Armoured spiderman: Morphological and behavioural adaptations of a specialised araneophagous predator (Araneae: Palpimanidae)

Abstract

In a predator-prey system where both intervenients come from the same taxon, one can expect a strong selection on behavioural and morphological traits involved in prey capture. For example, in specialised snake-eating snakes, the predator is unaffetced by the venom of the prey. We predicted that similar adaptations should have evolved in spider-eating (araneophagous) spiders. We investigated potential and actual prey of two Palpimanus spiders (P. gibbulus, P. orientalis) to support the prediction that these are araneophagous predators. Specific behavioural adaptations were investigated using a high-speed camera during staged encounters with prey, while morphological adaptations were investigated using electron microscopy. Both Palpimanus species captured a wide assortment of spider species from various guilds but also a few insect species. Analysis of the potential prey suggested that Palpimanus is a retreat-invading predator that actively searches for spiders that hide in a retreat. Behavioural capture adaptations include a slow, stealthy approach to the prey followed by a very fast attack. Morphological capture adaptations include scopulae on forelegs used in grabbing prey body parts, stout forelegs to hold the prey firmly, and an extremely thick cuticle all over the body preventing injury from a counter bite of the prey. Palpimanus overwhelmed prey that was more than 200% larger than itself. In trials with another araneophagous spider, Cyrba algerina (Salticidae), Palpimanus captured C. algerina in more than 90% of cases independent of the size ratio between the spiders. Evidence indicates that both Palpimanus species possesses remarkable adaptations that increase its efficiency in capturing spider prey.

Full text

ORIGINAL PAPER
Armoured spiderman: morphological and behavioural
adaptations of a specialised araneophagous predator
(Araneae: Palpimanidae)
Stano Pekár &Jan Šobotník &Yael Lubin
Received: 9 February 2011 /Revised: 27 March 2011 /Accepted: 30 April 2011
#Springer-Verlag 2011
Abstract In a predator–prey system where both interve-
nients come from the same taxon, one can expect a
strong selection on behavioural and morphological traits
involved in prey capture. For example, in specialised
snake-eating snakes, the predator is unaffetced by the
venom of the prey. We predicted that similar adaptations
should have evolved in spider-eating (araneophagous)
spiders. We investigated potential and actual prey of two
Palpimanus spiders (P. gibbulus,P. orientalis)tosupport
the prediction that these are araneophagous predators.
Specific behavioural adaptations were investigated using a
high-speed camera during staged encounters with prey,
while morphological adaptations were investigated using
electron microscopy. Both Palpimanus species captured a
wide assortment of spider species from various guilds but
also a few insect species. Analysis of the potential prey
suggested that Palpimanus is a retreat-invading predator
that actively searches for spiders that hide in a retreat.
Behavioural capture adaptations include a slow, stealthy
approach to the prey followed by a very fast attack.
Morphological capture adaptations include scopulae on
forelegs used in grabbing prey body parts, stout forelegs to
hold the prey firmly, and an extremely thick cuticle all
over the body preventing injury from a counter bite of the
prey. Palpimanus overwhelmed prey that was more than
200% larger than itself. In trials with another araneopha-
gous spider, Cyrba algerina (Salticidae), Palpimanus
captured C. algerina in more than 90% of cases indepen-
dent of the size ratio between the spiders. Evidence
indicates that both Palpimanus species possesses remark-
able adaptations that increase its efficiency in capturing
spider prey.
Keywords Prey specificity .Stenophagy .Predatory
behaviour .Trophic niche .Spider
Introduction
Trophic specialisation is not a common phenomenon
among terrestrial carnivorous predators. Examples include
aphidophagy, i.e., specialisation on aphids (e.g. Hodek and
Honěk1996), myrmecophagy (e.g. Montanucci 1989),
termitophagy (e.g. Richardson 1987), oniscophagy (e.g.
Dejean 1997), and araneophagy (e.g. Schulz and Wainer
1997). Studies of trophic specialisation in prey-specialised
species often reveal fascinating examples of adaptations
that are used in the effective utilisation of a narrow prey
niche (e.g. Pekár and Král 2002;Řezáčet al. 2008).
The evolution of trophic specialisation, at least in
herbivores, has been explained by a number of hypotheses,
Communicated by: Sven Thatje
Electronic supplementary material The online version of this article
(doi:10.1007/s00114-011-0804-1) contains supplementary material,
which is available to authorized users.
