ...
... In addition, we observed that D. antarcticus foragers were active at arenas for almost 1 h longer than D. tener. This can be explained by the darker coloration of this species compared to D. tener, since dark colour, and specifically, melanin, has a protective role against harmful UV-B radiation (Bishop et al., 2016), which is very intense in the Patagonian steppe (Austin & Vivanco, 2006), and melanization provides greater resistance to desiccation by decreasing cuticular permeability (see Law et al., 2020 for an example in ants). In this way, having a slightly wider time frame available for foraging for D. antarcticus, might allow this species to gather resources when D. tener is not active, as occurs in other heat-tolerant, subordinate ant species (Cerdá et al., 1997). ...
Article
Predation is an important force structuring ecological communities. However, it is still controversial whether larger predator groups are more efficient at exploiting abundant resources. Here, we explored the association between the number of foragers and predation ability in generalist ant species that differ in forager numbers when exploiting resources. We conducted a field experiment by increasing caterpillar density around nests of two abundant Dorymyrmex ant species in the semiarid Patagonian steppe, where D. tener allocates a higher number of foragers to resource exploitation than D. antarcticus. We (1) compared the association between predation effectiveness (success to complete a task) and efficiency (speed of task performance and economy of foragers) with the number of foragers involved between species, and (2) studied how they responded numerically to increasing prey densities, by sequentially adding 3, 6 and 12 larvae in the same foraging arena. Finally, (3) we compared behavioural and morphological traits related to predation between these ant species. Although D. tener discovered a similar number of arenas with larvae than D. antarcticus, it was more effective as it recruited more and removed more larvae. This species was also more efficient than D. antarcticus in all predation subtasks, and the time used to remove one larva depended on prey density, being faster for the high‐larvae density. Besides the number of foragers, ant species did not differ in other behavioural traits, and D. tener's foragers were slightly larger than those of D. antarcticus. This study illustrates how, in social predators, the predator group size and individual behavioural characteristics may act in conjunction, with relevant consequences at ecological, evolutionary, and applied levels, including potential implications for pest control. Los números importan: la habilidad depredadora aumenta con el tamaño de grupo en especies de hormigas omnívoras con similares rasgos asociados a la depredación La depredación es una fuerza importante que estructura las comunidades ecológicas. Sin embargo, hay controversia sobre la relación entre el tamaño de grupo y la capacidad de depredación, específicamente si los grupos de depredadores más grandes son más eficientes en la explotación de recursos abundantes. Aquí, exploramos la asociación entre el número de forrajeras y la capacidad de depredación en especies de hormigas generalistas que difieren en el número de forrajeras cuando explotan recursos. Realizamos un experimento de campo aumentando la densidad de orugas alrededor de nidos de dos especies de hormigas Dorymyrmex abundantes en la estepa patagónica semiárida, donde D. tener asigna un mayor número de forrajeras a la explotación de recursos que D. antarcticus. Específicamente, (1) comparamos la asociación entre la efectividad de la depredación (éxito para completar una tarea) y la eficiencia (velocidad de ejecución de la tarea y menor cantidad de forrajeras involucradas) con el número de forrajeras involucradas por especie, y (2) estudiamos cómo respondieron al aumento de la densidad de presas, agregando secuencialmente 3, 6 y 12 larvas en la misma arena de forrajeo. Finalmente, (3) comparamos los rasgos morfológicos y de comportamiento de cada especie de hormiga relacionados con la depredación. Aunque D. tener descubrió un número similar de arenas con larvas que D. antarcticus, fue más eficaz ya que reclutó más y removió más larvas. Esta especie también fue más eficiente que D. antarcticus en todas las subtareas de depredación, y el tiempo utilizado para remover una larva dependió de la densidad de presas, siendo más rápido para la mayor densidad de larvas. Más allá del número de forrajeras, las especies de hormigas no difirieron en otros rasgos comportamentales, y las forrajeras de D. tener fueron apenas más grandes que las de D. antarcticus. Este estudio ilustra cómo el tamaño del grupo de depredadores y las características de comportamiento individuales pueden actuar en conjunto en los depredadores sociales, con consecuencias relevantes a nivel ecológico, evolutivo y aplicado, incluidas las posibles implicaciones para el control de plagas.
