Ethology Ecology & Evolution

ISSN: 0394-9370 (Print) 1828-7131 (Online) Journal homepage: http://www.tandfonline.com/loi/teee20

Orb-web inclination in Uloboridae spiders: the role

of microhabitat structure and prey capture

Suzana Diniz, João Vasconcellos-Neto & Vanessa Stefani

To cite this article: Suzana Diniz, João Vasconcellos-Neto & Vanessa Stefani (2016): Orb-web
inclination in Uloboridae spiders: the role of microhabitat structure and prey capture, Ethology
Ecology & Evolution, DOI: 10.1080/03949370.2016.1240109

To link to this article: http://dx.doi.org/10.1080/03949370.2016.1240109

Published online: 08 Nov 2016.

Submit your article to this journal

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=teee20

Download by: [University of Waterloo] Date: 08 November 2016, At: 06:54

Ethology Ecology & Evolution, 2016
http://dx.doi.org/10.1080/03949370.2016.1240109

Orb-web inclination in Uloboridae spiders: the role of

microhabitat structure and prey capture

SUZANA DINIZ1,2, JOÃO VASCONCELLOS-NETO2 and VANESSA STEFANI3,*

1
Department of Zoology, University of British Columbia, #2370-6270 University Blvd, BC,
V6T 1Z4 Vancouver, Canada
2
Departamento de Biologia Animal, Universidade Estadual de Campinas, Caixa Postal
6109, CEP 13083-970, Campinas, SP, Brazil
3
Laboratório de Ecologia, Comportamento e Interações da Universidade Federal de
Uberlândia, CEP 38400-902, Uberlândia, Brazil

Received 24 November 2015, accepted 15 September 2016

Spider webs are a good example of how phenotypic traits can be adjustable to
the physical structure of the environment. Orb-web spiders of the family Uloboridae
sometimes form aggregations inside webs of other spider species. In the present
study, we focused on the system of Uloboridae associated with webs of the spider
Aglaoctenus (Lycosidae). Because of the conformation of Aglaoctenus webs, the direc-
tion of prey flow in the central and peripheral positions of the Uloboridae aggrega-
tions is vertical and horizontal, respectively. The inclination of Uloboridae orb-webs,
in relation to the ground, tends to be closer to horizontal in the centre and vertical in
the periphery, whereas other characteristics of the orb-webs and spiders do not differ
between the positions. Structural differences between centre and periphery were not
enough to explain the inclination pattern of the orb-webs, at least regarding the
method we used to measure the micro-habitat structure in this study. Our results
support the view of intraspecific variability in orb-web inclination as a case of fora-
ging optimisation strategy and, possibly, of adaptive plasticity.

KEY WORDS : Aglaoctenus, prey interception, orb-weavers, Zosis, Philoponella, web

adjustment.

INTRODUCTION

The foraging and reproductive success of orb-web spiders depend largely on their
web (Schneider & Vollrath 1998). As a result of the energy demands related to the
production of silk and web building (Tanaka 1989; Craig 2003; Li & Lee 2004), the web
represents a high initial cost and cannot be quickly and easily modified once it is built
(Schneider & Vollrath 1998). Thus, the web efficiency in prey capture depends on
decisions made by the spider before its building. Web building has been traditionally

*
Corresponding author: Vanessa Stefani, Laboratório de Ecologia, Comportamento e Interações da
Universidade Federal de Uberlândia, CEP 38400-902, Uberlândia, Brazil (E‑mail: [email protected]).

© 2016 Dipartimento di Biologia, Università di Firenze, Italia

2 S. Diniz et al.

considered a stereotyped behaviour whose only significant variability was interspecific

(Reed et al. 1970; Levi 1978; Blackledge et al. 2009). A large number of studies demon-
strate how spiders are able to modify the geometry of their webs in response to a range
of different conditions, what includes the ability to adapt their webs to spatial con-
straints (Bonte et al. 2008; Blamires 2010; Herberstein & Tso 2011; Hesselberg 2015).
However, intraspecific and even intraindividual variations in orb-web geometry, size
and orientation have been detected in response to both abiotic (e.g. Eberhard 1971;
Biere & Uetz 1981; Vollrath et al. 1997; Bonte et al. 2008; Blamires 2010) and biotic
factors (e.g. Leborgne & Pasquet 1987; Herberstein et al. 2000; Harmer & Herberstein
2009; Blamires et al. 2011).
For example, the orb-weaver Parawixia bistriata increases both the web area and
mesh size in response to termite swarms (Sandoval 1994), while the orb-weaver
Telaprocera maudae builds significantly more elongated webs when space is limited,
which allows this spider to maximise the web area, and hence foraging potential, when
building in space-limited microhabitats (Harmer & Herberstein 2009). Therefore, web
geometry is closely related not only to its properties of prey interception and retention
(Bishop & Connolly 1992), but also to spatial constraints, such as reduction in the
capture area in small spaces (Ades 1986; Krink & Vollrath 2000) and web reorientation
from horizontal to vertical (Nogueira & Ades 2012). Furthermore, factors intrinsic to
the spider body, such as age (Rayor & Uetz 1993; Anotaux et al. 2014), leg loss (Vollrath
1987), intoxication (Witt 1971) and starvation (Vollrath & Samu 1997; Venner et al.
2000; Mayntz et al. 2009), can also influence the web geometry. In this context, there
are, basically, three key characteristics that contribute to the web geometry variability:
(1) the efficiency in prey capture, (2) the physical structure of the habitat, and (3) the
spider anatomical/physiological traits. Since these three characteristics co-occur in
nature, it is hard to assess, in field situations, what their respective roles are in web
adjustments. As orb-webs are a geometric representation of the spider’s foraging strat-
egy, with easily measurable parameters (Eberhard 1971), orb-weaver spiders provide a
good example for studies of how environmental heterogeneity triggers phenotypic
adjustments.
The family Uloboridae is composed of small cribellate spiders that build orb-webs
(Marples 1962; Lubin 1986). Many Uloboridae species of the genera Philoponella and
Zosis are facultative or strictly colonial (Lubin 1986), and some species also use barrier
or interception threads of other spiders’ webs as support for their own orb-webs (Lubin
1986; Rypstra & Binford 1995; Stefani & Del-Claro 2015). The threads of the host web
also make it easier for the orb-web spiders to forage because prey can first collide with
the threads of the host spider web before falling onto the orb-web (Uetz 1989; Rypstra &
Binford 1995). The web of the spider Aglaoctenus (Lycosidae) provides a series of
attachment points for small orb-web spiders and consists of a lower sheet with a funnel,
that houses the Aglaoctenus individual, and a set of interception threads right above the
sheet, where colonies of orb-web Uloboridae spiders are housed (Stefani & Del-Claro
2015). The capture of prey by Aglaoctenus occurs when insects collide with interception
threads and fall onto the sheet below, where they are captured (Stefani & Del-Claro
2015; authors’ pers. obs.). Thus, we expect prey flow to be mainly horizontal in the
periphery and vertical in the centre of the capture threads of Aglaoctenus.
The Uloboridae spiders that are associated with Aglaoctenus webs may be subject
to the direction difference in the prey flow at a micro-habitat scale (i.e. between the
centre and periphery of the interception area). Because orb-web spiders are able to
adjust web angle parameters in order to optimise prey interception (Chacón &
Habitat structure, prey capture, and web inclination 3

