Dev Genes Evol (2016) 226:245–256
DOI 10.1007/s00427-016-0543-6
ORIGINAL ARTICLE
Size relationships of different body parts in the three
dipteran species Drosophila melanogaster, Ceratitis capitata
and Musca domestica
Natalia Siomava 1 & Ernst A. Wimmer 1 & Nico Posnien 1
Received: 11 January 2016 / Accepted: 5 April 2016 / Published online: 26 April 2016
# Springer-Verlag Berlin Heidelberg 2016
Abstract Body size is an integral feature of an organism that
influences many aspects of life such as fecundity, life span and
mating success. Size of individual organs and the entire body
size represent quantitative traits with a large reaction norm,
which are influenced by various environmental factors. In the
model system Drosophila melanogaster, pupal size and adult
traits, such as tibia and thorax length or wing size, accurately
estimate the overall body size. However, it is unclear whether
these traits can be used in other flies. Therefore, we studied
changes in size of pupae and adult organs in response to different rearing temperatures and densities for D. melanogaster,
Ceratitis capitata and Musca domestica. We confirm a clear
sexual size dimorphism (SSD) for Drosophila and show that
the SSD is less uniform in the other species. Moreover, the
size response to changing growth conditions is sex dependent.
Comparison of static and evolutionary allometries of the studied traits revealed that response to the same environmental
variable is genotype specific but has similarities between species of the same order. We conclude that the value of adult
Communicated by Angelika Stollewerk
This article is part of the Special Issue BSize and Shape: Integration of
morphometrics, mathematical modelling, developmental and
evolutionary biology^, Guest Editors: Nico Posnien—Nikola-Michael
Prpic.
* Ernst A. Wimmer
ewimmer@gwdg.de
* Nico Posnien
nposnie@gwdg.de
1
Johann-Friedrich-Blumenbach-Institute of Zoology and
Anthropology, Göttingen Center for Molecular Biosciences
(GZMB), Department of Developmental Biology,
Georg-August-University Göttingen, Ernst-Caspari-Haus,
Justus-von-Liebig-Weg 11, 37077 Göttingen, Germany
traits as estimators of the absolute body size may differ among
species and the use of a single trait may result in wrong assumptions. Therefore, we suggest using a body size coefficient computed from several individual measurements. Our
data is of special importance for monitoring activities of natural populations of the three dipteran flies, since they are
harmful species causing economical damage (Drosophila,
Ceratitis) or transferring diseases (Musca).
Keywords Body size . Diptera . Allometry . Temperature .
Larva crowding . Sexual dimorphism
Introduction
Body size is an important characteristic of an animal that determines not only the ecological niche the animal occupies but
also its life style and success rate. It was observed that smallersized populations of a species reside in warmer areas, while
larger-sized populations are found in colder regions. In 1847,
Bergmann investigated this phenomenon and stated that the
rule holds true for most living organisms (Bergmann 1847).
The rule was also confirmed for most poikilotherms, including
insects such as different Drosophila species (Ray 1960;
Atkinson 1994; Kingsolver and Huey 2008). In Drosophila,
various life history traits have been shown to be body size
dependent. For instance, mating success of males depends
on body and wing size (Partridge et al. 1987), while female
body size highly correlates with fecundity (Nunney and
Cheung 1997). Even lifespan is tightly tuned with the absolute
body size of Drosophila (Miller and Thomas 1958; Khazaeli
et al. 2005).
The genotype alone cannot entirely explain the wide
variation of the size reaction to environmental conditions observed in Drosophila. Many environmental
246
factors play a significant role in body size regulation
influencing both, the overall body size and the size of
certain organs. In natural populations, increasing latitudes and altitudes have an effect similar to those of
temperature (Anderson 1966; Robinson and Partridge
2001). Nutrition (Beadle et al. 1938), crowding (Santos
et al. 1994), infections (DiAngelo et al. 2009) and different oxygen levels (Peck and Maddrell 2005) are also
known to interact with fly development and to regulate
body size. Many of these effects are already wellcharacterized in the classical model system Drosophila
melanogaster (Edgar 2006). On the molecular level, recent research in Drosophila revealed major gene regulatory networks and regulation mechanisms underlying
body and organ size control (Mirth and Shingleton
2012). Many of the studied pathways interact with hormonal regulation during the larva development. For instance, both insulin and ecdysone signalling were found
to be involved in the growth rate and nutritional reaction norms in insects (Edgar 2006; Mirth et al. 2014;
Koyama et al. 2014; Gokhale and Shingleton 2015).
In holometabolous insects, such as Drosophila, different
adult organs develop from different larval anlagen, the
imaginal discs (Cohen 1993). For some of these anlagen,
it has been shown that their growth is highly coordinated
(Oliveira et al. 2014). Hence, the sum of the development
of individual organ precursors results into certain proportions of body parts in an adult fly. Due to the special
development of holometabolous insects, interactions of
the environment and a growing individual mainly occur
during the feeding larval and developing pupal stages.
On the other hand, the solid pupal case comprises the
whole body and its volume does not depend on the environment. Thus, pupal volume is considered the best estimator of the overall body size in Drosophila (Shingleton
et al. 2008; Stillwell et al. 2011), but this parameter is
usually impossible to assess in wild populations. Thus,
many researchers tend to use adult structures, such as
thorax and tibia lengths or wing size, to estimate overall
body size (Cavicchi et al. 1989; Pitnick and Markow
1995; de Moed et al. 1997; Kacmarczyk and Craddock
2000). While these traits are generally accepted as estimators of absolute body size in Drosophila, it is not yet clear
whether they are suitable for other dipteran species.
Therefore, we investigated the influence of different environmental conditions on the size of adult traits, pupa and
overall body size in three dipteran species that exhibit
clear size differences and occupy different ecological
niches. The smallest fly in our survey was D. melanogaster
that we included as a well-established reference model. D.
melanogaster and some of its close relatives are serious pests
in some parts of the world (Lutz 1948; Demerec 1950). The
second, relatively bigger fly was the Mediterranean fruit
Dev Genes Evol (2016) 226:245–256
fly Ceratitis capitata, known to be a worldwide pest
causing extensive economic damages (ChurchillStanland et al. 1986). The biggest fly in our survey
was the common housefly Musca domestica, one of
the widely distributed pests carrying serious diseases
(Hewitt 1914). Since all three studied flies represent
pest species, it is of major interest to monitor and control them in nature in order to eventually prevent negative consequences of their propagation. In this case, estimation of the fly size might be an indicator for female
fecundity and mating success of males (Partridge et al.
