Advertisement

Evolutionary Ecology

, Volume 32, Issue 2–3, pp 159–170 | Cite as

Tracing the evolutionary origin of a visual signal: the coincidence of wrap attack and web decorating behaviours in orb web spiders (Araneidae)

  • André Walter
Original Paper

Abstract

The silk decorations that adorn the webs of many orb-web spiders are thought to have a signal function, but the evolution of the decorating behaviour remains unresolved. The decoration signal is maintained apparently because it improves foraging efficiency, through either increased encounter rates with prey or reduced damage to the web. Recent investigations suggest that the decorations may originate in a regulation of the activity of the aciniform silk glands, which produce silk for both decorating the web and wrapping prey. This view predicts a link between decorating behaviour and a preference for restraining prey by wrapping with silk, which is evident among species of Argiope spiders. Here I compare the frequency of the wrap attack behaviour in four species of orb-web spiders that occupy the same habitat, but differ in their silk decorating behaviour: two species, Plebs bradleyi and Gea theridioides, build silk decorations, while the other two, Araneus hamiltoni and Backobourkia brounii do not. Spiders were presented with prey items that varied in the ease with which they could be captured, with houseflies being more easily subdued than house crickets. As predicted, the silk decorating species used wrap attacks significantly more often than non-decorating spiders, irrespective of the prey species. These data support the view that both behaviours are evolutionary linked. I propose that silk decorating originated from the evolution of wrap attacking, and that silken web decorations have later evolved into a signal and are now maintained for that function.

Keywords

Wrap attack Web decoration Signal evolution Orb web spiders 

Introduction

One of the major challenges in evolutionary ecology is the revelation of the adaptive value of animal signals. However, an understanding of a signal’s ultimate function alone does often not allow conclusions on how it has evolved. Selection pressures explaining either the maintenance or the evolutionary origin of a signal may be very different (Walter and Elgar 2012). Unravelling the origin of signals can be particularly ambitious when they derive from non-signalling traits (Schaedelin and Taborsky 2009). While many animal signals are elaborations of those characters, others seem to lack even a non-signalling predecessor and must be regarded as novel traits (cf. Endler 1992). For example, male bowerbirds throw colourful objects close to the entrance of their bower in order to attract female attention (Endler and Day 2006). This behaviour is not related to any aspect of the biology of these birds and is therefore regarded as novel signalling. Its evolutionary origin is thus obscure, since it remains unsolved where this behaviour has been derived from (Endler and Théry 1996). The construction of silken web decorations in many orb-web spiders represents another intriguing example. The literature presents a controversial debate about various signal functions of these decorations (see reviews in Eberhard 2003; Bruce 2006; Walter and Elgar 2012). Despite the large number of studies, all the described functions only refer to ultimate signal effects that may well explain the maintenance of web decorating, yet the evolutionary origin of the behaviour remains largely unknown (see review, Walter and Elgar 2012).

Silk decorations are known for a number of orb-web spider species, including the families Araneidae, Uloboridae and Nephilidae (Herberstein et al. 2000). These extra structures predominantly consist of silk arranged in a species- or genus-specific pattern, but some species, like Cyclosa, also use other materials such as plant parts or prey remains (Chou et al. 2005; Tan et al. 2010). For over a 100 years, biologists have tried to solve the riddle about the function of web decorations with indefinite success. It is widely accepted that they act as visual signals, yet the discussion about the nature of the receiver(s) continues. The main camps of researchers regard web decorations either as prey attractants (Craig and Bernard 1990; Tso 1998; Herberstein 2000; Cheng and Tso 2007; Blamires et al. 2008) or predator avoidance devices (Schoener and Spiller 1992; Blackledge 1998; Blackledge and Wenzel 2001; Eberhard 2003; Nakata 2009). As a compromise, it has been suggested that either different web decoration patterns serve different signal functions (Starks 2002; Cheng et al. 2010) or they do not represent signals at all (Walter and Elgar 2016). While still struggling with reaching consensus about the selective forces explaining the maintenance of web decorations, it remains even more puzzling why the spiders built these extra structures in the first place.