Eberhard 1980; Eberhard 1989; Bishop & Connolly 1992), we expect that Uloboridae
webs located in the centre will be differentiated from those in the periphery of inter-
ception threads. The spider could, for example, alter the web inclination and adjust it to
the predominant prey flow direction. Since webs that are perpendicular to the direction
of prey flow direction are more effective at intercepting prey and less subject to stresses
caused by the impact of prey (Chacón & Eberhard 1980; Eberhard 1986), we expect that
orb-web spiders in the centre will build webs with inclinations closer to horizontal than
those in the periphery. Alternatively, if the inclination of the orb-web is not adjusted to
prey flow, the spider could compensate by increasing the interception region of
the web.
Furthermore, spiders in the periphery could intercept more prey, since they are
more exposed and closer to the food source, which would cause a “shadow effect” over
the spiders in the centre (Lubin et al. 2001; Rao 2009). Shadow competition, however,
could be compensated for by the ricochet effect, which is caused by the escape of
usually larger prey from the more exposed webs and their retention in subsequent
webs (Uetz 1989). Thus, we expect that larger prey will be falling on orb-webs located
in the centre of the Aglaoctenus interception region, which would also increase the
advantage of building horizontal webs in the centre.
Therefore, we aim to answer the following questions: (1) In the interception
region of the Aglaoctenus host web, does the direction of prey flow tend to be horizontal
in the periphery and vertical in the centre? (2) Are the inclination and area of orb-webs
different between the centre and periphery? Do webs located in the centre tend to have
inclination closer to horizontal than those located in the periphery? (3) Do horizontal
webs in the centre and vertical webs in the periphery tend to intercept a higher
abundance and diversity of prey? Is there a difference in the mean size of intercepted
prey in the centre and periphery? (4) Does spider size differ between spiders at the
centre and periphery? (5) Is the web variability related to the direction of the prey flow
or to the physical structure of the microhabitat?

MATERIALS AND METHODS

Model organisms and study regions

The genus Aglaoctenus belongs to the family Lycosidae, specifically the subfamily
Sossipinae, and is represented by South American spiders that build sheet-webs to capture prey,
a foraging habit that is an exception in a primarily cursorial family (Vieira et al. 2007). Aglaoctenus
spiders build their webs with two distinct regions: (1) a lower region in the shape of a horizontal
sheet with a funnel that houses the spider, and (2) an upper region constructed of a series of
vertical threads that are used to intercept prey (Fig. 1A). Interception threads add architectural
complexity to the environment and frequently anchor small orb-webs of spider families such as
Uloboridae, Tetragnathidae and Theridiosomatidae in different positions in the host web (Fig. 1A)
(Alves-Costa & Gonzaga 2001; Stefani & Del-Claro 2015; authors’ pers. obs.).
We conducted this study in two areas, and with distinct Aglaoctenus species and their
respective associated Uloboridae. The distance between the two areas is about 710 km. We studied
two species of Uloboridae because they showed similar patterns of orb-web inclinations between
the centre and periphery of the host web, even though the species were located far apart from each
other and in very different environments. This suggests that the environment per se does not play
an important role in this specific case of orb-web architecture variability, but rather is an inherent
characteristic of the system of Uloboridade associated with Aglaoctenus webs. Area 1 – Perequê
nucleus of Cardoso Island State Park (25°4ʹ19.37ʹʹS, 47°55ʹ27.76ʹʹW), city of Cananéia, State of São
4 S. Diniz et al.

Fig. 1. — (A) Host web architecture of the spider Aglaoctenus (Lycosidae) indicating the presence of
vertical threads (interception region), which were used as anchor points for the orb-webs of spiders from
the families Uloboridae, Tetragnathidae and Theridiosomatidae. (B) Illustration showing how we defined
the peripheral and central regions of the Aglaoctenus host web. We inserted a carpenter’s ruler through
the interception region and determined the total width (W) of the interception region, which we defined
as the segment between the two most distal threads (filled points) touching the carpenter’s ruler. We
considered as periphery the distal segments comprising, on each side, 10% of the total width of the
interception region, while the centre was the remaining 80% between the two peripheral segments.

Paulo, southeastern Brazil. Cardoso Island has an area of approximately 15,100 ha and contains a
small portion of most of the vegetal formations that occur along the Brazilian coast (Sampaio et al.
2005). The Aglaoctenus species we studied was A. castaneus Mello-Leitão (1942), which is found in
low sandbank forest areas where its webs are usually built on the bromeliad Quesnelia arvensis
(Vellozo) Mez (Bromeliaceae). Interception threads of the A. castaneus web usually anchor small
orb-webs of Philoponella aff. fasciata (Mello-Leitão 1917) (Araneae Uloboridae). We recorded
observational data of this area in July, 2008. Area 2 – Bosque John Kennedy Urban Park (18°
39ʹ3.45ʹʹS; 48°10ʹ57.90ʹʹW) in the city of Araguari, state of Minas Gerais, southeastern Brazil. The
park has 11.2 ha of semideciduous mesophilic forest (Souza & Araújo 2005). The Aglaoctenus
species studied was A. lagotis Holmberg (1876), which builds its web next to the forest leaf litter
(for more details on the biology, reproductive behaviour and life history of A. lagotis, refer to
Stefani et al. 2011; Stefani & Del-Claro 2012, 2015). Orb-webs of Zosis sp. (Araneae Uloboridae) are
frequently found in the interception region of the web of A. lagotis. We recorded observational data
of this area in July 2013.