1987; Nunney and Cheung 1997). We also chose
models, for which established laboratory strains exist
and which can be easily used for experiments under
controlled conditions.
Here, we study the size of pupae and various adult organs
changing in response to different rearing temperatures and
crowding conditions. We show a clear sexual size dimorphism
for many traits in all three species and sex-dependent response
to changing growth conditions. Next, we compare allometries
of the traits between the chosen species and demonstrate that
response to the same environmental variable is genotype specific but has similarities between species of the same order.
Eventually, we show that adult estimators of the absolute body
size may differ among species and the use of a single trait
known to correlate with the body size in other species may
result in misinterpretation of the data. As a consequence, we
suggest using a body size coefficient computed from several
individual measurements.
Materials and methods
Flies
We used a well-established laboratory strain for D.
melanogaster (w 1118 ; Bloomington Drosophila Stock
Center; the stock was kept at 18 °C for approximately
20 years) and a wild-type strain from Egypt for C. capitata
(Egypt II, IAEA). The M. domestica strain (ITA1) was
collected in the south of Italy (Altavilla Silentia) and
established as a laboratory culture in 2013 (Y. Wu and L.
Beukeboom, GELIFES, The Netherlands).
Drosophila flies were kept at 18 °C on standard food.
Ceratitis were reared at 28 °C, 55 ± 5 % RH on an artificial
diet composed by 52.5 g of yeast extract, 52.5 g of carrot
powder, 2 g of sodium benzoate, 1.75 g of agar, 2.25 ml of
32 % HCl, 5 ml of Nipagin (2.86 g of Nipagin in 10 ml of
ethanol), and water up to 500 ml for larvae. For adult flies, we
used a 1:3 mixture of yeast extract and sugar. The Musca
strain was reared at room temperature (RT) (22 ± 2 °C) on food
composed by 500 g of wheat bran, 75 g of wheat flour, 60 g of
milk powder, 25 g of yeast extract, 872 ml of water and
Dev Genes Evol (2016) 226:245–256
18.85 ml of Nipagin (the same as for Ceratitis food). Adult
Musca flies were kept with sugar water only.
Experimental design
Two days before the experiment, the Drosophila strain was
moved to 25 °C and females oviposited on an apple-agar
plate for 1 h in an egg-collecting chamber. Collected eggs
were left at 25 °C for 24 h for development, whereupon the
first-instar larvae were transferred to 50 ml vials with 15 ml
of standard culture medium in two experimental sets. One
set was moved to 18 °C, and the other was left at 25 °C.
Both sets had three vials with two experimental densities:
25 (low density) and 300 (high density) larvae. To estimate
pupal size, we randomly collected 30 pupae from both
densities at 25 °C. Each pupa was weighed on a Sartorius
CP225D scale, photographed and placed in a separate vial
with wet food for the further development. Sex of the individuals was determined after eclosion.
Collection of Ceratitis eggs was carried at 28 °C. Flies
laid eggs in water through a net for 1 h. Collected eggs
were placed on the larval food and kept at 28 °C for
22 h. First-instar larvae were transferred then to a small
Petri plate (55 mm in diameter) containing 15 ml of the
larval food in two sets. One set was moved to 18 °C, and
the other was left at 28 °C. Both sets had two plates with
three experimental densities each: 25 (low density), 100
(middle density) and 300 (high density) larvae.
Musca eggs were collected at RT for 1 day in the larval
food and left to develop. Next day, all larvae were removed
from food, and only larvae hatched within the next hour were
counted and transferred to 50 ml vials with 5 g of the wet
larval food. The procedure with the newly hatched larvae
was repeated several times to get a number required for two
experimental sets, one of which was moved to 18 °C and the
other was left at RT. Both sets had three vials with three experimental densities each: 10 (low density), 20 (middle density) and 40 (high density) larvae.
All Ceratitis and Musca pupae were weighed,
photographed and placed in separate vials with a wet
sponge for the further development. The water sponge
was refreshed on every second day until flies emerged.
From each density and temperature, at least 5 males and
5 females of Musca, together with 10 males and 10
females of Drosophila and Ceratitis were randomly taken, anesthetized and photographed from the dorsal side.
Both wings and the metathoracic right leg were dissected, mounted on a microscope slide, embedded in the
Roti®-Histokitt II (Roth) and photographed as well.
All images were taken under a Leica MZ16 FA stereo
microscope with a QImagingMicroPublisher 5.0 RTV
Camera.
247
Size measurements
The pupal volume was calculated with the ellipse volume equation PV = 4/3π*PL*(PW/2)2, where PV is the
pupal volume; PL is the pupal length, measured as a
distance from the most apical to the most distal point
of the pupa; and PW is the pupal width, measured in
the widest centre region of the pupae. All distances
were measured with an accuracy of ±5 μm. Images with
broken or deformed during preparation samples were
excluded from the analysis. The pupal size was computed as the principal component 1 (PC1) of the pupal
weight and volume with the principal component analysis (PCA) implemented in R (R Development Core
Team 2008). Additionally, for each fly, we measured
the tibia length and the thorax length, defined as the
distance from the anterior margin of the thorax to the
posterior tip of the scutellum.
To estimate wing parameters, we digitized 11 landmarks
on Drosophila and 13 landmarks on Ceratitis and Musca
wings with tpsUtil (Rohlf 2004) and tpsDig2 (Rohlf 2010)
and processed them with MorphoJ (Klingenberg 2011).