... Supporting our alternative hypothesis H2, we were able to relate differences between sites in the strength of trait-assembly patterns or dominant trait axis to biogeography (position relative to northern extent of Andes) and climate (precipitation seasonality). Our analysis of geographical variance in trait-assembly patterns joins only a few other geographically extensive studies of animal communities along environmental gradients, including those of fish (Lamouroux et al., 2002;McLean et al., 2021), ants (Bishop et al., 2016;Gibb et al., 2018), bees (Moretti et al., 2009) and birds (Barnagaud et al., 2019;Matthews et al., 2015). These studies show that geography can have a range of effects on community trait filtering by local environments, from minor effects of geographical location (McLean et al., 2021) to dominant effects of biogeography (Barnagaud et al., 2019) and bioclimatic context (Moretti et al., 2009). ...
Article
1. It has been argued that the mechanisms structuring ecological communities may be more generalizable when based on traits than on species identities. If so, patterns in the assembly of community‐level traits along environmental gradients should be similar in different places in the world. Alternatively, geographic change in the species pool and regional variation in climate might result in site‐specific relationships between community traits and local environments. These competing hypotheses are particularly untested for animal communities. 2. Here we test the geographic constancy of trait‐based assembly patterns using a widespread multi‐trophic community: aquatic macroinvertebrates within bromeliads. We used data on 615 invertebrate taxa from 1656 bromeliads in 26 field sites from Mexico to Argentina. We summarized invertebrate traits with four orthogonal axes, and used these trait axes to examine trait convergence and divergence assembly patterns along three environmental gradients: detrital biomass and water volume in bromeliads, and canopy cover over bromeliads. 3. We found no overall signal of trait‐based assembly patterns along any of the environmental gradients. However, individual sites did show trait convergence along detrital and water gradients, and we built predictive models to explore these site differences. 4. Sites that showed trait convergence along detrital gradients were all north of the Northern Andes. This geographic pattern may be related to phylogeographic differences in bromeliad morphology. Bromeliads with low detritus were dominated by detritivorous collectors and filter feeders, where those with high detritus had more sclerotized and predatory invertebrates. 5. Sites that showed the strongest trait convergence along gradients in bromeliad water were in regions with seasonal precipitation. In such sites, bromeliads with low water were dominated by soft‐bodied, benthic invertebrates with simple life cycles. In less seasonal sites, traits associated with short‐term desiccation resistance, such as hard exoskeletons, were more important. 6. In summary, we show that there are strong geographic effects on the trait‐based assembly patterns of this invertebrate community, driven by the biogeography of their foundational plant species as well as by regional climate. We suggest that inclusion of biogeography and climate in trait‐based community ecology could help make it a truly general theory.
... Research on dragonflies, butterflies, and ants point to the possibility that color plays a role in how insect species respond to climate change (Zeuss et al. 2014, Bishop et al. 2016, Heidrich et al. 2018, as cold-blooded animals (ectotherms, q.v.) often are darker in color in colder climates (Poikela et al. 2021). In a study of oak-associated beetles across a climate gradient in Norway and Sweden, it was expected that increased summer temperature would positively influence all wood-living beetle species whereas precipitation would be less important with a negligible or negative impact (Gough et al. 2015). ...
Technical Report
Full-text available
In this report, we summarize the current state of knowledge and best estimates of how climate change is expected to impact Norwegian forest ecosystems from now to the year 2100
... We focused on thermal tolerance and morphological traits because they are historically used to assess survival and performance in different climates (Roeder et al., 2021). However, additional traits such as pilosity, cuticle reflectance and colour contribute to insect thermal tolerance (Bishop et al., 2016;Shi et al., 2015). ...