Data collection

Observational data. We selected host webs of Aglaoctenus that had at least one Uloboridae orb-
web associated with the interception region. We marked a total of 34 webs of A. castaneus and 25 of A.
lagotis. We also quantified the number of Uloboridae orb-webs in each position (periphery or centre) of
the interception region from each marked host web (Fig. 1A). In order to define what was periphery or
centre, we inserted a carpenter’s ruler through the interception region, parallel to the sheet-web, and at
the height of the Uloboridae orb-web that we were investigating (Fig. 1B). Thus, we considered
peripheral orb-webs the ones that were located in the region determined by the distal segments
comprising, on each side, 10% of the total width of the interception region from each Aglaoctenus
Habitat structure, prey capture, and web inclination 5

web (Fig. 1B). Since the total width of the Aglaoctenus interception region was defined as the segment
between the two most distal threads touching the carpenter’s ruler (Fig. 1B), the peripheral orb-webs
were obligatorily inside the interception region of the host web. Furthermore, we only included in the
study Uloboridae individuals that only used the Aglaoctenus threads as anchor points for their webs.
The central webs, in turn, were located in the region covered by the remaining 80% of the total width of
the interception region between the two peripheral segments (Fig. 1B). Additionally, to obtain a
measurement of the micro-habitat physical structure, we calculated the peripheral and central thread
density of all marked Aglaoctenus web. In order to measure the thread density, we also inserted a
carpenter’s ruler through the interception region parallel to the sheet-web (Fig. 1B), but at a standard
height of 30 cm above the sheet and at a random angle in relation to the north of a compass. We then
counted the number of threads that touched the peripheral and central segments of the carpenter’s
ruler. The central and peripheral segments were defined as described above and illustrated in Fig. 1B.
Subsequently, we divided the thread number by its respective segment length in order to obtain the
thread density in centre and periphery.
Regarding the data of the associated orb-webs, we only selected Aglaoctenus webs with
Uloboridae orb-webs in both the centre and periphery of the interception region, which included
10 webs of A. castaneus and 10 of A. lagotis. This allowed us to control for possible local effects of
the micro-habitat. For each Uloboridae orb-web, we measured the largest and smallest diameters
of the capture area and approximated it by the area of an ellipse. We measured the orb-web
inclination relative to the ground by means of a protractor positioned so that the inclinations 0
and 90°, for example, corresponded to webs that were parallel and orthogonal to the ground,
respectively. After measuring the web, we collected a total of 35 individuals of Philoponella and 22
of Zosis for identification and measurements. We then measured the length and width of the
cephalothorax for each Zosis and Philoponella specimen in a stereomicroscope and used the
measurements for calculation of cephalothorax area, which we approximated by the area of an
ellipse. We also measured the length, width and height of the abdomen of each spider for
calculating the abdomen volume, which we approximated by the volume of an ellipsoid. We
used the cephalothorax area as an estimate of the structural size of the spider because this
measurement is stable and directly correlated to the developmental stage of the spider
(Hagstrum 1971; Vollrath 1987). Conversely, the abdomen volume is a very variable measure,
since it expands when the spider acquires food (Vollrath 1987). Therefore, the abdomen size is,
in general, positively correlated with the spider feeding regime, weight and body condition (e.g.
Jakob et al. 1996; Uetz et al. 2002; Moya‐Laraño et al. 2003). Indeed, abdomen size is the
morphological trait most influenced by recent prey consumption (McNett & Rypstra 1997). We
also divided the values of web area by the respective values of cephalothorax area. We divided the
abdomen volume and web area by the cephalothorax area to correct for confounding effects
associated with the ontogenetic stage, since the Uloboridae samples mostly consisted of adults
females (for Zosis) or immature individuals (for Philoponella). However, we did not find any
difference between centre and periphery regarding the occurrence pattern of adult and immature
spiders (Fisher’s exact test, Philoponella: P = 0.402; Zosis: P = 1). We deposited the voucher speci-
mens in the arthropod collection of the Butantã Institute, in São Paulo, Brazil.

Experimental data. To assess whether the orb-web inclination affects prey interception
efficiency between the centre and periphery of the interception region from the Aglaoctenus web,
we designed an experiment with sticky traps that simulated Uloboridae orb-webs. Sticky traps have
previously been used in other studies to estimate the number and type of prey available for orb-web
spiders (e.g. Chacón & Eberhard 1980; Uetz & Biere 1980; Biere & Uetz 1981; Bishop & Connolly
1992). Despite the criticism related to the differences between prey intercepted by traps and those
actually captured by spiders, sticky traps can be useful indicators of the effect of web features on
prey interception, such as inclination, height and positioning (Castillo & Eberhard 1983). Thus,
sticky trap data can indicate selective pressures on the design and positioning of orb-webs (Castillo
& Eberhard 1983).
We carried out the experiment only in Area 2 with webs of the A. lagotis species, and during
the period of July to August of 2013. To ensure sticky traps were proportional in size to Uloboridae
webs, we quantified the area of 12 webs of Zosis and obtained a mean web area of 369 ± 25.4 cm2
6 S. Diniz et al.

(mean ± SD). Subsequently, we built the sticky traps using flat and circular structures made of
clear plastic with proportions similar to those of the mean area of Zosis webs. We covered each
web on both surfaces with a layer of entomological glue (Colly Química, Mombuca, São Paulo)
which is odorless and colourless. We then positioned the traps inside the interception region of the
Aglaoctenus webs and supported them with a wood stake. However, we found that some intercep-
tion threads were lost during the setting of the stakes, which could cause an unwanted effect on the
experimental results. To minimise this effect, 4 days passed after the initial placement of the stakes
before the sticky traps were attached to them. We established a period of 4 days, during a pre-
experimental phase, while assessing the time required by the spider A. lagotis to rebuild the
interception threads damaged by the stake. Thus, we counted the mean number of threads from
12 A. lagotis webs before and after the stakes settled and inspected the reconstruction of threads
over a period of 7 days. The host spider A. lagotis began the web rebuilding within 1 day after we
had settled the stakes, and the number of threads remained constant after the 3rd day.
We designed the experiment with two factors crossed: (1) trap inclination, with three
inclination levels relative to the ground: 0° (horizontal), 45° (inclined), 90° (vertical); (2) position
in the interception region of the Aglaoctenus web, with two levels: centre and periphery. We
determined the centre and periphery according to the methodology described for the observational
data (Fig. 1B). Therefore, the experiment included a total of six treatments for the traps: central
horizontal, central inclined, central vertical, peripheral horizontal, peripheral inclined, and per-
ipheral vertical. The experiment followed a completely randomised sampling design. Each web of
Aglaoctenus randomly received only one replicate of a single experimental treatment. Each treat-
ment, in turn, consisted of 12 replicates. We left the traps in the webs of Aglaoctenus for a period of
7 days. After this period, we collected the traps and transported them to the lab, where we
measured, counted and identified – at the order level – the arthropods retained in the traps. So,
at the end of the experiment, we obtained three variables regarding the potential prey intercepted
by the sticky traps: (1) prey abundance, which was the raw number of individuals intercepted per
trap; (2) prey richness, which was the raw number of species intercepted per trap; (3) prey size, the
average prey length (in cm) per trap.