Web decoration signals are visible to both prey and predators (Bruce et al. 2001, 2005) and must be considered to represent evolutionary novel signals (Walter and Elgar 2012). Although built together with the capture web (McCook 1889), there is no known part of the web building behaviour they may have been elaborated from. Yet the silk these spiders use for decorating is known. Orb-web spiders are able to produce up to seven different types of silk (Vollrath and Knight 2001; Foelix 2011), each of them used for different purposes depending on their physical properties. Peters (1993) found the silk used for constructing web decorations mainly consists of products of aciniform silk glands. Aciniform silk is extraordinarily tough and extensible (Hayashi et al. 2004). Hence, it is only straight-forward that spiders use this silk for wrapping prey (Vollrath and Knight 2001). Intriguingly, this material, which is crucial for effectively arresting struggling prey, is also used for creating a visual signal. Consequently, a direct link between both the prey capture and web decorating behaviours has been suggested (Peters 1993; Tso 2004; Walter et al. 2008).

The prey capture behaviour of orb-web spiders has been studied repeatedly. In essence, these spiders use two basic strategies to subdue prey, differing in the sequence of attack behaviours. They either bite the prey first and wrap it in silk afterwards, or wrap it first with the biting following (Robinson and Olazarri 1971; Olive 1981; Müller and Westheide 1993). The latter strategy does not primarily rely on the sedating effect of spider venom to immobilise prey but on the arresting effect of a tight entanglement with wrapping silk. This so called ‘wrap attack’ or ‘attack wrapping’ strategy has the advantage that spiders can overcome large and defensive prey as they do not need to face the danger of getting close to a potentially fiercely fighting prey item. However, the quick throw of dense silk layers from a safe distance onto the prey is a silk consuming strategy as a lot of silk is deployed. Wrap attacking spiders must therefore rely on a constantly high provision of aciniform silk.

Walter et al. (2008) found in a study on three different Argiope species that the manipulative depletion of the aciniform silk glands resulted in an intensified web decoration activity. This result has been attributed to a gland activation by mechanical stimulation, causing a production increase of aciniform silk, noticeable by larger decorations in subsequent webs. This mechanism is suggested to allow the spiders to adjust the aciniform silk production according to their needs for wrap attacks depending on the prevailing prey abundance. In the evolutionary past, spiders may have been forced to deposit silk in the web with peaking aciniform silk productions (Walter et al. 2008). The deposited silk may have attracted the attention of certain receivers, allowing secondary selection to act on the decoration pattern, eventually shaping a visual signal (Walter and Elgar 2012). This silk gland regulation hypothesis offers an explanation for the evolutionary origin of the decoration signal (Walter and Elgar 2012).

The silk gland regulation hypothesis is not yet satisfyingly supported by empirical evidence. It still requires a more detailed knowledge about the functioning of the aciniform silk glands, e.g. their production physiology or the silk gene expression levels in the secretory cells. Physiological and molecular studies are required, yet elaborate. A reasonable first step may be the verification of a behavioural coherency between wrap attacking and web decorating across different species to justify more elaborate experiments in the future. So far, the preferential use of wrap attacks in combination with silk decorating has only been described for the well-studied genus Argiope (Walter et al. 2008). I here provide results on the prey capture behaviour of four other araneid spider species of South Eastern Australia. All four occupy the same habitat and notably also share it with two Argiope species, A. protensa and A. trifasciata. Two of the study species build silk decorations, while the other two do not. I hypothesise a positive correlation between wrap attacking and silk decorating and thus, that the silk decorating species perform wrap attacks more frequently than the non-decorating species when facing the same prey type.

Materials and methods

I collected 12 adult females each of four araneid spider species, Araneus hamiltoni (Rainbow, 1893), Backobourkia brounii (Urquhart, 1885) (both subfamily Araneinae, Framenau et al. 2010), Plebs bradleyi (Keyserling, 1887) (subfamily Araneinae, Joseph and Framenau 2012) and Gea theridioides (L. Koch, 1872) (subfamily Argiopinae, Scharff and Coddington 1997). The latter two species build silken web decorations, while the first two do not (Fig. 1).
Fig. 1

The four study species in their common natural habitat, a grassland approx. 60 km north of Melbourne/Australia. Non-decorating species: a Araneus hamiltoni, b Backobourkia brounii. Silk decorating species: c Plebs bradleyi, d Gea theridioides

The silk decorating behaviour has evolved several times independently within the Araneidae (Scharff and Coddington 1997), and sampling genera of distant clades allowed the comparison of two decorating species from different subfamilies (G. theridioides vs. P. bradleyi). With Plebs and Backobourkia, the study further comprised two either decorating or non-decorating, yet closely related genera (both placed within the Australian ‘eriophorines’, see Joseph and Framenau 2012). The inclusion of Araneus as the ‘classical’ representative of the family completes the set of study species. However, the most important criterion for choosing these four species for the analysis was their co-occurrence in the same habitat, not only ensuring a similar history of abiotic habitat conditions but also implying some overlap in their prey spectrum.