Article
Full-text available
Aim Predictions of future species distributions rest on the assumption that climatic conditions in the current range reflect fundamental niche requirements. So far, it remains unclear to what extent this is true. We tested if three important factors determining fundamental niche—ecophysiology, morphology and evolutionary history—can predict the realized niche, using thermal specialist ants. They are suitable model organisms because their body temperature, metabolism and fitness are closely tied to the habitat temperatures. Location Iberian Peninsula and Maghreb. Time period 2013–2015. Major taxa studied Ants (Hymenoptera:Formicidae). Methods We measured heat tolerance, chill coma recovery, body size and phylogenetic relationships in 19 desert specialist ants in the genus Cataglyphis to test if these important determinants of fundamental niches are good predictors of species realized niches. We modelled species climatic niches using 19 bioclimatic variables from WorldClim for recorded occurrence of each species. Results None of the determinants of the species' fundamental niche were linked to their realized climatic niche, modelled using species distribution models. However, both heat tolerance and chill coma recovery were highly correlated with body size and all three thermoregulatory traits were phylogenetically constrained, suggesting they reflect fundamental requirements of each species. Main conclusions Our results challenge the basic assumption of climatic niche modelling, that the realized niche can be used as a proxy for determining fundamental niche requirements. These findings are particularly concerning for studies that use the species' current realized niche to predict their responses to climate change.
Article
Full-text available
Seasonal dynamics of diversity patterns are a key component to understand when assessing ecological communities across temporal scales given that long-term trends in diversity are often a product of the intricate dynamisms that occur at shorter temporal scales. However, seasonal trends in diversity are usually dependent on local-scale conditions, such as habitat types or the demographic characteristics of a given fauna, thus requiring better data coverage from consistent local-scale sampling. Furthermore, the assessment of seasonal dynamics in the context of functional diversity derived from trait-based data is often lacking in many important taxa such as insects. In this study, I quantify and describe the diversity of a Floridian subtropical aboveground ant community from monthly sampling across seasons using both contemporary taxonomic diversity metrics and functional diversity metrics. Results show differences in the timing of peaks across different diversity metrics. Species richness and abundances peak in months leading up to wet seasons while functional richness and divergence peak near the end of the wet season. This asynchrony is likely a result of species-specific differences in natural histories and demographic dynamics. While clear temporal dynamics are observed across diversity metrics, differences between wet or dry seasons were lacking for all metrics except functional richness. Fine-scale sampling data of seasonal trends in insect communities compiled from studies like this will be essential tools for future assessments and predictions of insect biodiversity.
Article
Hair (i.e., pelage/fur) is a salient feature of primate (including human) diversity and evolution—serving functions tied to thermoregulation, protection, camouflage, and signaling—but wild primate pelage evolution remains relatively understudied. Specifically, assessing multiple hypotheses across distinct phylogenetic scales is essential but is rarely conducted. We examine whole body hair color and density variation across Indriidae (Avahi, Indri, Propithecus)—a lineage that, like humans, exhibits vertical posture (i.e., their whole bodies are vertical to the sun). Our analyses consider multiple phylogenetic scales (family‐level, genus‐level) and hypotheses (e.g., Gloger's rule, the body cooling hypotheses). We obtain hair color and density from museum and/or wild animals, opsin genotypes from wild animals, and climate data from WorldClim. To analyze our data, we use phylogenetic generalized linear mixed models (PGLMM) using Markov chain Monte Carlo algorithms. Our results show that across the Indriidae family, darker hair is typical in wetter regions. However, within Propithecus, dark black hair is common in colder forest regions. Results also show pelage redness increases in populations exhibiting enhanced color vision. Lastly, we find follicle density on the crown and limbs increases in dry and open environments. This study highlights how different selective pressures across distinct phylogenetic scales have likely acted on primate hair evolution. Specifically, our data across Propithecus may implicate thermoregulation and is the first empirical evidence of Bogert's rule in mammals. Our study also provides rare empirical evidence supporting an early hypothesis on hominin hair evolution. Adherence to Bogert's rule across sifaka lemurs. Darker coat colors are more common where it is colder–a potential implication for thermoregulation.