Data analysis

Observational data. We performed four mixed models separately for each variable related
to the Uloboridae spiders and their respective orb-webs (i.e. cephalothorax area, abdomen
volume, web area, and inclination). So, each mixed model consisted of one out of four
Uloboridae variables as the continuous response variable. We included the position of the
orb-web (i.e. centre or periphery of the interception region of Aglaoctenus host web) and
Uloboridae species (i.e. Zosis or Philoponella) as fixed factors, while we included the identity
of the Aglaoctenus host web as the random factor of the model. As stated before, we only used
Aglaoctenus webs with associated Uloboridae spiders in both positions of the host web’s inter-
ception region (i.e. centre and periphery). We considered the mixed design appropriate because
the orb-webs were grouped into many different Aglaoctenus webs units in an unbalanced
design, since some host webs contained a different number of Uloboridae spiders between
the two positions (i.e. centre or periphery). The mixed model also allowed us to exclude
possible confounding effects of idiosyncrasy from each Aglaoctenus web, which consisted of a
typical random effect on the response variable.
For the thread density between centre and periphery of each Aglaoctenus host web, which
we considered a measure of physical structure of the micro-habitat, we performed a linear model
with random block design. So, we included the thread density as the continuous response variable,
the position (i.e. centre or periphery) and Uloboridae species (i.e. Zosis or Philoponella) as the main
factors, and the Aglaoctenus identity as the block effect. The design was balanced, as each
Aglaoctenus host web had just one value of thread density per position, which made us consider
the random block design the simplest and most appropriate design for this case. We analysed the
Habitat structure, prey capture, and web inclination 7

thread density separately from the Uloboridae traits because we obtained the data from different
Aglaoctenus host webs.
Additionally, we calculated the effect size between the positions periphery and centre for each
response variable (i.e. cephalothorax area, abdomen volume, web area, web inclination, and thread
density). We chose Cohen’s d as a measure of effect size, since it is a standardised mean difference
between two groups and it is commonly used in meta-analysis (Cohen 1988). This would allow us to
compare the magnitude of the position effect on different response variables, according to the thresholds
|d| < 0.2 “negligible”, |d| < 0.5 “small”, |d| 0.8 “medium”, otherwise “large” (Cohen 1992). We validated the
models by examining diagnostic plots of the residuals (Fox & Weisberg 2011), and performed Tukey
post-hoc tests for multiple comparisons among the factor levels. We performed all statistical analyses of
the observational data in R Studio software, version 0.98.507 (R Core Team 2014).

Experimental data. We performed one generalised linear model (GLM) for each response
variable from the sticky traps (i.e. prey abundance and size), with the trap inclination (i.e.
horizontal, vertical or inclined) and position in the interception region (i.e. centre or periphery)
as main factors. We used Poisson distribution to model the prey abundance, since it is count data.
For the prey size, we used Gaussian distribution, since it is count data. We also performed model
selection, through analysis of deviance, in order to assess the significance of the factors and their
interactions in each GLM. We calculated the effect size Cohen’s d in order to assess the magnitude
of the position effect, for each trap inclination, on prey size and abundance. We also validated the
model by examining diagnostic plots of the residuals (Fox & Weisberg 2011) and performed Tukey
post hoc tests for multiple comparisons among the factor levels.
We performed all the statistical analyses of the experimental data in R Studio software,
version 0.98.507 (R Core Team 2014).

RESULTS

Observational data

The position (i.e. centre or periphery) affected the web inclination of Uloboridae
orb-webs as a whole (t = 2.649, df = 35, P = 0.012; effect size: Cohen’s d = − 1.205 “large”;
Fig. 2A). However, the position effect was not uniform for both species, as the interac-
tion term was also significant (t = 3.721, df = 35, P < 0.001, Fig. 2A). The inclinations of
Zosis orb-webs were strongly affected by the position (Tukey post-hoc: Z = 6.885,
P < 0.001; effect size: Cohen’s d = − 2.65 “large”; Fig. 2A), as orb-webs located in the
periphery exhibited greater inclinations (mean ± SE = 66.67° ± 3.22, n = 12) than orb-
webs located in the centre (mean ± SE = 24 ± 3.56, n = 10). A similar, although weaker,
pattern was also found for Philoponella orb-webs (Tukey post-hoc: Z = 2.649, P < 0.001;
effect size Cohen’s d = − 0.67 “medium”; Fig. 2A) because the webs tended to exhibit
greater inclinations in the periphery (mean ± SE = 39.72° ± 3.91, n = 18) than in the
centre (mean ± SE = 26.76° ± 4.22, n = 17). The position main effect on orb-web area
(t = 2.328, df = 35, P = 0.026; effect size: Cohen’s d = − 0.464 “small”; Fig. 2B), spider
cephalothorax area (t = − 0.654, df = 34, P = 0.518; effect size: Cohen’s d = − 0.331
“small”; Fig. 2C), and abdomen volume (t = − 0.847, df = 35, P = 0.403; effect size:
Cohen’s d = 0.105 “negligible”; Fig. 2D) was small or negligible. Despite the significance
of the position effect on the orb-web area, the only difference pointed out by the post-
hoc test was between the peripheral orb-webs of Philoponella and the central ones of
Zosis (Tukey post-hoc: Z = 2.636, P = 0.04; Fig. 2B).
The effect of position on thread density of the interception region differed
between the two species of Aglaoctenus (F(1,89) = 26.379, P < 0.001; Fig. 3). For A. lagotis
webs (the host of Zosis orb-webs), the thread density in the periphery (mean ±
8 S. Diniz et al.

Fig. 2. — Web inclination (A); web area, corrected by the cephalothorax area (B); cephalothorax area (C);
and abdomen volume, corrected by the cephalothorax area (D) of Zosis and Philoponella individuals
located at the centre and periphery of the interception region from Aglaoctenus webs.
Habitat structure, prey capture, and web inclination 9

Fig. 3. — Thread density at the centre and periphery of the interception region from host webs
of Aglaoctenus castaneus and Aglaoctenus lagotis.

SE = 0.589 ± 0.034, n = 25) was higher than in the centre (mean ± SE = 0.308 ± 0.013,
n = 25) (Tukey post-hoc: P < 0.001; effect size Cohen’s d = − 1.635 “large”; Fig. 3).
Conversely, for A. castaneus webs (the host of Philoponella orb-webs), there was no
difference in thread density between the centre (mean ± SE = 0.936 ± 0.078, n = 34) and
the periphery (mean ± SE = 0.951 ± 0.071, n = 34) (Tukey post-hoc: P = 0.996; effect size
Cohen’s d = − 0.039 “negligible”; Fig. 3).