The two non-decorating species are characterised by the construction of protective retreats next to their capture web, whereas this behaviour is lacking in the other two, which build silken web decorations in a vertical linear pattern, consisting of two zigzag-shaped silk ribbons, one above and one below the web hub. All individuals were collected in a grassland area along Ennis Rd, close to Hume Hwy, 2 km south of Tallarook Victoria/Australia in February 2012. Spiders were transferred to individual web frames (58 × 58 × 15 cm; and 25 × 25 × 10 cm for the smaller species, G. theridioides) and maintained in the laboratory under 12/12 h dark–light cycle and a constant temperature of 25 °C. Individual webs were misted with water every other day and spiders were fed once with two houseflies (Musca domestica) prior to the tests. Trials commenced after 5 days of acclimation.

For 10 days, one housefly (ca. 15–16 mg) was thrown into the capture web of each of the spiders every other day (= five repeats). The initial response of the spiders was recorded and their capture behaviour categorised as bite or wrap attack. Bite attacks were counted if spiders directly approached the prey and bit it, with the wrapping following or lacking. Wrap attacks were recorded if spiders did not get in direct contact with the prey, but threw silk threads/bands over the prey as their initial action of subduing, with biting as a later response, or not following at all (cf. Robinson and Olazarri 1971). After 5 min of prey handling time the fly was removed from the web in order to keep the prey capture motivation up for the following test days. Spiders were weighed and their body length recorded daily.

The whole experimental approach was repeated with house crickets (Acheta domestica juv., 80–100 mg; smaller individuals with 40–50 mg were given to Gea theridioides), to test the spiders’ response to a larger and more defensive prey type compared to flies. After the first 10 days of experiment and prior to the rerun, spiders were fed with two house crickets (approx. 70 mg) and allowed to recover for 2 days.

Statistical analyses were performed using the R platform (http://www.R-project.org). The differences in frequencies of wrap attacks between species were calculated using a two-way ANOVA with spider and prey species as main effects. As I used percentages for this analysis the values were square root-/arcsine-transformed to normalise them prior to performing the ANOVA. Since the ANOVA revealed significant interactions between prey and spider species, I performed a Tukey-HSD post hoc test to identify the significance levels of effects. I used Fisher’s exact Tests to further compare the frequencies of the two different prey capture strategies in fly versus house cricket feedings.

Results

The four study species had an equal body mass and size with the exception of the smaller decorating species Gea theridioides (see Table 1). The two decorating species did not build protective retreats and frequently left both captured flies (Plebs: 14.55%; Gea: 25.93%) and crickets (Plebs: 58.62%; Gea: 33.93%) in the capture area of the web. By contrast, the two non-decorating species Araneus hamiltoni and Backobourkia brounii carried their prey items away from the place of capture, either to the web hub or the retreat of the spiders (Table 2). Irrespective of the presence of the decorating behaviour, all four species performed wrap attacks more frequently when faced with the larger prey, the house crickets. However, in comparison, wrap attacks were used significantly more frequently by the decorating species both with houseflies and house crickets as prey (ANOVA, F3,1 = 74.29, p < 0.01; Table 3, Fig. 2; see also supplemental material for details).
Table 1

Body condition measurements of the four orb-web spider species prior to the prey capture experiments (n = 12 each)

Species

Body mass (SE mg)

Body length (SE mm)