Article
The western white butterfly, Pontia occidentalis, has distinctly different wing phenotypes during spring and summer generations as a result of phenotypic plasticity (seasonal polyphenism). We experimentally generated different seasonal phenotypes in the lab by altering photoperiodic conditions during rearing, and released the resulting butterflies in the field. Mark-recapture studies were then used to estimate the effects of the polyphenism on activity patterns and adult survival in both late-spring (one study) and summer (two studies) conditions. There were no significant effects of rearing treatment on temporal patterns of behavioral activity during either the late-spring or the summer field studies. Recapture probabilities were consistently higher for males than females in all three field studies; in the summer 1992 study, recapture probabilities were higher for long-day (LD) than for short-day (SD) treatment groups. During the late-spring 1992 study, there were no significant differences between LD and SD treatment groups for survival probabilities. In the two summer studies, there were significant effects of photoperiodic treatment on survival probabilities; in the summer 1992 study, LD individuals consistently had higher survival probabilities than SD individuals; in the summer 1991 study, there was a significant interaction between treatment and time period on survival probabilities, such that survival probabilities were higher for LD than for SD individuals in two of four time periods. The consistent differences in survival probabilities between treatment groups in the summer 1992 study can be accounted for by the differences in wing traits between the treatment groups. Micrometeorological data from the study site showed that midday ambient temperatures averaged ~3°C hotter during the 1992 than the 1991 summer study and that ambient conditions during the late-spring 1992 study were relatively warm and sunny for the season. These results document the varying relationships between phenotype and fitness in the temporally fluctuating environments experienced by this population.
Code
Tools for performing model selection and model averaging. Automated model selection through subsetting the maximum model, with optional constraints for model inclusion. Model parameter and prediction averaging based on model weights derived from information criteria (AICc and alike) or custom model weighting schemes. [Please do not request the full text - it is an R package. The up-to-date manual is available from CRAN].
Article
Body colouration is of high evolutionary relevance for most animals. Several competing hypotheses exist regarding the evolutionary reasons for animal colouration ranging from predator avoidance and sexual advertisement to neutral selection. Among these hypotheses, biophysical principles suggest the thermoregulatory importance of dark colouration which in turn strongly depends on species body size. This body size – darkness trade-off is based on sound theoretical background conceptualized in the thermal melanism hypothesis and is confirmed by numerous case studies for individual species. However, evidence for the general relevance of this trade-off on large spatial and taxonomic scale is still missing. Here we specifically focus on this body size – colouration trade-off for a hyper-diverse and cosmopolitan group of insects, namely ground beetles. We combined colour information with trait data and distributional as well as bioclimatic attributes for more than 1,000 carabid species from the entire Western Palearctic. We quantified species-specific body colouration from high-quality, standardised digital photographs using the Munsell colour system. We detect a strong increase of colour darkness with body size from small to medium-sized carabids up to a body size threshold of 15 mm which is consistent with the thermal melanism hypothesis. However, body size showed no effect above this threshold and colour darkness remained constantly high which is in accordance with previous ideas about the size-dependency of thermoregulative control mechanisms (size dependence hypothesis). By demonstrating a strong tendency towards darkness with increasing body size, we illustrate the inter-specific relevance of body colouration for this cosmopolitan group of ectotherms on a continental scale. The putative thermoregulative trade-off between body size and melanism seems to be of significant importance for carabids on a broad spatial scale and may be a general but still underestimated phenomenon for ectotherms in general, although other mechanistic drivers cannot be completely neglected.
Article
Aim. Our primary aim is to understand how assemblages of rare (restricted range) and common (widespread) species are correlated with each other among different taxa. We tested the proposition that marine species richness patterns of rare and common species differ, both within a taxon in their contribution to the richness pattern of the full assemblage and among taxa in the strength of their correlations with each other. Location. The UK intertidal zone. Methods. We used high-resolution marine datasets for UK intertidal macroalgae, molluscs and crustaceans each with more than 400 species. We estimated the relative contribution of rare and common species, treating rarity and commonness as a continuous spectrum, to spatial patterns in richness using spatial cross-correlations. Correlation strength and significance was estimated both within and between taxa. Results. Common species drove richness patterns within taxa, but rare species contributed more when species were placed on an equal footing via scaling by binomial variance. Between taxa, relatively small sub-assemblages (fewer than 60 species) of common species produced the maximum correlation with each other, regardless of taxon pairing. Cross-correlations between rare species were generally weak, with maximum correlation occurring between small sub-assemblages in only one case. Cross-correlations between common and rare species of different taxa were consistently weak or absent. Main conclusions. Common species in the three marine assemblages were congruent in their richness patterns, but rare species were generally not. The contrast between the stronger correlations among common species and the weak or absent correlations among rare species indicates a decoupling of the processes driving common and rare species richness patterns. The internal structure of richness patterns of these marine taxa is similar to that observed for terrestrial taxa.