Experimental data

The proportion of prey smaller than the average spider size was 62.16% in the
centre and 54.17% in the periphery, which did not differ significantly between the two
positions (χ2 = 0.79, df = 1, P = 0.373). So, at least half of the prey intercepted by the
sticky traps were potentially manageable by Uloboridae spiders, as the family is com-
posed of small spiders that lack venom glands, which can limit the size of prey they can
capture (Opell 1979, 1998). The position (i.e. centre or periphery) affected the abun-
dance of intercepted prey (deviance: χ2 = 26.341, df = 3, P < 0.001), although the effect
was not the same across the different trap inclinations (deviance: χ2 = − 23.725, df = 2,
P < 0.001). Horizontal traps intercepted more prey in the centre (mean ±
SE = 3.17 ± 0.297, n = 12) than in the periphery (mean ± SE = 1.333 ± 0.1884,
n = 12) (Tukey post-hoc: Z = − 4.437, P < 0.001; effect size Cohen’s d = 2.128 “large”;
Fig. 4A). In contrast, vertical traps intercepted more prey in the periphery (mean ±
SE = 3.083 ± 0.358, n = 12) than in the centre (mean ± SE = 1.167 ± 0.207, n = 12)
(Tukey post-hoc: Z = 5, P < 0.001; effect size Cohen’s d = − 1.892 “large”; Fig. 4A).
Inclined traps presented the same pattern, but rather weaker, as the vertical ones
(Tukey post-hoc: Z = 3.430, P = 0.008; effect size Cohen’s d = − 1.328 “large”; Fig. 4A),
because inclined traps in the periphery (mean ± SE = 3.5 ± 0.452, n = 12) intercepted
more prey than the ones in the centre (mean ± SE = 1.833 ± 0.241, n = 12). Conversely,
prey interception was similar between vertical traps in the periphery and horizontal
ones in the centre (Tukey post-hoc: Z = − 0.252, P = 0.999).
10 S. Diniz et al.

Fig. 4. — Average prey abundance (A), diversity (B) and size (C) for each sticky trap inclination
(horizontal, inclined and vertical) at the centre and periphery of the interception region from
Aglaoctenus host webs. Error bars represent the standard error.

The effect of position on prey size was variable among different inclinations
(deviance: χ2 = − 6.241, df = 2, P = 0.041). The major effect pointed out by the post-
hoc contrasts was between vertical and horizontal traps in the centre (Tukey post-hoc:
Z = − 3.504, P = 0.006; effect size Cohen’s d = 1.317 “large”; Fig. 4C), as the horizontal
traps intercepted larger prey (mean ± SE = 3.122 ± 0.344, n = 12) than the vertical ones
(mean ± SE = 1.708 ± 0.271, n = 12).

DISCUSSION

Many studies interpret the variability of orb-web parameters, such as mesh size
(Sandoval 1994; Sensenig et al. 2010), asymmetry (Heiling & Herberstein 1999; Kuntner
et al. 2010; Pasquet et al. 2014) and capture area (Sherman 1994; Barrantes & Eberhard
2012), as optimal foraging strategies. However, few studies have addressed the role of
web inclination in foraging strategies adopted by spiders (e.g. Eberhard 1989; Bishop &
Connolly 1992). For both spider species in our study (i.e. Zosis and Philoponella), the
orb-webs in the centre tend to have lower inclination, in relation to the ground, than the
ones in periphery. Conversely, prey flow tends to be vertical in the centre and horizontal
in the periphery of the host web. This inclination pattern in orb-webs of Uloboridae is
consistent with the hypothesis of optimisation of prey interception, as the orb-web
inclination is kept orthogonal to the predominant direction of prey flow.
Web inclination can vary both inter- and intraspecifically. Usually, orb-web spi-
ders of the family Araneidae build webs with vertical inclination relative to the ground,
Habitat structure, prey capture, and web inclination 11

whereas many representatives of the family Uloboridae (e.g. the genera Uloborus and
Lubinella) and some of the family Tetragnathidae (e.g. the genera Leucauge and
Tetragnatha) build horizontal webs (Lubin 1986; Eberhard 1990). The studies of
Chacón and Eberhard (1980) and Eberhard (1989) demonstrated, however, that vertical
webs are more efficient because they intercept a larger quantity of prey and retain them
for longer periods. The reason why certain spiders build horizontal webs was incon-
clusive. Many studies have related the advantages of building horizontal webs to abiotic
factors. For example, Eberhard (1971) proposed that horizontal webs are built in
response to wind damage, Opell (1998) suggested that space limitation may favour
the construction of horizontal webs, and Krakauer (1972) proposed that Nephila cla-
vipes webs that diverge from vertical inclination help thermoregulation in environments
with high insolation.
Despite the limited references on the role of web inclination in foraging strategies,
Chacón and Eberhard (1980) suggested that horizontal webs could be more efficient in
environments with prey that fall from above. In fact, Bishop and Connolly (1992)
demonstrated that, in a forest environment, horizontal and inclined sticky traps inter-
cepted more prey than vertical traps did; whereas Chacón and Eberhard (1980) found
that, in an open field, vertical traps were more efficient in prey interception than the
inclined and horizontal ones were. The authors attributed this contradiction to differ-
ences in insect flight patterns between the forest and the open field environment, as
forest structure favours more vertical movements by insects than an open field does.
Furthermore, some studies have pointed out that some spiders near water surfaces tend
to build horizontal orb-webs to better intercept emerging aquatic insects (e.g. Buskirk
1975; Williams et al. 1995; Henschel et al. 2001; Kato et al. 2003). Contradictorily,
Prokop (2005) showed that horizontal orb-webs from Larinioides cornutus, a species
that is able to build both horizontal and vertical webs, were neither related to water nor
more successful than vertical orb-webs in intercepting aquatic insects, which makes it
puzzling to interpret the process underlying spider web inclinations. Despite indirect
evidence that web inclination can be adjusted to the flight pattern of prey, the compar-
ison between forests and open fields does not exclude the influence of abiotic factors on
web construction. In a study by Vollrath et al. (1997), the authors evaluated the effect of
several abiotic factors, such as wind, humidity, temperature and anchorage structures,
on web geometry. All the environmental factors affected the web geometry, which led
the authors to suggest the need to control abiotic environmental factors before con-
cluding that biotic factors have an effect on the web geometry. Although the results of
this study are partially based on observational data, the difference in inclination of
Uloboridae webs between the centre and periphery was observed at a microhabitat
scale through the use of block design (i.e. Aglaoctenus web). This allowed for the control
of environmental factors since Uloboridae individuals at the centre and periphery in
each Aglaoctenus web were subjected to the same environmental conditions, which
supports the hypothesis that the orb-webs’ inclination variability is an adjustment to
the direction of prey flux as a strategy to optimise interception.
Web adjustment to space and attachment structures is another explanation for the
variability in web building (e.g. Ades 1986; Vollrath et al. 1997; Krink & Vollrath 2000).
Although the physical structure differed between centre and periphery in A. lagotis
webs, the host used for Zosis individuals, there was no difference in A. castaneus
webs, the host used for Philoponella individuals. Since orb-webs from both positions
(i.e. centre and periphery) were inside the interception region and used only the host
web threads as attachment points, structural constraints of the microhabitat are unli-
kely to explain the difference in web inclination for Philoponella. The same is not the
12 S. Diniz et al.