S. Pekár (*)
Department of Botany and Zoology, Faculty of Sciences,
Masaryk University,
Kotlářská 2,
611 37 Brno, Czech Republic
e-mail: pekar@sci.muni.cz
J. Šobotník
Institute of Organic Chemistry and Biochemistry,
Flemingovo nám. 2,
166 10 Prague 6, Czech Republic
Y. Lubin
Mitrani Department of Desert Ecology,
Blaustein Institute for Desert Research, Ben Gurion University,
84990, Sede Boqer Campus,
Beer-Sheva, Israel
Naturwissenschaften
DOI 10.1007/s00114-011-0804-1

such as increased physiological efficiency, the utilisation of
enemy-free space, optimal foraging, neural constraints and
trade-offs (Singer 2008). In some of these hypotheses, it is
assumed that trophic specialisation is the driving force for
the evolution of traits in the consumer that increase the
efficacy of searching for and handling the food. Evidence
for such adaptations, namely morphological changes in
beaks, was already revealed by Darwin in finches on
Galapagos (Darwin 1859). Evolution of morphological and
behavioural traits may lead to remarkable adaptations, in
particular when the target prey is difficult to find or handle.
This is the case when both predator and prey are from the
same taxon and are of similar size, such as snake-eating
snakes, firefly-eating fireflies or araneophagous spiders.
Then, selection should improve traits used in capture and in
preventing counter-attack by the dangerous prey. Ophioph-
agous snakes, for example, are immune to the venom of
their prey (Underwood 1997). Females of fireflies imitate
signals of other species to attract and devour firefly males
(Lloyd 1975). Araneophagous Portia spiders use cryptic
stalking to avoid being recognised as a predator and
perform a sudden attack (Tarsitano et al. 2000). In the case
of the spider prey, the preys possess chelicerae empowered
by potent venom to which the spider predators are
susceptible (Jackson 1992).
Spiders are carnivorous predators and most species are
euryphagous (Nentwig 1987). Many include other spider
species or conspecifics in their diet (e.g. Li and Jackson
1996; Guseinov 2006; Heuts and Brunt 2001). In contrast
to this facultative araneophagy, obligatory spider-eating is
far less common, though present in a few families, namely
Mimetidae, Salticidae, Palpimanidae, Scytodidae, Gnapho-
sidae and Archaeidae (Legendre 1961; Cutler 1972; Jarman
and Jackson 1986; Li et al. 1999; Cerveira and Jackson
2005). Such species may use aggressive mimicry, i.e., the
imitation of ensnared prey (e.g. Jackson 1990a) or the
male’s courtship signals (Jackson and Wilcox 1990)to
attract and capture the prey spider. Evidence gathered from
the salticid Portia fimbriata (Doleschall), the best-studied
araneophage, shows that it uses fascinating behavioural
adaptations that minimise detection and identification by its
prey and thereby prevent a counter-attack. P. fimbriata
engages in cryptic stalking during encounters with other
cursorial species (Clark and Jackson 2000) or exploits
situations in which the web-building spider’s ability to
detect the predator’s approach is impaired by wind or by
the struggling of ensnared prey that create a “smoke-
screen”for approach of P. fimbriata (Wilcox et al. 1996).
Nevertheless, evidence for morphological adaptations to
araneophagy is lacking.
Here, we focus on species of the family Palpimanidae.
This family is represented by more than 100 species
occurring in subtropical and tropical zones all over the
world, except for Australia (Platnick 2011). Palpimanid
spiders are rare; thus, our knowledge of their biology is
extremely fragmentary (Jocqué and Dippenaar-Schoeman
2006). There are a few anecdotal reports of araneophagy in
two Mediterranean species (Murphy 1991)andinan
African species (Henschel 1997;HenschelandLubin
1997), but there is only a single analysis of the natural
prey of an African species, Palpimanus sp., that supports its
araneophagous habit (Cerveira and Jackson 2005).
We investigated two species of the genus Palpimanus,
P. gibbulus Dufour from Portugal and Palpimanus orientalis
Kulczyński from Israel. Both species were extremely rare so
we combined data for the two species. First, we gained
evidence to support expected araneophagy by means of
studying prey capture efficiency under laboratory conditions
and potential prey in nature. We then conducted a series of
experiments with different types of prey in order to reveal
behavioural and morphological adaptations used in prey
capture. Using a high-speed camera, electron microscopy
and manipulative design, we studied the function and
efficacy of these adaptations.