Please note that this landmark dataset was only used to
estimate wing size parameters in this study. A thorough
analysis of wing shape is part of another study (Siomava
et al., in preparation). Every wing was described with
three metrics obtained from raw landmark coordinates:
the wing length (distance from landmark 1 to landmark
10), the wing width (distance from landmark 8 to landmark 9) and the wing area, restricted by landmarks 1, 3
and 11 (Fig. 1, red line). For Ceratitis, wing area was
manually measured using Analysis tools of Adobe
Photoshop CS5. For two other species, we manually measured the area for 35 wings randomly taken from different
groups. For the same set of wings, we also computed the
wing centroid size (WCS) (Bookstein 1996), measured as
the square root of the sum of squared deviations of landmarks around their centroid. Using MorphoJ, we extracted
the WCS from landmarks 1–10 for Drosophila and landmarks 1–13 for Musca (Fig. 1). To check correlation of
the WCS and the manually measured area, we performed
Spearman’s rank correlation test and found a high correlation between the two parameters: 0.993 for Drosophila
and 0.992 for Musca (p < 0.05 for both). Therefore, for
these two species, we computed the wing area from the
WCS using the deduced correlation equations (for
Drosophila wing area = 1348 × WCS − 1125; for Musca
wing area = 3185 × WCS − 8674). All numbers obtained
for the right and left wings were averaged for each individual. If only one wing was available for a fly, it was
used as a mean. Finally, for each fly, we computed the
body size coefficient as the PC1 of thorax and tibia
lengths with the PCA implemented in R.
248
Dev Genes Evol (2016) 226:245–256
1 mm
L5
PCV
11
12
Wi
7
7
L5
Anal part
8
2
ACV
8
Wing length
5
11
6
PCV
7
13
12
Anal part
10
h
1
t
10
L4
id
6
L4
w
ACV
5
g
Wing length
10
in
L4
ng
PCV
Anal part
h
Wing length
6
9
L3
L2
4
W
5
dt
11
ACV
L1
3
L3
4
wi
2
1
c
9
L2
2
L3
4
1
3
b
9
L2
widt h
L1
3
Wing
a
L1
L5
8
13
Fig. 1 Wing outlines with landmarks and measurements. Wing length,
width and outline for the manually measured area (red line) and
landmarks from 1 to 11 in D. melanogaster (a) and from 1 to 13 in C.
capitata (b) and M. domestica (c). The landmark coordinates were used to
extract wing length, width and WCS. ACV corresponds to anterior cross
vein, PCV posterior cross vein, L2–L5 longitudinal veins
Computation of allometric vectors
Results
Usually, scaling relationships are modelled with the allometric equation y=axb, where y and x are measurements for two given traits and b is the allometric coefficient that shows relationships between the traits
(Huxley 1924; Huxley and Tessier 1936). Log transformation of the allometric equation results into a linear
relationship: log(y)=log(a) + b × log(x), where b is a
slope and log(a) is an intercept. Here, the allometric
coefficients for the wing area, thorax and tibia were
computed for all possible combinations of the analysed
conditions using method described in (Shingleton et al.
2009) using the pca() function in the labdsv package in
R to process log-transformed data for each dataset. This
method relies on multivariate log-transformed data and
results in allometric vectors. Isometry is present for a
given trait when the allometric vector equals 1/√n with
n being the number of variables. For our analysis, isometry is given when the allometric vector equals 0.577
(n = 3). Accordingly, hypo- or hyperallometry occurs
when the allometric vector is <0.577 or >0.577,
respectively.
Sexual size dimorphism
Statistical analysis
Statistical analysis of size changes was performed with
STATISTICA 12 (StatSoft Inc. 1997). Since our data was not
normally distributed and the sample size was low in some
cases (i.e. for Musca and the effect of the sex for all species),
we used non-parametric statistical tests. Thus, correlation between the WCS and the manually measured wing area, as well
as pairwise correlation between the measured body parts, were
tested with Spearman’s rank correlation coefficient. Effects of
the rearing temperature and density on size and their significance were checked with Mann-Whitney U test.
D. melanogaster is known to exhibit a clear sexual dimorphism for various body parts (Badyaev 2002; Stillwell et al.
2010). Therefore, we first combined all measurements across
rearing conditions and tested whether they vary in size between male and female flies in the three species.
We found a clear sexual dimorphism in Drosophila with
females being significantly larger than males for all compared
variables. In Ceratitis, we did not find any difference in the
thorax length and the wing area. However, pupae were larger
and the tibiae were shorter in females. Similar to Ceratitis, we
found significantly longer tibiae in males of Musca.
Additionally, Musca females had larger wings compared to
males, while we did not detect any difference in size of pupae
and in the thorax length (Fig. 2). Hence, we confirm a clear
sexual dimorphism of pupal size and body parts for
Drosophila. In contrast, Ceratitis and Musca did not exhibit
such a uniform sexual dimorphism in our survey.
Alterations of size in response to environmental cues
Next, we tested whether different rearing conditions, i.e. temperature and larval density, influence the size of the measured
body parts and the pupae.
Comparing different rearing temperatures, we found that
Ceratitis flies raised at high temperature were smaller in all
measured parameters, while in Drosophila, significant effects
of temperature were present only for wing size. In Musca, the
response was the opposite in the case of thorax and tibia
lengths, while pupal size and wing area remained unaffected
by temperature (Fig. 3a).
Varying larval densities resulted in a steady and identical
response in the three species. All adult body parts as well as
Dev Genes Evol (2016) 226:245–256
249
12
10
4
−1
−5
NS
U=2757.5
8
Wing area, sq mm
2
0
0
1
1
0
U=1056
6
U=1159
3
U=775
5
*
U=0
***
U=0
−2
2
−2
−10
−1
Pupae size, PC1
*
NS
***
2
Fig. 2 Sexual size dimorphism in
pupae and different traits.