case for Zosis, as its host web presented higher thread density in the periphery than in
the centre. It is possible that Zosis individuals in the periphery are subject to more
constrained space than the ones in the centre are. However, the structural difference
between centre and periphery is not enough to explain the pattern of orb-web inclina-
tion in Zosis. The reason is that the magnitude of difference between centre and
periphery is larger for the Zosis orb-webs’ inclination (Cohen’s d = − 2.65) than for
the thread density of the A. lagotis host web (Cohen’s d = − 1.635). In concordance, the
magnitude of difference between centre and periphery was larger for prey interception
than for thread density, as observed for horizontal (Cohen’s d = 2.128) and vertical
sticky traps (Cohen’s d = − 1.892). Furthermore, since the periphery of Zosis host webs
has higher thread density than the centre, the periphery would provide more anchor
points for horizontal orb-webs. However, Zosis orb-webs in the periphery were mostly
close to vertical, which weakens the hypothesis of microhabitat structural restraining.
Therefore, the consistent pattern of web inclination between Zosis and Philoponella
cannot be solely explained by physical constraints of the microhabitat, but rather by the
flow pattern of prey created by the configuration of the host web.
For colonial orb-web spiders, researchers are often interested in the cost/benefit
of central or peripheral positions inside a colony (Rypstra 1979; Rayor & Uetz 1990).
Spiders in the periphery are more exposed to the food source and can intercept prey
first, which implies that spiders in the centre are subject to shadow competition (Lubin
et al. 2001; Rao 2009). However, we found no evidence for the shadow competition
hypothesis because there was no difference, between centre and periphery, in abdomen
size for Zosis or Philoponella. One possible explanation of why we did not find evidence
for differential foraging success between the centre and periphery is that shadow
competition could have been compensated for by the ricochet effect. Rao (2009)
showed that, even in situations with only two spiders, the ricochet effect compensates
for shadow competition because some prey are not retained by the outer web and may
fall on inner webs. In the system of Uloboridae associated with Aglaoctenus webs, the
ricochet effect can result from prey colliding against, and escaping from, the peripheral
orb-webs or interception threads of Aglaoctenus. Although orb-web spiders in the
periphery are the first to intercept prey, the central spiders receive larger prey that
ricocheted from peripheral threads and/or orb-webs. Thus, prey tends to fall on hor-
izontal orb-webs or onto the sheet of the host web itself, which is in accordance with
the results from the sticky trap data. This effect could explain why there was no
difference between centre and periphery in spider abdomen size or in abundance of
intercepted prey by the traps, as well as the fact that larger prey was intercepted by
horizontal traps in central positions. Therefore, the ricochet effect may be the causal
factor of change in web inclination of Uloboridae associated with Aglaoctenus webs.
The pattern of differential orb-web inclination between the centre and periphery
of Aglaoctenus host webs support the hypothesis of web adjustment as a foraging
strategy. This result could also be interpreted as adaptive plasticity (Gotthard & Nylin
1995; Boutry & Blamires 2013), although to make such a statement it will be necessary
to conduct further controlled experiments to observe whether the same individual
changes web inclination in response to varying directions of prey flow. It is also
noteworthy that the inclination pattern is stronger for Zosis than for Philoponella.
Such a difference in inclination pattern between the two Uloboridae species could not
be explained by the fact that they are associated with host webs of distinct Aglaoctenus
species, as the web structure and prey capture method are very similar between A.
logotis and A. castaneus (Santos & Brescovit 2001; Stefani & Del-Claro 2015; authors’
pers. obs.). Maybe distinct environmental conditions could explain the stronger web
Habitat structure, prey capture, and web inclination 13

inclination plasticity for Zosis. While the Philoponella species was found in a conserved
coastal area of Atlantic forest with approximately 15,000 ha (Sampaio et al. 2005), the
Zosis species was in a small urban fragment of Brazilian Cerrado of only 11 ha in size
(Stefani & Del-Claro 2015). The dry season (April–September) is more pronounced in
the Cerrado than it is in the Atlantic forest. Furthermore, we obtained the observational
data in July, when some of the lowest precipitation rates and temperatures of the year
are recorded, which makes July one of the months with the greatest shortage of prey
(Oliveira-Filho & Fontes 2000; Pinheiro et al. 2002). Species or populations originating
from more stable and resource-rich environments tend to exhibit a weaker degree of
plasticity than species from more unstable environments (e.g. Newman 1992; Seigel &
Ford 2001; Ghalambor & Martin 2002). So, the differential degree of web adjustment
between the two Uloboridae species of this study may be a case of plasticity as an
adaptation (Gotthard & Nylin 1995).

ACKNOWLEDGEMENTS

We thank Antônio Brescovit for identifying the spider species. We thank the municipal
government of Araguari, the environmental department, and the employees of the John Kennedy
Woodland Park (Parque Bosque John Kennedy) for their assistance in this study. We also acknowl-
edge employees from the State Park of Cardoso Island, Perequê Nucleus. Part of the data was
obtained during the field course “Ecologia da Mata Atlântica”, carried out in Ilha do Cardoso, July
2008.

DISCLOSURE STATEMENT

No potential conflict of interest was reported by the authors.

FUNDING

This work was supported by: (1) Coordenaçã o de Aperfeiçoamento de Pessoal de Nível
Superior (CAPES) – Plano Nacional de Pós-Doutorado da Pós-Graduaçã o em Ecologia e
Conservação/UFU [grant number CAPES PNPD/1533482] for Vanessa Stefani; (2) Fundação de
Amparo à Pesquisa do Estado de São Paulo (FAPESP) – Master Scholarship [grant number 07
/57294-5] for Suzana Diniz; (3) Conselho Nacional de Desenvolvimento Científico e Tecnológico
(CNPq) – Doutorado Pleno (GD) [grant number 159631/2011-4] for Suzana Diniz; (4) Coordenação
de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) – Programa Institucional de Bolsas de
Doutorado Sanduíche no Exterior (PDSE) [grant number 8321/14-0] for Suzana Diniz.

REFERENCES

Ades C. 1986. A construção da teia geométrica como programa comportamental [The geometrical
construction of the web as behavioral program]. Cienc Cult. 38:760–775. Portuguese.
Alves-Costa CP, Gonzaga MO. 2001. Prey capture and spatial distribution of Philoponella vittata
(Araneae Uloboridae) in host webs. Ethol Ecol Evol. 13:239–246.
Anotaux M, Toscani C, Leborgne R, Châline N, Pasquet A. 2014. Aging and foraging efficiency
in anorb-web spider. J Ethol. 32:155–163.
14 S. Diniz et al.