Methods
Prey data
Direct observations of prey capture by Palpimanus in
nature are very rare due to extremely low abundance and
nocturnal hunting activity. Therefore, we used alternative
approaches to gain evidence for their prey. To reveal
potential natural prey of P. orientalis in Israel, we
investigated their association with other spiders in the
Negev desert. We used data gained for a biodiversity study
(see Pekár and Lubin 2003 for more details). Ten sites (45
locations) were chosen for the investigation in the northern
and central Negev desert, Israel. The sites represented
different habitats of the desert ecosystem. Spiders were
collected using dry pitfall traps consisting of two plastic
cups—one inside the other (10 cm diameter by 10 cm
deep). In each location, 30–100 traps were installed. Traps
wereexposedfor3days/monthduring1990–1992 and
checked each morning to retrieve spiders and release other
organisms. After combining the data from 3 years, there
were five to 27 collection periods in total at each location,
though not for the full 3 years at all sites. Trapped spiders
(13,755 individuals belonging to 125 species) were
collected and identified in the laboratory. Data on stand-
ardised species abundance (higher than 5%) were sub-
jected to detrended correspondence analysis within
CANOCO (Ter Braak and Šmilauer 2002)inorderto
reveal the association between P. orientalis and other
spider species.
Naturwissenschaften

To reveal potential natural prey of P. gibbulus in
Portugal, we searched for these spiders under stones in
Alentejo and the Algarve between 2003 and 2009. When a
Palpimanus was found under a stone, we recorded the
frequency of other spiders present under the same stones.
The study was performed at Corte da Velha and Mésquita
(near Mértola, Alentejo) and Alcoutim (Algarve) in
November and April of 2003, 2005, 2008 and 2009.
Predatory experiments
To investigate the fundamental trophic niche, we conducted
staged predatory encounters under laboratory conditions
using a variety of prey (Tables 1and 2). Altogether, 92
individuals of P. gibbulus and 65 individuals of P. orientalis
at various stages of their development were tested. In the
acceptance trials, we placed Palpimanus spiders singly in a
Petri dish (diameter 45 mm) 1 day prior to the trial. The
size of Palpimanus (8–10 mm in adult stage) and of the
prey (total body length in both) were measured before each
trial. The prey was offered to a given spider at 10-day
intervals, as prey offered sooner than this was not accepted
presumably due to satiation. In all other trials, similar
fasting period was used. The different preys were offered to
the spiders in a random order. A repeated-measures design
was used, in which each Palpimanus spider was used
repeatedly for several prey species, because it provides
higher efficiency of the test by using fewer spiders than a
complete random design (Mead 1988). If the spider did not
attack the prey within 2 h, the prey was replaced by a
different one. In each trial, we recorded whether the spider
attacked and captured the prey. The capture success was
compared among prey types and spider guilds (Table 1)by
means of generalised linear models (GLM) with binomial
errors (Pekár and Brabec 2009). Due to the repeated-
measures design and resulting overdispersion, a quasibino-
mial setting (GLM-qb) was used in order to correct for too
positive pvalues favouring acceptance of the alternative
hypothesis (Hardin and Hilbe 2003). Logit model within
GLM was also used to model the effect of size ratio (total
Family Subfamily, genus or species Guild % Captured Number
Agelenidae Tegenaria sp. Web-building 87 38
Anyphaenidae Anyphaena accentuata (Walckenaer) Cursorial 92 13
Araneidae Araniella sp. Web-building 73 15
Atypidae Atypus sp. Retreat-web 100 10
Corinnidae Liophrurillus flavitarsis (Lucas) Cursorial 20 25
Dictynidae Dictyna sp. Web-building 100 10
Dysderidae Dysdera crocata C. L. Koch Cursorial 100 10
Eresidae Stegodyphus lineatus (Latreille) Web-building 76 42
Filistatidae Filistata sp. Retreat-web 100 10
Gnaphosidae Drassodes sp. Cursorial 80 10
Pterotricha sp. Cursorial 73 11
Lycosidae Pardosa agrestis (Westring) Cursorial 98 64
Linyphiidae Erigoninae Web-building 100 15
Miturgidae Cheiracanthium sp. Cursorial 92 13
Nesticidae Nesticus cellulans (Clerck) Web-building 90 10
Oecobiidae Oecobius sp. Retreat-web 67 43
Philodromidae Philodromus sp. Cursorial 80 10
Thanatus sp. Cursorial 85 13
Tibellus sp. Cursorial 90 10
Prodidomidae Prodidomus sp. Cursorial 40 10
Salticidae Cyrba algerina (Lucas) Cursorial 91 42
Evarcha sp. Cursorial 70 10
Morgus sp. Cursorial 87 15
Tetragnathidae Tetragnatha sp. Web-building 100 10
Theridiidae Steatoda sp. Web-building 60 10
Theridion sp. Web-building 57 14
Thomisidae Xysticus sp. Cursorial 94 32
Zodariidae Selamia reticulata (Simon) Cursorial 80 10
Zodarion styliferum (Simon) Cursorial 69 29
Tab le 1 Overview of spider
prey species used in the accep-
tance experiments and trials
with Oecobius sp., P. agrestis
and C. algerina. Percent cap-
tured, guild name (Uetz et al.