Statistical significance of
difference was checked with
Mann Whitney U test and shown
as NS non-significant at p = 0.05,
*p < 0.05, **p < 0.005 and
***p < 0.0001. Mdn refers to the
median of each group
Male Female
Mdn= -1.66 1.10
n= 13
17
Male Female
0.19 0.64
47
43
-0.06 -0.41
56
52
Ceratitis
Musca
Male Female Male Female Male Female
1.25
83
1.61
85
Drosophila
6.00
81
5.79
77
Ceratitis
8.26
55
8.84
52
Musca
***
U=882
2.0
***
***
U=2136
1.5
Tibia length, mm
2.5
NS
1.0
3.0
2.0
NS
U=1174
U=2732
1.5
Thorax length, mm
Drosophila
Male Female
***
U=456
0.5
1.0
U=1777
Male Female Male Female Male Female
0.86
83
1.02
83
2.02
79
Drosophila
*
U=1090
NS
U=1357
Ceratitis
b
2.84
56
Male Female Male Female Male Female
2.71
52
0.60
83
Musca
Drosophila
***
U=65
***
U=2
0.64
85
1.43
81
1.38
80
Ceratitis
1.87
56
1.74
52
Musca
***
U=72
10
Ceratitis
Musca
8
Low High Low Mid High Low Mid High
1.52 1.26 6.47 6.39 5.17 9.89 8.54 7.64
82
86
56
51
51
28 37
42
Drosophila
Ceratitis
Musca
***
U=167
2.0
3.0
2.5
Thorax length, mm
***
U=251
***
U=1373
***
U=447
***
U=608
18°C
0.95
81
25°C
0.94
85
Drosophila
18°C
2.11
88
28°C
1.95
70
Ceratitis
18°C
2.71
48
RT
2.87
60
Musca
0.5
0.5
Mdn=
n=
6
2
Low Mid High
1.79 0.01 -0.67
29
37 42
1.0
*
U=2815.5
4
2
1
0
−1
−2
Low Mid High
1.26 0.51 -1.61
30
30 30
***
U=95
2.0
1.5
Tibia length, mm
NS
U=3013
0
Mdn=
n=
Musca
***
U=929.5
1.0
3.0
2.5
2.0
Thorax length, mm
RT
8.69
60
**
U=915
1.0
1.5
18°C
8.54
47
***
U=1420
1.5
Ceratitis
**
U=983
***
U=1012
−5
−10
28°C
5.45
70
Tibia length, mm
Drosophila
18°C
6.51
88
***
U=394
1.0
Musca
25°C
1.29
86
Wing area, sq mm
5
10
8
4
2
18°C
1.50
82
RT
0.03
60
2.0
Ceratitis
18°C
-0.04
48
1.5
18°C 28°C
0.82 0.15
45
45
Pupae size, PC1
1
0
−1
***
U=1948
−2
−10
Mdn=
n=
***
U=533.5
6
2
Wing area, sq mm
5
0
−5
Pupae size, PC1
3
3
*
U=753
12
a
2.00
79
12
Mdn=
n=
18°C
0.63
82
25°C
0.61
86
Drosophila
18°C
1.46
88
28°C
1.34
73
Ceratitis
18°C
1.75
48
RT
1.86
60
Musca
Fig. 3 Size variation in pupae and different traits in response to changing
environmental conditions. Response to different larval density (a) and
rearing temperatures (b). RT means room temperature. Low, mid and
high corresponds to the low, middle and high densities of larvae. Error
Mdn=
n=
Low High Low Mid High Low Mid High
0.98 0.89 2.12 2.03 1.91 3.10 2.84 2.62
81
85
54
51
53
29 37
42
Drosophila
Ceratitis
Musca
Low High Low Mid High Low Mid High
0.64 0.58 1.45 1.43 1.31 2.01 1.82 1.69
82
86
56 51
54 29
37
42
Drosophila
Ceratitis
Musca
bars show the max and min values. Statistical significance was checked
with Mann Whitney U test and shown as NS non-significant at p = 0.05,
*p < 0.05, **p < 0.005 and ***p < 0.0001. Mdn refers to the median of
each group
250
Dev Genes Evol (2016) 226:245–256
pupal size were smaller in crowded conditions, while low
density resulted in bigger flies. The observed size difference
between the density extremes was statistically significant for
each species (Fig. 3b).
Response to changing environmental conditions is sex
dependent
Then, we asked whether the influence of the different rearing
conditions on organ and pupal size was sex dependent. At low
temperature (18 °C for every species), the increase in length
and width of wings is the same for both sexes when density
decreases (blue and pink numbers in Table 1). Accordingly,
the male-female differences in wing width and length remain
constant at both rearing densities (black numbers in Table 1) at
18 °C. In contrast, at higher temperature, the length and width
of female wings changed more in response to the rearing density than those of males (blue and pink numbers in Table 1).
For instance, the wing width increased twice as much in
Drosophila females than in males (i.e. a change of 0.05 mm
Table 1
in males vs. 0.1 mm in females, blue and pink numbers in
Table 1) when flies are raised at 25 °C. However, the relative
increase differs between species. In Drosophila, wing size
increased by ≈50 %, while in Ceratitis, the increase was lower,
with ≈30 %. In line with the increase, the relative male-female
differences in wing width and length were more pronounced
in non-crowded conditions (black numbers in Table 1) at high
temperature. Interestingly, the sex-specific response to environmental growth conditions was only observed for the linear
wing measurements and not for wing area (data not shown),
suggesting significant changes in the overall wing shape in
these species in different conditions (see also Siomava et al.,
in preparation).
Besides the linear wing measurements, we observed a sexdependent response for thorax length in Ceratitis. The increase of female and male thorax length was very similar with
decreasing density at 18 °C (i.e. a change of 0.18 mm for
males and 0.19 mm for females; blue and pink numbers in
Table 1). However, at 28 °C, the increase of thorax length at
lower rearing density was much more pronounced in females
Sexual size dimorphism depends on the environment
Mean absolute measurements and difference between males and females in wing length, wing width and thorax length. Comparison of male (blue
numbers) and female (violet numbers) traits between groups with varying density. RT refers to room temperature
Dev Genes Evol (2016) 226:245–256
251
compared to males (i.e. 0.16 mm differences for males
compared to 0.24 mm in females, blue and pink numbers in
Table 1). In Musca, the change in thorax length was similar for
males and females at both temperatures (18 °C and room
temperature). Also, no effect of the sex on thorax length in
different rearing conditions was observed in Drosophila
(Table 1), what was supported by the finding that the strain
we used in this study seems to be insensitive to the changing
rearing temperature in thorax length.
In contrast to wing and thorax variation, tibia length did not
show sex-dependent changes to rearing conditions. At both
temperatures and in all analysed densities, changes in size were
similar for males and females in all species (data not shown).
correlation dropped to 72–75 %. In Ceratitis, correlation coefficients were much lower in general and especially between
the measured traits and the pupal size (PC1, see ‘Materials and
methods’ section for details). Since the correlation of all body
parts and pupal size was rather low in Ceratitis, we combined
measurements of thorax and tibia with wing area and computed a single body size coefficient (BSC) out of these three. The
correlation of the pupal size with this BSC became 77 %, and
correlation coefficients with other body parts increased as
well. In case of Musca, computation of the BSC resulted in
the highest correlation with the pupal size (95 %) and correlation of the BSC with other parameters remained ≥88 %
(Table 2).