Barrantes G, Eberhard WG. 2012. Extreme behavioral adjustments by an orb-web spider to

restricted spaces. Ethology. 118:438–449.
Biere JM, Uetz GW. 1981. Web orientation in the spider Micrathena gracilis (Araneae: Araneidae).
Ecology. 62:336–344.
Bishop L, Connolly SR. 1992. Web orientation, thermoregulation, and prey capture efficiency in a
tropical forest spider. J Arachnol. 20:173–178.
Blackledge TAS, Nikolaj S, Coddington JA, Szüts T, Wenzel JW, Hayahi CY, Agnarsson I.
2009. Reconstructing web evolution and spider diversification in the molecular era. PNAS.
106:5229–5234.
Blamires SJ. 2010. Plasticity in extended phenotypes: orb web architectural responses to varia-
tions in prey parameters. J Exp Biol. 213:3207–3212.
Blamires SJ, Chao IC, Liao CP, Tso IM. 2011. Multiple prey cues induce foraging flexibility in a
trap-building predator. Anim Behav. 81:955–961.
Bonte D, Lanckacker K, Wiersma E, Lens L. 2008. Web building flexibility of an orb-web spider
in a heterogeneous agricultural landscape. Ecography. 31:646–653.
Boutry C, Blamires SJ. 2013. Plasticity in spider webs and silk: an overview of current evidence.
In: Santerre M, editor. Insects and other terrestrial arthropods: biology, chemistry and
behavior – spiders: morphology, behavior and geographic distribution. New York: Science
Publishers, Inc; p. 1–46.
Buskirk RE. 1975. Coloniality, activity patterns and feeding in a tropical orb‐weaving spider.
Ecology. 56:1314–1328.
Castillo JA, Eberhard WG. 1983. Use of artificial webs to determine prey available to orb weaving
spiders. Ecology. 64:1655–1658.
Chacón P, Eberhard WG. 1980. Factors affecting numbers and kinds of prey caught in artificial
spider webs, with consideration of how orb webs trap prey. Bull Br Arachnol Soc. 5:189–204.
Cohen J. 1988. Statistical power analysis for the behavioral sciences. New York: Academic Press.
Cohen J. 1992. Statistical power analyses. Curr Dir Psychol Sci. 1:98–101.
Craig CL. 2003. Spider webs and silk: tracing evolution from molecules to genes to phenotypes.
New York: Oxford University Press.
Eberhard WG. 1971. The ecology of the web of Uloborus diversus (Araneae: Uloboridae).
Oecologia. 6:328–342.
Eberhard WG. 1986. Effects of orb-web geometry on prey interception and retention. In: Shear
WA, editor. Spiders: webs, behavior, and evolution. Stanford: Stanford University Press; p.
70–100.
Eberhard WG. 1989. Effects of orb web orientation and spider size on prey retention. Bull Br
Arachnol Soc. 8:45–48.
Eberhard WG. 1990. Function and phylogeny of spider webs. Annu Rev Ecol Syst. 21:341–372.
Fox J, Weisberg HS. 2011. An R companion to applied regression. Thousand Oaks (CA): Sage.
Ghalambor CK, Martin TE. 2002. Comparative manipulation of predation risk in incubating
birds reveals variability in the plasticity of responses. Behav Ecol. 13:101–108.
Gotthard K, Nylin S. 1995. Adaptive plasticity and plasticity as an adaptation: a selective review of
plasticity in animal morphology and life history. Oikos. 74:3–17.
Hagstrum DW. 1971. Carapace width as a tool for evaluating the rate of development of spiders in
the laboratory and the field. Ann Entomol Soc Am. 64:757–760.
Harmer AMT, Herberstein ME. 2009. Taking it to extremes: what drives extreme web elongation
in Australian ladder web spiders (Araneidae: Telaprocera maudae)? Anim Behav. 78:499–504.
Heiling AM, Herberstein ME. 1999. The role of experience in web-building spiders (Araneidae).
Anim Cogn. 2:171–177.
Henschel JR, Mahsberg D, Stumpf H. 2001. Allochthonous aquatic insects increase predation
and decrease herbivory in river shore food webs. Oikos. 93:429–438.
Herberstein ME, Gaskett AC, Glencross D, Hart S, Jaensch S, Elgar MA. 2000. Does the
presence of potential prey affect web design in Argiope keyserlingi (Araneae, Araneidae)? J
Arachnol. 28:346–350.
Habitat structure, prey capture, and web inclination 15

Herberstein ME, Tso IM. 2011. Spider webs: evolution, diversity and plasticity. In: Herberstein
ME, editor. Spider behavior, flexibility and plasticity. New York: Cambridge University
Press; p. 57–98.
Hesselberg T. 2015. Exploration behavior and behavioural flexibility in orb-web spiders: a review.
Curr Zool. 61:313–327.
Jakob EM, Marshall SD, Uetz GW. 1996. Estimating fitness: a comparison of body condition
indices. Oikos. 77:61–67.
Kato C, Iwata T, Nakano S, Kishi D. 2003. Dynamics of aquatic insect flux affects distribution of
riparian web‐building spiders. Oikos. 103:113–120.
Krakauer T. 1972. Thermal responses of the orb - weaving spider, Nephila clavipes. Am Midl Nat.
88:245–250.
Krink T, Vollrath F. 2000. Optimal area use in orb webs of the spider Araneus diadematus.
Naturwissenschaften. 87:90–93.
Kuntner M, Gregoric M, Li D. 2010. Mass predicts web asymmetry in Nephila spiders.
Naturwissenschaften. 97:1097–1105.
Leborgne R, Pasquet A. 1987. Influence of conspecific silk-structures on the choice of a web site
by the spider Zygiella x-notata (Clerk). Rev Arachnol. 7:85–90.
Levi HW. 1978. Orb-webs and phylogeny of orb-weavers. Symp Zool Soc. 42:1–15.
Li D, Lee W. 2004. Predator-induced plasticity in web building behaviour. Anim Behav. 67:309–
317.
Lubin YD. 1986. Web building and prey capture in the Uloboridae. In: Shear WA, editor. Spiders:
webs, behavior, and evolution. Stanford: Stanford University Press; p. 132–171.
Lubin Y, Henschel JR, Baker MB. 2001. Costs of aggregation: shadow competition in a sit-and-
wait predator. Oikos. 95:59–68.
Marples BJ. 1962. Notes on spiders of the family Uloboridae. Ann Akad Zool Agric India. 4:1–11.
Mayntz D, Nielsen VH, Sørensen A, Toft S, Raubenheimer D, Hejlesen C, Simpson SJ. 2009.
Balancing of protein and lipid by a mammalian carnivore, the mink (Mustela vison). Anim
Behav. 77:349–355.
McNett BJ, Rypstra AL. 1997. Effects of prey supplementation on survival and web site tenacity
of Argiope trifasciata (Araneae, Araneidae): a field experiment. J Arachnol. 25:352–360.
Moya‐Laraño J, Taylor PW, Fernández‐Montraveta C. 2003. Body patterns as potential ampli-
fiers of size and condition in a territorial spider. Biol J Linn Soc. 78:355–364.
Newman RA. 1992. Adaptive plasticity in amphibian metamorphosis. Bioscience. 42:671–678.
Nogueira SSC, Ades C. 2012. Evidence of learning in the web construction of the spider Argiope
argentata (Araneae: Araneidae). Rev Etol. 11:23–36.
Oliveira-Filho AT, Fontes MAL. 2000. Patterns of floristic differentiation among atlantic forests
in southeastern Brazil and the influence of climate. Biotropica. 32:793–810.
Opell BD. 1979. Revision of the genera and tropical American species of the spider family
Uloboridae. Bull Mus Comp Zool. 48:443–549.
Opell BD. 1998. Economics of spider orb-webs: the benefits of producing adhesive capture thread
and of recycling silk. Funct Ecol. 12:613–624.
Pasquet A, Marchal J, Anotaux M, Leborgne R. 2014. Does building activity influence web
construction and web characteristics in the orb-web spider Zygiella x-notata (Araneae,
Araneidae)? Zool Stud. 53:11.
Pinheiro F, Diniz IR, Coelho D, Bandeira MPS. 2002. Seasonal pattern of insect abundance in
the Brazilian cerrado. Austral Ecol. 27:132–136.
Prokop P. 2005. Web inclination alters foraging success of a nocturnal predator. Ital J Zool.
72:249–252.
R Core Team. 2014. R: a language and environment for statistical computing. Vienna, Austria: R
Foundation for Statistical Computing. Available from: http://www.R-project.org/
Rao D. 2009. Experimental evidence for the amelioration of shadow competition in an orb-web
spider through the ‘ricochet’ effect. Ethology. 115:691–697.
Rayor LS, Uetz GW. 1990. Trade-offs in foraging success and predation risk with spatial position
in colonial spiders. Behav Ecol Sociobiol. 27:77–85.
16 S. Diniz et al.