1999) and number are given for
each species used
Naturwissenschaften

body length of prey to total body length of Palpimanus)on
capture success.
Some trials (N=20) of the acceptance experiments were
recorded at 25 frames/s on a standard video camera (Sony
SSC-DC80P connected to Sony GV-HD700 recorder)
attached to the stereomicroscope. From these trials, we
measured the direction of approach, approaching speed
and distance from prey before attack. As the attack
sequence was very fast (less than a second), it was filmed
in other trials with a high-speed B/W camera (IDT
MotionXtra N3) using 500 frames/s. For the trials
recorded by high-speed camera, we used a small dish
(25 mm diameter) in order to maximise encounter rate
between Palpimanus andthespiderandtoconstraintheir
movement. As prey, we used Pardosa agrestis (Westring;
Lycosidae) spiders. Starved Palpimanus spiders (subjected
to 10-day fasting period) were put into dishes a day before
the trial. A slow replaying of the footage allowed us to
recognise behavioural events. We performed 63 trials with
P. a g r e s t i s spiders. From each piece of footage, we
determined the distance frompreyattheonsetofattack
(i.e., when the spider lunged toward prey), the body part
that Palpimanus grabbed with its forelegs, the body part it
bit with chelicerae, the duration of predatory sequence (i.e.,
the time between onset of attack and the bite) and capture
success. The chi-square test of homogeneity was used to
compare frequencies of grabbing and attacking among prey
body parts.
Trials with an araneophagous salticid C. algerina (N=42)
and an oecobiid Oecobius sp. (N=43), two potential prey
species of P. gibbulus, were performed in Petri dishes
(diameter 35 mm). The prey spiders were collected in
Portugal at the site of P. gibbulus occurrence in Alentejo
(Mértola). Various developmental stages of both species, C.
algerina and Oecobius,werecollected.Theywerehoused
singly in a Petri dish (diameter 35 mm) 2 days prior to the
trial. This time was sufficient for Oecobius individuals to
build a web. Both C. algerina and Oecobius were fed
with Drosophila flies 3 days prior to the trial. Palpimanus
spiders of variable size with a ratio (total body length of
prey to total body length of Palpimanus) varying from 0.5
to 1.8 were used. Palpimanus spiders were released singly
into the dish hosting the prey spiders. If Palpimanus did
not attack the prey within 2 h, the trial was terminated. For
each trial, we measured the size of prey and Palpimanus
and the capture success. In trials with Oecobius,wealso
recorded where (inside or outside the web) the capture
took place. The effect of size ratio on the capture success
was modelled using GLM with binomial errors (GLM-b)
for each prey type separately.
To investigate the role of scopulae during the capture,
the scopulae on both forelegs of 17 juvenile Palpimanus
spiders were glued with paraffin using a fine soldering gun.
The spiders were anaesthetised using CO
2
for 5 min prior to
the treatment. The capture efficacy of ablated individuals
was compared with the efficacy of another 17 individuals
without glue that were used as a control. These individuals
were also starved for 10 days. We used P. agrestis spiders
(matching the size of each Palpimanus individual) as prey
using a design similar to that used for the acceptance
experiments. In each trial, we recorded the number of
attacks (i.e., attempts to grab prey) until the prey was
captured. Recorded counts were then compared between the
two treatment levels by means of GLM with Poisson errors
(GLM-p).