Evolutionary and static allometries for thorax, tibia
and wing size
Discussion
Allometries describe scaling relationships between given traits
at different evolutionary or developmental stages. They are
usually classified into three types, which are ontogenetic, static and evolutionary (Cheverud 1982; Schlichting and
Pigliucci 1999). In this study, we focused on the static and
evolutionary allometries. The first describes the relative size
of thorax, tibia and wing among individuals of the same species, while the latter compares relative size of organs among
different species at the same stage of development.
We found that the allometric coefficient varied among traits
and environmental conditions but remained similar between
species (Fig. 5). In all conditions, thorax grew slightly
hyperallometrically relative to the absolute body size
(b < 0.577), while wing area showed a strong hyperallometric
relationship (b > 0.577). Growth of tibia was close to isometric
at different rearing densities. Also, at different rearing densities,
the observed scaling relationships were temperature independent for all three species. In contrast, the allometric coefficient
for different temperature regimes was density dependent in
Ceratitis and Musca and to a lesser extent in Drosophila.
Estimators of the absolute body size in C. capitata and M.
domestica
Thorax, tibia and wing sizes are widely used as estimators of
the Drosophila body size, because they are known to be highly correlated among each other and with pupal size, which is
supposed to best represent the body size because it is a stage
when flies stop feeding and, therefore, do not increase in mass
and size (Shingleton et al. 2008).
To define adult body parts that can easily be used to estimate the whole body size of Ceratitis and Musca flies, we
computed Spearman’s rank correlation coefficients for every
pair of the analysed measurements. For most body parts, we
found a high correlation (>80 %) in Musca. Only for comparisons between the tibia length and the wing measurements, the
SSD in Drosophila, Ceratitis and Musca
Sexual dimorphism is a phenotypic difference between male
and female individuals of the same species. A sexual dimorphism can be observed in a variety of traits such as body or
organ size, body structure and shape, pigmentation or behaviour. In this study, we focus on comparison of size differences
of traits known to represent body size in flies. Although body
and organ size is highly variable in animals, a number of
common trends have been observed in large groups. For example, most invertebrates have a female-biassed sexual size
dimorphism (SSD) with males being smaller than females
(Shine 1979, 1994; Head 1995; Teder and Tammaru 2005).
It holds true for D. melanogaster as well (Badyaev 2002;
Stillwell et al. 2010), and our results confirm this. Here, we
showed that the size differences are already evident in late
instars, which are not feeding and do not grow anymore, and
reflected in the observed sexual dimorphism in pupal size.
At the intraspecific level, the SSD in whole body size (Table
1, pupal size) and in individual body parts (Table 1) was highly
variable in all three studied fly species. Interestingly, SSD was
not uniform even within species but depended on external cues.
It has been shown that in most species that have a femalebiassed SSD, female size increases more in comparison to
males, when flies are compared between different environmental conditions (Santos et al. 1994; Teder and Tammaru 2005).
Our results demonstrate that this phenomenon is also present in
Ceratitis and Musca, which belong neither to the femalebiassed nor to the male-biassed system. These flies rather represent a mixed system of the SSD where both sexes have similar size and certain body parts appear to be larger in males,
while others are larger in females.
It has been proposed that different sensitivity of females and
males to environmental conditions could explain the disproportional growth (Teder and Tammaru 2005). Here, we demonstrate
that certain conditions, e.g. temperature, affect sensitivity of flies
252
Table 2
Dev Genes Evol (2016) 226:245–256
Correlations of pupal size, body size coefficients (BSC) and different traits in C. capitata and M. domestica
Spearman’s rank correlation coefficients were computed for each pair of variables (p < 0.05). BSC refers to body size coefficient
All pairwise comparisons are significant with p < 0.05
to varying density regimes during larval development. When
flies were raised at cold temperature, size changes were the same
at different rearing densities for both sexes. However, warm temperature stimulated sex-dependent sensitivity (Fig. 4). This sexdependent sensitivity seemed to be organ or tissue specific, since
we observed it for thoraces and wings, but not for the tibia measurements. Intriguingly, this observation recapitulates the common origin of the dorsal thorax and the wings from the same
wing imaginal disc. In contrast, the legs develop from a different
imaginal disc during larval development (Madhavan and
Schneiderman 1977; Cowley and Atchley 1990). These data
suggest that different imaginal discs can react differentially to
varying environmental conditions in a sex-dependent manner.
Body and organ size: response to changing rearing
temperature
Most poikilotherms follow Bergmann’s rule (Ray 1960), and we
confirm a similar trend for Ceratitis (Navarro-Campos et al.
2011) and Drosophila (French et al. 1998; Robinson and
Partridge 2001). In both species, body size and all measured
body parts decreased in size with increasing rearing temperature.
While this trend was highly significant for all traits in Ceratitis, in
Drosophila, changes in thorax and tibia length were not significant but wings remained temperature sensitive and followed
Bergmann’s rule. Although these two adult structures develop
from the same larval tissue, it has been shown that timing of
temperature sensitivity differs in developing thoraces and wings.
The rearing temperature has a cumulative effect on thorax size
throughout larval life, while wing size is only affected during the
third larval instar and most of the pupal stage (Pantalouris 1957;
French et al. 1998). Intriguingly, the Drosophila strain used in
our survey (w1118) was kept at 18 °C for more than 20 years.
These cold adapted flies might have accumulated a temperature
insensitivity of the leg imaginal discs and the part of the wing
imaginal disc that contributes to the dorsal thorax. Such a striking
alteration is also known for body size regulation in the nematode
Caenorhabditis elegans, where a single nucleotide substitution
results in worms that are insensitive to the rearing temperature
(Kammenga et al. 2007).
Since the compliance of Bergmann’s rule seems to be
retained by selection rather than physiological or thermodynamic constraints (Scheiner and Lyman 1991), the phenotypic plasticity of body size in response to temperature is heritable and can
Dev Genes Evol (2016) 226:245–256
Cold temperature
Warm temperature
high density
low density
high density
low density
♂
Number of individuals
Number of individuals
♂
x1
♀
x
be artificially selected (Scheiner and Lyman 1989; deMoed et al.