Silk decorating

 Plebs bradleyi

95.79 ± 6.36

10.74 ± 0.39

 Gea theridioides

38.12 ± 2.93

5.81 ± 0.21

Non-decorating

 Araneus hamiltoni

96.7 ± 7.71

9.61 ± 0.37

 Backobourkia brounii

106.25 ± 11.43

9.95 ± 0.49

Table 2

Overview of behavioural elements of prey capture for the four study species

Behaviour

Silk decorating species

Non-decorating species

P. bradleyi

G. theridioides

A. hamiltoni

B. brounii

Prey: Musca domestica

n = 55

n = 54

n = 50

n = 55

 No. of wrap turns before bite

5.64 ± 0.44

6.26 ± 0.46

5.36 ± 1.03

5 ± 1.11

 No. of wrap turns after bite

7.85 ± 0.54

6.03 ± 0.58

7.69 ± 0.42

7.78 ± 0.36

 Carried to web hub

47

40

48

44

 Carried to retreat

2

11

 Left in capture spiral

8

14

0

0

Prey: Acheta domestica

n = 58

n = 56

n = 55

n = 55

 No. of wrap turns before bite

10.72 ± 0.67

13.95 ± 0.75

11.22 ± 1.59

11.88 ± 1.13

 No. of wrap turns after bite

9.15 ± 0.83

6.63 ± 0.73

15.5 ± 0.9

12.39 ± 0.96

 Carried to web hub

34

35

51

37

 Carried to retreat

4

18

 Left in capture spiral

24

21

0

0

Several times, the silk-decorating species showed the habit to leave subdued prey items at the place of capture. As they never build retreats, this aspect of the prey transport could not be compared to the non-decorating species

Table 3

Differences in frequencies of wrap attacks (p values) between species and treatments

Prey: Musca domestica

Silk decorating species

Non-decorating species

Plebs bradleyi

Gea theridioides

Araneus hamiltoni

Backobourkia brounii

Silk decorating species 

 Plebs bradleyi

 Gea theridioides

< 0.001

Non-decorating species

 Araneus hamiltoni

< 0.05

< 0.001

 Backobourkia brounii

< 0.01

< 0.001

0.9969455

Prey: Acheta domestica

Silk decorating species

Non-decorating species

Plebs bradleyi

Gea theridioides

Araneus hamiltoni

Backobourkia brounii

Silk decorating species

 Plebs bradleyi

 Gea theridioides

< 0.05

Non-decorating species

 Araneus hamiltoni

< 0.001

< 0.001

 Backobourkia brounii

< 0.01

< 0.001

0.0715994

The silk-decorating species used wrap attacks more frequently in any case. B. brounii more often performed wrap attacks when faced with house crickets compared to A. hamiltoni. There was no such difference when both species were given flies. p values were calculated with TukeyHSD post hoc tests after performing a two-way ANOVA (see text and supplemental material; cf. also Fig. 2)

Fig. 2

The frequency of wrap attacks used by the four study species when presented either with a houseflies (Musca domestica) or b house crickets (Acheta domestica). The silk decorating species P. bradleyi and G. theridioides use wrap attacks significantly more often for both prey types then the non-decorating orb weavers (ANOVA, F3.1 = 74.29, p < 0.01). ***Equals a significance level of p < 0.001

Non-decorating species

When given houseflies, most females responded with bite attacks (A. hamiltoni: 88.3%, ntot = 55 attacks; B. brounii: 92.92%, ntot = 56 attacks; Fig. 2a). Some individuals never performed wrap attacks at any of the five experimental days (A. hamiltoni: 7 of 12; B. brounii: 9 of 12). Flies were usually carried to the web hub for consumption, and a few individuals brought them to their retreats (Table 2). All the houseflies were thoroughly wrapped after biting and prior to the transport. Faced with house crickets, all spiders tend to use wrap attacks more frequently, but only in B. brounii it was significantly different from handling flies (A. hamiltoni: 21.3% vs. 11.7% for flies; Fisher Test; p = 0.895, ntot = 50 attacks; B. brounii: 47.92% vs. 7.08% for flies; Fisher Test; p < 0.01, ntot = 55 attacks; cf. Fig. 2). With 82.9 ± 2.9 SE mg the crickets were only little lighter than the capturing spiders of both species. Four of the 12 A. hamiltoni never performed wrap attacks on house crickets, while it was only one B. brounii that never did. Spiders of both non-decorating species used more wrap turns after than before biting the prey (Table 2). Similar to the handling of flies, the captured crickets were stored at the web hub and sometimes carried to the retreat.

Silk decorating species

The larger of the two decorating species, P. bradleyi, attacked houseflies more frequently by biting (63.47%), while G. theridioides rarely set the bite first (19.44%), but predominantly used wrap attacks (Fig. 2a). Two individuals of P. bradleyi never performed wrap attacks, whereas wrap attacks were performed by all G. theridioides, at least once during the 5-day observation period. For house crickets, the wrap attack strategy turned out to be the most common response in both species and was significantly more frequent than in the fly-treatment (P. bradleyi: 78.89 vs. 36.53% for flies; Fisher Test; p < 0.001, ntot = 58 attacks; G. theridioides: 100% vs. 74.02% for flies; Fisher Test; p < 0.001, ntot = 56 attacks). Together with the more frequent use of wrap attacks, spiders used more wrap turns before than after biting (Table 2). Both, P. bradleyi and G. theridioides do not build retreats, and in this study, they showed the frequent habit to leave the prey items at the site of capture, with this behaviour being stronger pronounced after catching crickets. This was never observed in the non-decorating species, for neither prey type (Table 2).