Rayor LS, Uetz GW. 1993. Ontogenetic shifts within the selfish herd: predation risk and foraging
trade-offs change with age in colonial web-building spiders. Oecologia. 95:1–8.
Reed CF, Witt PN, Scarboro MB, Peakall DB. 1970. Experience and the orb web. Dev
Psychobiol. 3:251–265.
Rypstra AL. 1979. Foraging flocks of spiders. Behav Ecol Sociobiol. 5:291–300.
Rypstra AL, Binford GJ. 1995. Philoponella republicana (Araneae, Uloboridae) as a commensal in
the webs of other spiders. J Arachnol. 23:1–8.
Sampaio D, Souza VC, Oliveira AA, Paula-Souza J, Rodrigues RR. 2005. Árvores de restinga:
guia ilustrado para identificação das espécies da Ilha do Cardoso [Restinga trees: illustrated
guide to identify the species of Cardoso Island]. São Paulo: Neotrópica. Portuguese.
Sandoval CP. 1994. Plasticity in web desing in the spider Parawixia bistriata: a response to
variable prey type. Funct Ecol. 8:701–707.
Santos AJ, Brescovit AD. 2001. A revision of the South American spider genus Aglaoctenus
Tullgren, 1905 (Araneae, Lycosidae, Sosippinae). Andrias. 15:75–90.
Schneider JM, Vollrath F. 1998. The effect of prey type on the geometry of the capture web of
Araneus diadematus. Naturwissenschaften. 85:391–394.
Seigel RA, Ford NB. 2001. Phenotypic plasticity in reproductive traits: geographical variation in
plasticity in viviparous snake. Funct Ecol. 15:36–42.
Sensenig A, Agnarsson I, Gondek TM, Blackledge TA. 2010. Webs in vitro and in vivo: spiders
alter their orb-web spinning behavior in the laboratory. J Arachnol. 38:183–191.
Sherman PM. 1994. The orb-web: an energetic and behavioral estimator of a spider’s dynamic
foraging and reproductive strategies. Anim Behav. 48:19–34.
Souza JP, Araújo GM. 2005. Estrutura arbustivo/arbórea do subosque de clareiras e áreas
sobdossel fechado em floresta estacional semidecidual urbana em Araguari – MG
[Structure of understory woody species in gaps and closed-canopy sites of an urban seasonal
semideciduous forest in Araguari, MG]. Biosci J. 21:93–102. Portuguese.
Stefani V, Del-Claro K. 2012. Oviposition and post-embryonic development of Aglaoctenus lagotis
(Araneae: Lycosidae). Zoologia. 28:565–570.
Stefani V, Del-Claro K. 2015. The effects of forest fragmentation on the population ecology and
natural history of a funnel-web spider. J Nat Hist. 49:211–223.
Stefani V, Del-Claro K, Silva LA, Guimarães B, Tizo-Pedroso E. 2011. Mating behaviour and
maternal care in the tropical savanna funnel-web spider Aglaoctenus lagotis Holmberg
(Araneae: Lycosidae). J Nat Hist. 45:1119–1129.
Tanaka K. 1989. Energetic cost of web construction and its effect on web relocation in the web-
building spider Agelena limbata. Oecologia. 81:459–464.
Uetz GW. 1989. The “ricochet effect” and prey capture in colonial spiders. Oecologia. 81:154–159.
Uetz GW, Biere JM. 1980. Prey of Micrathena gracilis (Walckenaer) (Araneae: Araneidae) in
comparison with artificial webs and other trapping devices. Bull Br Arachnol Soc. 5:101–
107.
Uetz GW, Papke R, Kilinc B. 2002. Influence of feeding regime on body size, body condition and
a male secondary sexual character in Schizocosa ocreata wolf spiders (Araneae, Lycosidae):
condition-dependence in a visual signaling trait. J Arachnol. 30:461–469.
Venner S, Pasquet A, Leborgne R. 2000. Web-building behaviour in the orb-weaving spider
Zygiella x-notata: influence of experience. Anim Behav. 59:603–611.
Vieira C, Japyassú HF, Santos AJ, Gonzaga MO. 2007. Teias e forrageamento. In: Gonzaga MO,
et al., editors. Ecologia e comportamento de aranhas [Ecology and spiders behaviour]. Rio
de Janeiro: Editora Interciência; p. 45–66. Portugese.
Vollrath F. 1987. Spiders with regenerated legs can build normal webs. Nature. 328:247–248.
Vollrath F, Downes M, Krackow S. 1997. Design variability in web geometry of an orb-weaving
spider. Physiol Behav. 62:735–743.
Vollrath F, Samu F. 1997. The effect of starvation on web geometry in an orb-weaving spider. Bull
Br Arachnol Soc. 10:295–298.
Williams DD, Ambrose LG, Browning LN. 1995. Trophic dynamics of two sympatric species of
riparian spider (Araneae: Tetragnathidae). Can J Zool. 73:1545–1553.
Witt PN. 1971. Drugs alter web-building of spiders: a review and evaluation. Behav Sci. 16:98–112.