1997). This means that rearing flies at constant laboratory conditions for a long time (i.e. 20 years at 18 °C) might allow the
accumulation of mutations that result in the elimination of the
temperature sensitivity in this strain. Thus, we believe that D.
melanogaster w1118 warrant further investigations, which can
add to our understanding of how wings attain their final size
and how wing networks differ from those of thorax.
Although there are known exceptions from Bergmann’s
rule (Atkinson 1994; Kingsolver and Huey 2008), our finding
that Musca shows a positive thermal reaction norm for the size
of all measured body parts is unexpected because natural populations of the species follow Bergmann’s rule with a negative
thermal reaction norm (Bryan 1977; Alves and Bélo 2002). In
contrast to previous publications, we used a laboratory strain
that was recently collected in Italy (see ‘Materials and
methods’ section for details). While room temperature (22
±2 °C) is well-suited to keep stocks of this strain, a rearing
temperature of 18 °C could be too stressful for the flies, especially at low densities. Thus, we believe that the observed
opposite reaction might be induced by cold stress, and in this
condition with low survivorship, the temperature-size rule
cannot be properly applied (David and Clavel 1967;
Kingsolver and Huey 2008). Another possible explanation
can be a shift in the diet because the strain was only recently
established for constant rearing under laboratory conditions.
During this time, flies were provided only with sugar water as
food, and this change from natural to laboratory nutrition
might affect the temperature-size rule (Diamond and
Kingsolver 2010).
x1
x
Size
high density
low density
Size
x1
Size
high density
low density
♀
Number of individuals
x
Number of individuals
Fig. 4 Scheme of a size increase
with respect with a sexual size
dimorphism. Scheme of a size
increase of male (blue) and
female (violet) organs (e.g. wings
or thoraces) at different densities
(high density—solid line, low
density—dashed line) for cold
and warm temperatures. The
black solid and dashed line and
the x and x1 labels on the x-axis
represent the mean organ size at
the respective density. The red
arrows indicate increase in organ
size
253
x
x2
Size
Body and organ size: response to changing rearing
densities
We found a clear influence of the rearing density on the final
body size and the size of individual body parts (Fig. 3b). For
Drosophila, we analysed only two conditions that are known
to be density extremes for this species and we confirm the
previously observed trend that flies are smaller when they
are raised at higher densities (Santos et al. 1994). We found
the same trend for Ceratitis and Musca. While body and organ
size in Ceratitis remained rather stable between the low- and
mid-density conditions, the size of all measured traits decreased clearly when the middle- and high-density conditions
were compared, suggesting a non-linear relationship between
density and size. However, in Musca, we observed a linear
relationship for thorax and tibia length, while pupal size and
wing area already showed the most obvious response between
the low- and middle-density conditions (Fig. 3b). Several
previous studies attempted to test, whether reduction in
size is linearly dependent on density or not. The first
results suggested that body size, measured as dry weight
of adult flies, was non-linearly reduced in increasing
density (Miller and Thomas 1958). Later, Santos et al.
(1994) found that body size, measured as thorax length,
had a linear response to changing density. Our data
showed that the density response was highly variable
and that different body parts had a potential to change
with a different rate. Thus, both linear and non-linear
relationships are likely to occur in different organs and
different fly lineages.
254
Dev Genes Evol (2016) 226:245–256
Until now, there is no clear model that fully explains the
mechanism of body size regulation under conditions of different crowding. It has been proposed that this phenomenon
could be explained by a possible pheromone regulation
(Shingleton et al. 2009) and regulation via oxygen level in
the media with hypoxia occurring at high densities
(Biddulph and Harrison 2014). Additionally, increased concentration of waste products as a consequence of crowding
can also be recognized and interpreted by larvae during
growth. The data obtained in this study represents an excellent
starting point to further explore the genetic and physiological
mechanisms underlying organ-specific size regulation in response to changing rearing densities in a comparative way.
Body size estimators and growth scaling
Body size has been shown to correlate with fundamental life
history traits, mating behaviour and mating success (Burk and
Webb 1983). Therefore, it is of general interest to estimate
body size of individuals originating from different wild populations to infer various environmental and ecological conditions. This is even more important for insects that represent
serious pests, such as the three dipteran species studied here.
The final adult body size of a fly is a consequence of
growth during the feeding larval stages, while ‘wandering
larva’ and the immobile pupal stage do not increase in mass
and volume anymore. Therefore, these two stages are good
estimators of body size (Churchill-Stanland et al. 1986;
Shingleton et al. 2008; Stillwell et al. 2011). However, in wild
populations, the access to different stages of the lifecycle is
highly limited. As an alternative, different parts of the adult
body are often used as size estimators in different studies. For
instance, it has been extensively shown that adult thorax, tibia
and wing size are highly correlated with the absolute body size
in Drosophila (Cavicchi et al. 1989; Pitnick and Markow
1995; de Moed et al. 1997; Kacmarczyk and Craddock
2000). In fact, these correlations were even extrapolated to
Effect of temperature
Effect of temperature
(at low density)
(at highdensity)
Drosophila
Ceratitis
Musca
Fig. 5 Allometric vectors for different traits. Isometric growth occurs
when the allometric vector equals 0.577. Allometric vector traits with
allometric vectors <0.577 scale hypoallometricaly with the overall body
Wing area
Thorax
Tibia
n=86
n=51
n=42
n=54
n=42
Tibia
n=86
Thorax
n=85
n=53
n=42
n=82
n=56
n=29
Wing area
n=82
n=56
n=28
n=81
n=54
n=29
Tibia
n=86
n=70
n=60
Thorax
n=86
n=73
n=60
0.577
n=85
n=70
n=60
allometric vector
Effect of density
other dipterans in some cases (Gleiser et al. 2000; NavarroCampos et al. 2011). Here, we show that body part correlations in other dipteran flies are not always consistent with
correlations known from Drosophila. For instance, tibia
length, which is widely used in Drosophila studies as a body
size estimator, exhibits a low correlation with other body size
measurements in Ceratitis. Similarly, tibia length alone is of
limited value to estimate body size of Musca flies.