Discussion

My experimental results reveal a concurrence between a preference for the wrap attack prey capture strategy and the presence of the silk decorating behaviour in four different araneid spider species, suggesting an evolutionary link between both. The two decorating species Plebs bradleyi and Gea theridioides use wrap attacks more frequently to subdue both flies and crickets compared to the non-decorating species Araneus hamiltoni and Backobourkia brounii they share the habitat with. In general, wrap attacks have been used more frequently when faced with the larger prey, house crickets.

The evolution of wrap attacks is thought to be an adaptation to subdue large and defensive prey insects (Robinson et al. 1969; Olive 1980). The spiders can keep a safe distance from an item that might bite, sting or kick them. Smaller prey such as flies, on the other hand, is easier to handle and may just be grabbed and held using the chelicerae, and in those cases bite attacks are used more frequently (Robinson and Olazarri 1971). The higher proportion of wrap attacks recorded in Plebs bradleyi and Gea theridioides may reflect a specialisation on relatively larger prey types in the habitat they share with the two non-decorating species. While, for example, the North American Araneus trifolium builds webs in higher strata to intercept flying insects, Argiope trifasciata, that occupies the same habitat, prefers lower strata and is specialised to catch the larger and more defensive orthopteran prey (Olive 1980). Consequently, the latter species is more often observed to use wrap attacks (Olive 1980). In accordance with those behavioural specialisations, Peters (1993) revealed also morphological differences between Argiope and Araneus spiders. Argiope, more frequently using wrap attacks than Araneus, is characterised by twice as many aciniform spinnerets. Aciniform glands produce both, silk for wrapping prey and for constructing web decorations (Vollrath and Knight 2001; Hayashi et al. 2004), suggesting a link between both behaviours (Peters 1993; Walter et al. 2008).

Silken web decorations are relatively rarely found in the phylogeny of orb web spiders, only occurring in a few genera (Scharff and Coddington 1997; Griswold et al. 1998). They are found in three different families: Nephilidae, Uloboridae and Araneidae (Herberstein et al. 2000). Within the Araneidae, silk decorations are scattered across several subfamilies: Gasteracanthinae, Cyrtophorinae, Araneinae, Argiopinae, and they occur in the genus Cearostris, a basal araneid of which the exact subfamily placement is ambiguous (Gregorič et al. 2015). If silk decorating is linked to wrap attacking, extra silk structures must be expected to occur primarily in species that preferably use this prey capture strategy. In my study, I now demonstrate in a behavioural comparison that this coherency exists. The results show that even within the same subfamily, Araneinae, Backobourkia brounii uses wrap attacks less frequently than Plebs bradleyi and, accordingly, builds no silk decorations. On the other hand, the evolution of the wrap attack strategy does obviously not inevitably lead to the evolution of silk decorating, as the former is more widespread in the araneid phylogeny than the latter (Scharff and Coddington 1997). Other factors like morphology and behavioural adaptations also open alternative evolutionary routes to fill ecological niches.

The higher capability of producing aciniform silk has been suggested to be a prerequisite to perform wrap attacks (Peters 1993). Accordingly, Peters (1993) suggested that a constant production of this silk type would lead to an excess if the wrapping activity is low, and that the spiders would use the extra silk to build decorations. This hypothesis was supported by experimental data by Tso (2004), who found that Argiope aetheroides reduces the web decorating activity after depletion of the aciniform silk glands. However, results on three other Argiope species are contradicting (Walter et al. 2008). The manipulative depletion of aciniform silk caused a significant increase in the decorating activity in subsequently built webs of A. bruennichi, A. sector and A. keyserlingi. Walter et al. (2008) argue that the silk removal causes an upregulation of the aciniform glands in order to adjust the production to higher demands for the use in wrap attacks, whenever there is a multiplication of prey capture events. This silk gland regulation hypothesis is further suggested to potentially mark the evolutionary origin of the silk decorating behaviour. The here presented link between wrap attacking and silk decorating across species is in line with this hypothesis, yet does not allow inferences about the underlying mechanism.

Araneus hamiltoni and Backobourkia brounii usually transport captured prey to either the web hub or the protective retreat. The two decorating species, Plebs bradleyi and Gea theridioides, often leave the immobilised prey items at the place in the web where they were caught and wrapped. In the case of not carrying the prey away after wrapping, spiders can often be observed to drag a substantially thick remnant of silk behind after finishing the last wrapping turn and when they return to the web hub. The conspicuousness of this observation is variable and thus hard to quantify objectively. Nevertheless, it occasionally creates a line of concentrated (wrapping) silk reaching from the point of capture to the centre of the web. This structure not only resembles the typical silk decoration bands in shape but also consists of the same material. Hence, apart from suggesting an intentional deposition of aciniform silk in the web to regulate the gland activity, such ‘carelessness’ may frequently create silk bands that become apparent to visually orientating animals. Without employing silk gland physiology, this observation describes another way of how silk decorating may have been directly originated from wrap attacking. However, like with the silk gland regulation hypothesis, it needs yet to be revealed what selective forces have shaped the deposited aciniform silk into a signal and how its evolution has detached from its non-signalling origin.

The number of signal functions described in the literature (see review, Walter and Elgar 2012), addressing prey, predator and non-predatory species as receivers, indicates a low specifity of the decoration signal, which may be a consequence of its unusual, non-signalling origin, but may also suggest that web decorations have no signal function at all (Walter and Elgar 2016). Nevertheless, there is evidence that silk decorations protect the web against damage by non-prey animals (Eisner and Nowicki 1983; Kerr 1993), that they attract prey (Craig and Bernard 1990; Tso 1998; Herberstein 2000; Cheng and Tso 2007), but also predators (Bruce et al. 2001; Seah and Li 2001), and, paradoxically, that they can deter predators (Schoener and Spiller 1992; Blackledge and Wenzel 2001). For the latter, many orb-web spiders use protective retreats to reduce the predation risk. Interestingly, silk decorating species do not construct those retreats (Scharff and Coddington 1997). The constant exposure to potential predators suggests that the decoration signal might be of protective nature (Walter and Elgar 2016), ‘replacing’ the retreat. If silk decorations act as prey attractants (Craig and Bernard 1990; Tso 1998; Herberstein 2000; Cheng and Tso 2007), why should retreat-constructing species not take advantage of those structures? Alternatively, it can be argued that, given the demonstrated link between web decorating and wrap attacking, species without retreats have a greater use of wrap attacks. This may seem puzzling, but diurnality may partly explain it. While the non-decorating, retreat-building species A. hamiltoni and B. brounii are mainly nocturnal, the decorators P. bradleyi and G. theridioides are hub-dwelling and mainly forage during the day. While defensive prey like grasshoppers and hymenopterans are only active during the day, the prey spectrum at night may consist of much less aggressive species that can be approached directly by biting. Admittedly speculative, the causal chain may look like the following: Less defensive prey is active during the night, rendering the use of wrap attacks less important for nocturnal orb-web spiders. Since for diurnal species the more defensive and perhaps relatively larger prey may bear the risk of escaping the web or injuring the host spider, some orb weavers evolved the wrap attack strategy together with a hub-dwelling lifestyle to be able to more quickly respond to prey impacts. And, the silk decorating behaviour may in some cases be the visible result.

After all, a link between the preferential use of wrap attacks and the construction of silken web decorations exists in orb web spiders. This link is evident in species from different araneid clades (P. bradleyi and G. theridioides), with presumably independent origins of silk decorating. In conclusion, I propose that the silk decorating behaviour of orb-web spiders originated as a by-product of the evolution of wrap attacking, and that web decorations have subsequently evolved various beneficial signal functions. Future studies should aim to reveal the underlying mechanisms of the linkage and unravel how the evolution of the decoration signal has separated from its non-signalling origin.

Notes

Acknowledgements

I thank the Deutsche Forschungsgemeinschaft (DFG) for financial support (Grant No. WA2637/2-1). I am very grateful for the support of the lab of Mark Elgar (University of Melbourne), hosting me for the time of the experiments and having created a joyful working environment. I thank Mark Elgar for helpful comments on the manuscript and for always being such a supportive and encouraging mentor, co-operator and friend over the years.

Compliance with ethical standards

Conflict of interest

The author declares that they have no conflict of interest.