Additionally, we found different scaling relationships
between the size of individual traits and the overall body
size. A study on scaling relationships in D. melanogaster
Oregon R indicated that changes in traits were not always proportional with the absolute body size and the
relative size of each trait could change dramatically in
different rearing conditions (Shingleton et al. 2009). We
confirm these results in our Drosophila strain and provide additional evidence supporting the previous suggestion by Shingleton et al. that different genotypes respond
differently to the same environmental variables.
However, although we see differences in allometries
among different species, we still find consistent trends.
For instance, the effect of variable temperature seems to
be masked at high densities, suggesting that rearing density is a more crucial parameter than temperature. This
assertion is further supported by the steady and uniform
size changes under conditions of variable density, while
size changes resulting from different rearing temperatures
are more unstable and less universal (Fig. 5).
Altogether, our data suggests that pupal traits (volume
or dry weight) should be used since they represent body
size best and remain unaffected by the environment.
However, since this is rarely possible in field studies,
adult body parts are usually used to estimate body size.
We show that the use of a single body part to estimate the
entire body size might result in wrong conclusions.
Therefore, we propose to use a body size coefficient computed from measurements of several traits.
Wing area
size and traits with allometric vectors >0.577 scale hyperallometrically.
Error bars indicate ±SD
Dev Genes Evol (2016) 226:245–256
Acknowledgments We thank Y. Wu and L. Beukeboom for providing
the Musca flies. This work has been funded by a German Academic
Exchange Service (DAAD) fellowship #A/12/86783 to NS, the
Göttingen Graduate School for Neurosciences, Biophysics, and
Molecular Biosciences (GGNB) and the Volkswagen Foundation (project
number: 85 983; to NP). Special thanks to the two anonymous reviewers
for their helpful comments on the previous versions of the manuscript.
References
Alves SM, Bélo M (2002) Morphometric variation in the housefly, Musca
domestica (L.) with latitude. Genetica 115:243–251
Anderson WW (1966) Genetic divergence in M. Vetukhiv’s experimental
populations of Drosophila pseudoobscura. 3. Divergence in body
size. Genet Res 7:255–266
Atkinson D (1994) Temperature and organism size—a biological law for
ectotherms? Adv Ecol Res 3:1–58
Badyaev AV (2002) Growing apart: an ontogenetic perspective on the
evolution of sexual size dimorphism. Trends Ecol Evol 17:369–378
Beadle GW, Tatum EL, Clancy CW (1938) Food level in relation to rate
of development and eye pigmentation in Drosophila melanogaster.
Biol Bull 75:447–462
Bergmann C (1847) Über die Verhältnisse der Wärmeökonomie der
Thiere zu ihrer Grösse. Göttinger Stud 3:595–708
Biddulph TA, Harrison JF (2014) Oxygen modulates density effects on
body size in Drosophila melanogaster. Society for Integrative and
Comparative Biology, Austin
Bookstein FL (1996) Biometrics, biomathematics and the morphometric
synthesis. Bull Math Biol 58:313–365
Bryan EH (1977) Morphometric adaptation of the housefly, Musca
domestica L., in the United States. Evolution 31:580–596
Burk T, Webb JC (1983) Effect of male size on calling propensity, song
parameters and mating success in Caribbean fruit flies, Anastrepha
suspensa (Loew) (Diptera: Tephritidae). Ann Entomol Soc Am 76:
678–682
Cavicchi S, Guerra D, Natali V, Pezzoli C, Giorgi G (1989) Temperature‐
related divergence in experimental populations of Drosophila
melanogaster. II. Correlation between fitness and body dimensions.
J Evol Biol 2:235–251
Cheverud JM (1982) Relationships among ontogenetic, static, and evolutionary allometry. Am J Phys Anthropol 59:139–149
Churchill-Stanland C, Stanland R, Wong TTY, Tanaka N, McInnis DO,
Dowell R (1986) Size as a factor in the mating propensity of
Mediterranean fruit flies, Ceratitis capitata (Diptera: Tephritidae),
in the laboratory. J Econ Entomol 79:614–618
Cohen S (1993) The development of Drosophila melanogaster. CSHLP,
New York, pp 747–841
Cowley DE, Atchley WR (1990) Development and quantitative genetics
of correlation structure among body parts of Drosophila
melanogaster. Am Nat 135:242–268
David JR, Clavel M-F (1967) Influence de la temperature subie au cours
du development sur divers characters biometriques des adultes de
Drosophila melanogaster Meigen. J Insect Physiol 13:717–729
de Moed GH, de Jong G, Schatloo W (1997) The phenotypic plasticity of
wing size in Drosophila melanogaster: the cellular basis of its genetic variation. Heredity 79:260–267
Demerec M (1950) Biology of drosophila. Wiley, New York
Diamond SE, Kingsolver JG (2010) Environmental dependence of thermal reaction norms: host plant quality can reverse the temperaturesize rule. Am Nat 175:1–10
DiAngelo JR, Bland ML, Bambina S, Cherry S, Birnbaum MM (2009)
The immune response attenuates growth and nutrient storage in
255
Drosophila by reducing insulin signaling. Proc Natl Acad Sci U S
A 106:20853–20858
Edgar BA (2006) How flies get their size: genetics meets physiology. Nat
Rev Genet 7:907–916
French V, Feast M, Partridge L (1998) Body size and cell size in
Drosophila: the developmental response to temperature. J Insect
Physiol 44:1081–1089
Gleiser RM, Urrutia J, Gorla DE (2000) Body size variation of the floodwater mosquito Aedes albifasciatus in Central Argentina. Med Vet
Entomol 14:38–43
Gokhale RH, Shingleton AW (2015) Size control: the developmental
physiology of body and organ size regulation. Wiley Interdiscip
Rev Dev Biol 4:335–356
Head G (1995) Selection on fecundity and variation in the degree of
sexual size dimorphism among spider species (class Aranea).
Evolution 49:776–781
Hewitt CG (1914) The house-fly, Musca domestica Linn.: its structure,
habits, development, relation to disease and control. University
Press, Cambridge
Huxley JS (1924) Constant differential growth-ratios and their significance. Nature 114:895–896
Huxley JS, Tessier G (1936) Terminology of relative growth. Nature 137:
780–781
Kacmarczyk T, Craddock EM (2000) Cell size is a factor in body size
variation among Hawaiian and nonHawaiian species of Drosophila.