Supplementary material

10682_2018_9930_MOESM1_ESM.pdf (141 kb)
Supplementary material 1 (PDF 140 kb)

References

  1. Blackledge TA (1998) Stabilimentum variation and foraging success in Argiope aurantia and Argiope trifasciata (Araneae: Araneidae). J Zool 246:21–27CrossRefGoogle Scholar
  2. Blackledge TA, Wenzel JW (2001) Silk mediated defense by an orb web spider against predatory mud-dauber wasps. Behaviour 138:155–171CrossRefGoogle Scholar
  3. Blamires SJ, Hochuli DF, Thompson MB (2008) Why cross the web: decoration spectral properties and prey capture in an orb spider (Argiope keyserlingi) web. Biol J Linn Soc 94:221–229CrossRefGoogle Scholar
  4. Bruce MJ (2006) Silk decorations: controversy and consensus. J Zool 269:89–97CrossRefGoogle Scholar
  5. Bruce MJ, Herberstein ME, Elgar MA (2001) Signalling conflict between prey and predator attraction. J Evol Biol 14:786–794CrossRefGoogle Scholar
  6. Cheng R-C, Tso I-M (2007) Signaling by decorating webs: Luring prey or deterring predators? Behav Ecol 18:1085–1091CrossRefGoogle Scholar
  7. Cheng R-C, Yang E-C, Lin C-P, Herberstein ME, Tso I-M (2010) Insect form vision as one potential shaping force of spider web decoration design. J Exp Biol 213:759–768CrossRefPubMedGoogle Scholar
  8. Chou I-C, Wang P-H, Shen P-S, Tso I-M (2005) A test of prey-attracting and predator defence functions of prey carcass decorations built by Cyclosa spiders. Anim Behav 69:1055–1061CrossRefGoogle Scholar
  9. Craig CL, Bernard GD (1990) Insect attraction to ultraviolet reflecting spider webs and web decorations. Ecology 71:616–623CrossRefGoogle Scholar
  10. Eberhard WG (2003) Substitution of silk stabilimenta for egg sacs by Allocyclosa bifurca (Araneae: Araneidae) suggests that silk stabilimenta function as camouflage devices. Behaviour 140:847–868CrossRefGoogle Scholar
  11. Eisner T, Nowicki S (1983) Spider web protection through visual advertisement: role of the stabilimentum. Science 219:185–187CrossRefPubMedGoogle Scholar
  12. Endler JA (1992) Signals, signal conditions, and the direction of evolution. Am Nat 139:125–153CrossRefGoogle Scholar
  13. Endler JA, Day LB (2006) Ornament colour selection, visual contrast and the shape of colour preference functions in great bowerbirds, Chlamydera nuchalis. Anim Behav 72:1405–1416CrossRefGoogle Scholar
  14. Endler JA, Théry M (1996) Interacting effects of lek placement, display behavior, ambient light, and color patterns in three neotropical forest-dwelling birds. Am Nat 148:421–452CrossRefGoogle Scholar
  15. Foelix RF (2011) Biology of spiders. Oxford University Press, OxfordGoogle Scholar
  16. Framenau VW, Dupérré N, Blackledge TA, Vink CJ (2010) Systematics of the new Australasian orb-weaving spider genus Backobourkia (Araneae: Araneidae: Araneinae). Arthropod Syst Phylogeny 68:79–111Google Scholar
  17. Gregorič M, Agnarsson I, Blackledge TA, Kuntner M (2015) Phylogenetic position and composition of Zygiellinae and Caerostris, with new insight into orb-web evolution and gigantism. Zool J Linn Soc-Lond 175:225–243CrossRefGoogle Scholar
  18. Griswold CE, Coddington JA, Hormiga G, Scharff N (1998) Phylogeny of the orb-web building spiders (Araneae, Orbiculariae: Deinopoidea, Araneoidea). Zool J Linn Soc Lond 123:1–99CrossRefGoogle Scholar
  19. Hayashi CY, Blackledge TA, Lewis RV (2004) Molecular and mechanical characterization of aciniform silk: uniformity of iterated sequence modules in a novel member of the spider silk fibroin gene family. Mol Biol Evol 21:1950–1959CrossRefPubMedGoogle Scholar
  20. Herberstein ME (2000) Foraging behaviour in orb-web spiders (Araneidae): do web decorations increase prey capture success in Argiope keyserlingi Karsch, 1878? Austral J Zool 48:217–223CrossRefGoogle Scholar
  21. Herberstein ME, Craig CL, Coddington JA, Elgar MA (2000) The functional significance of silk decorations of orb-web spiders: a critical review of the empirical evidence. Biol Rev 75:649–669CrossRefPubMedGoogle Scholar
  22. Joseph MM, Framenau VW (2012) Systematic review of a new orb-weaving spider genus (Araneae: Araneidae), with special reference to the Australasian-Pacific and South-East Asian fauna. Zool J Linn Soc Lond 166:279–341CrossRefGoogle Scholar
  23. Kerr AM (1993) Low frequency of stabilimenta in orb webs of Argiope appensa (Araneae: Araneidae) from Guam: An indirect effect of an introduced avian predator? Pac Sci 47:328–337Google Scholar
  24. McCook HC (1889) American spiders and their spinning work, vol 1. Academy of Natural Sciences, PhiladelphiaGoogle Scholar
  25. Müller MC, Westheide W (1993) Comparative morphology of the sexually dimorphic orb-weaving spider Argiope bruennichi (Araneae: Araneidae). Mem Queensl Mus 33:615–620Google Scholar
  26. Nakata K (2009) To be or not to be conspicuous: the effects of prey availability and predator risk on spider’s web decoration building. Anim Behav 78:1255–1260CrossRefGoogle Scholar
  27. Olive CW (1980) Foraging specializations in orb-weaving spiders. Ecology 61:1133–1144CrossRefGoogle Scholar
  28. Olive CW (1981) Optimal phenology and body-size of orb-weaving spiders: foraging constraints. Oecologia 49:83–87CrossRefPubMedGoogle Scholar
  29. Peters HM (1993) Über das problem der stabilimente in spinnennetzen. Zool Jahrb Allg Zool 97:245–264Google Scholar
  30. Robinson MH, Olazarri J (1971) Units of behavior and complex sequences in the predatory behavior of Argiope argentata (Fabricius): (Araneae: Araneidae). Smithson Contr Zool 65:1–36CrossRefGoogle Scholar
  31. Robinson MH, Mirick H, Turner O (1969) The predatory behaviour of some araneid spiders and the origin of immobilization wrapping. Psyche 76:487–501CrossRefGoogle Scholar
  32. Schaedelin FC, Taborsky M (2009) Extended phenotypes as signals. Biol Rev 84:293–313CrossRefPubMedGoogle Scholar
  33. Scharff N, Coddington JA (1997) A phylogenetic analysis of the orb-weaving spider family Araneidae (Arachnida, Araneae). Zool J Linn Soc 120:355–434CrossRefGoogle Scholar
  34. Schoener TW, Spiller DA (1992) Stabilimenta characteristics of the spider Argiope argentata on small islands: support of the predator-defense hypothesis. Behav Ecol Sociobiol 31:309–318CrossRefGoogle Scholar
  35. Seah WK, Li D (2001) Stabilimenta attract unwelcome predators to orb-webs. Proc R Soc Lond B 268:1553–1558CrossRefGoogle Scholar
  36. Starks PT (2002) The adaptive significance of stabilimenta in orb-webs: a hierarchical approach. Ann Zool Fennici 39:307–315Google Scholar
  37. Tan EJ, Seah SWH, Yap L-MYL, Goh PM, Gan W, Liu F, Li D (2010) Why do orb-weaving spiders (Cyclosa ginnaga) decorate their webs with silk spirals and plant detritus? Anim Behav 79:179–186CrossRefGoogle Scholar
  38. Tso I-M (1998) Isolated spider web stabilimentum attracts insects. Behaviour 135:311–319CrossRefGoogle Scholar
  39. Tso I-M (2004) The effect of food and silk reserve manipulation on decoration building of Argiope aetheroides. Behaviour 141:603–616CrossRefGoogle Scholar
  40. Vollrath F, Knight DP (2001) Liquid crystalline spinning of spider silk. Nature 410:541–548CrossRefPubMedGoogle Scholar
  41. Walter A, Elgar MA (2012) The evolution of novel animal signals: silk decorations as a model system. Biol Rev 87:686–700CrossRefPubMedGoogle Scholar
  42. Walter A, Elgar MA (2016) Signal polymorphism under a constant environment: the odd cross in a web decorating spider. Sci Nat 103:93CrossRefGoogle Scholar
  43. Walter A, Elgar MA, Bliss P, Moritz RFA (2008) ‘Wrap attack’ activates web decorating behavior in Argiope spiders. Behav Ecol 19:799–804CrossRefGoogle Scholar

Copyright information

© Springer International Publishing AG, part of Springer Nature 2018

Authors and Affiliations

  1. 1.Department of BiomedicineAarhus UniversityAarhus CDenmark
  2. 2.School of BioSciencesUniversity of MelbourneVictoriaAustralia

Personalised recommendations