Dros Inf Serv 83:144–148
Kammenga JE, Doroszuk A, Riksen JA, Hazendonk E, Spiridon L,
Petrescu AJ, Tijsterman M, Plasterk RH, Bakker J (2007) A
Caenorhabditis elegans wild-type defies the temperature-size rule
owing to a single nucleotide polymorphism in tra-3. PLoS Genet 3,
e34. doi:10.1371/journal.pgen.0030034
Khazaeli AA, Van Voorhies W, Curtsinger JW (2005) The relationship
between life span and adult body size is highly strain-specific in
Drosophila melanogaster. Exp Gerontol 40:77–85
Kingsolver JG, Huey RB (2008) Size, temperature and fitness: three
rules. Evol Ecol Res 8:703–715
Klingenberg CP (2011) MorphoJ: an integrated software package for
geometric morphometrics. Mol Ecol Resour 11:353–357
Koyama T, Rodrigues MA, Athanasiadis A, Shingleton AW, Mirth CK
(2014) Nutritional control of body size through FoxO-ultraspiracle
mediated ecdysone biosynthesis. Elife 3:1–20. doi:10.7554/eLife.
03091
Lutz F (1948) Field book of insects. G. P. Putnam’s Sons, New York
Madhavan MM, Schneiderman HA (1977) Histological analysis of dynamics of growth of imaginal discs and histoblast nests during larval
development of Drosophila melanogaster. Roux’s Arch Dev Biol
183:269–305
Miller RS, Thomas JL (1958) The effect of larval crowding and body size
on the longevity of adult Drosophila melanogaster. Ecology 39:
118–125
Mirth CK, Shingleton AW (2012) Integrating body and organ size in
Drosophila: recent advances and outstanding problems. Front
Endocrinol 3:49. doi:10.3389/fendo.2012.00049
Mirth CK, Tang HY, Makohon-Moore SC, Salhadar S, Gokhale RH,
Warner RD, Koyama T, Riddiford LM, Shingleton AW (2014)
Juvenile hormone regulates body size and perturbs insulin signaling
in Drosophila. Proc Natl Acad Sci U S A 111:7018–23
Navarro-Campos C, Martínez-Ferrer MT, Campos JM, Fibla JM, Alcaide
J, Bargues L, Marzal C, Garcia-Marí F (2011) The influence of host
fruit and temperature on the body size of adult Ceratitis capitata
(Diptera: Tephritidae) under laboratory and field conditions.
Environ Entomol 40:931–938
Nunney L, Cheung W (1997) The effect of temperature on body size and
fecundity in female drosophila melanogaster: evidence for adaptive
plasticity. Evolution 51:1529–1535
256
Oliveira MM, Shingleton AW, Mirth CK (2014) Coordination of wing
and whole-body development at developmental milestones ensures
robustness against environmental and physiological perturbations.
PLoS Genet 10, e1004408. doi:10.1371/journal.pgen.1004408
Pantalouris EM (1957) Size response of developing Drosophila to temperature change. J Genet 55:507–510
Partridge L, Ewing A, Chandler A (1987) Male size and mating success
in Drosophila melanogaster: the role of male and female behaviour.
Anim Behav 35:555–562
Peck LS, Maddrell SHP (2005) Limitation of size by hypoxia in the fruit
fly Drosophila melanogaster. J Exp Zool A Comp Exp Biol 303A:
968–975
Pitnick S, Markow TA (1995) Delayed male maturity is a cost of producing large sperm in Drosophila. Proc Natl Acad Sci USA 92:10614–
10618
R Development Core Team (2008) R: A language and environment for
statistical computing. Vienna, Austria: R Foundation for Statistical
Computing, http://www.R-project.org
Ray C (1960) The application of Bergmann’s and Allen’s rules to the
poikilotherms. J Morphol 106:85–108
Robinson SJW, Partridge L (2001) Temperature and clinal variation in
larval growth efficiency in Drosophila melanogaster. J Evol Biol
14:14–21
Rohlf FJ (2004) tpsUtil, File utility program, version 1.54. Department of
Ecology and Evolution, State University of New York at Stony Brook
Rohlf FJ (2010) tpsDig, Digitize landmarks and outlines, version 2.17.
Department of Ecology and Evolution, State University of New
York at Stony Brook
Santos M, Fowler K, Partridge L (1994) Gene-environment interaction
for body size and larval density in Drosophila melanogaster: an
Dev Genes Evol (2016) 226:245–256
investigation of effects on development time, thorax length and adult
sex ratio. Heredity 72:515–521
Scheiner SM, Lyman RF (1989) The genetics of phenotypic plasticity.
Heritability. J Evol Biol 2:95–107
Scheiner SM, Lyman RF (1991) The genetics of phenotypic plasticity II.
Response to selection. J Evol Biol 3:23–50
Schlichting CD, Pigliucci M (1999) Phentypic evolution—a reaction
norm perspective. Heredity 82:344–344
Shine R (1979) Sexual selection and sexual dimorphism in the Amphibia.
Copeia 2:297–306
Shine R (1994) Sexual size dimorphism in snakes revisited. Copeia 2:
326–346
Shingleton AW, Mirth CK, Bates PW (2008) Developmental model of
static allometry in holometabolous insects. Proc R Soc B 275:1875–
1885
Shingleton AW, Estep CM, Driscoll MV, Dworkin I (2009) Many ways to
be small: different environmental regulators of size generate distinct
scaling relationships in Drosophila melanogaster. Proc R Soc B
276:2625–2633
StatSoft, Inc (1997) Electronic statistics textbook. Tulsa, OK: StatSoft.
WEB: http://www.statsoft.com/textbook/stathome.html
Stillwell RC, Blanckenhorn WU, Teder T, Davidowitz G, Fox CW (2010)
Sex differences in phenotypic plasticity affect variation in sexual
size dimorphism in insects: from physiology to evolution. Annu
Rev Entomol 55:227–245
Stillwell RC, Dworkin I, Shingleton AW, Frankino WA (2011)
Experimental manipulation of body size to estimate morphological
scaling relationships in drosophila. J Vis Exp 56:3162. doi:10.3791/
3162
Teder T, Tammaru T (2005) Sexual size dimorphism within species increases with body size in insects. Oikos 108:321–334