Morphology and anatomy
The two Palpimanus species have very similar morphology
and can be distinguished only at the adult stage by the shape
of sexual organs (Platnick 1981). The morphology of the
scopulae that are situated on the forelegs of both species was
studied first under a stereomicroscope. As we found no
difference between the two species we used only P. gibbulus
for detailed study using a scanning electron microscope
(SEM) Jeol JSM-6380LV. Forelegs from three individuals
were placed on a stub, coated with gold and photographed.
We subsequently analyzed the size and the density of setae
from SEM pictures by means of image analysis software
(SigmaScan).
To compare the thickness and the composition of the
cuticle of Palpimanus and their prey (P. a g r e s t i s and a
salticid Evarcha arcuata (Cerck)), we investigated the
cuticle of five individuals of each species using a standard
Order/Family Species or genus % Captured Number
Collembola/Entomobryidae Sinella curviseta Brook 60 10
Ensifera/Gryllidae Acheta domesticus Linnaeus 23 13
Isoptera/Rhinotermitidae Reticulitermes sp. 9 53
Lepidoptera/Pyralidae Plodia interpunctella (Hübner) 0 10
Coleoptera/Chrysomelidae Phyllotreta sp. 0 11
Hymenoptera/Formicidae Aphaenogaster senilis Mayr 4 27
Diptera/Drosophilidae Drosophila melanogaster Meigen 50 10
Tab l e 2 Overview of insect
prey used in the acceptance
experiments. Percent capture
and number is given for each
prey species used. Imagoes were
used in all species
Naturwissenschaften

protocol (Reynolds 1963;Spurr1969). Specimens were
stored in 75% ethanol, and a piece of femur, an
opisthosoma and a prosoma from each were studied. The
body parts were embedded into standard Spurr resin
(without fixation or postfixation), and, after polymeriza-
tion, semithin (1 μm thick) and ultrathin (about 70 nm
thick) sections were prepared using a Reichert Ultracut
ultramicrotome. Semithin sections were stained with
Azure II, and studied using a Zeiss Amplival optical
microscope combined with a Canon EOS 500D camera.
Ultra-thin sections were stainedwithuranylacetateand
lead citrate (standard recipe) and studied using a Jeol 1011
transmission electron microscope. This combined ap-
proach allowed us to compare the thickness of the whole
cuticle and also particular layers making up the cuticle. We
followed Dalingwater (1987) in the definition of layers of
cuticle. ANOVA was used to compare the thicknesses of
particular cuticular layers among three spider species as
the measures came from a symmetric distribution away
from zero.
All univariate analyses were performed within R (R
Development Core Team 2010).
Results
Trophic niche
In the Negev desert, P. orientalis spiders (N=48) were
captured in pitfall traps most frequently with 22 spider
species (Fig. 1). These included seven cursorial (Gnapho-
sidae, Hersiliidae, Lycosidae, Nemesiidae, Philodromidae,
Thomisidae and Zodariidae) and one web-building family
(Theridiidae). In Portugal, P. gibbulus spiders were found
under stones together with several cursorial families:
Sparassidae (44%, N=72), Salticidae (14%), Zodariidae
(13%), Gnaphosidae (5%), Sicariidae (6%), Corinnidae
(3%), Scytodidae (1%); and two web-building families:
Oecobiidae (11%) and Pholcidae (1%).
In the laboratory, Palpimanus captured a wide assort-
ment of spider species, from various families and of various
guilds (Table 1). The relative frequency of capture success
was not significantly different among three guilds (GLM-
qb, F
2,25
=0.2, P=0.85). Besides spiders, Palpimanus also
attacked other prey, such as flies and crickets, but at a
significantly lower frequency than spiders (all spider
species pooled 79.5 whereas all insect species pooled
20.9%; GLM-qb, F
1,33
=35.6, P<0.0001).
Palpimanus attacked a wide range of predator–prey
sizes, with a ratio ranging from 0.3 to approx. 2. The attack
success was not significantly related to the size ratios
(GLM-qb, F
1,369
<0.1, P=0.8). Thus, Palpimanus captured
prey that was even 200% longer than itself.
Predatory behaviour
The behaviour of juvenile and adult Palpimanus spiders
was similar to that described in Cerveira and Jackson
(2005). Palpimanus moved at an average speed of
0.2 cm s
−1
(SE=0.02, N= 10). Once Palpimanus recognised
a lycosid (P. agrestis) spider-prey moving nearby it raised
its forelegs (Figs. 2a and 3a, Video 1) and very slowly
approached it from any direction. It continued approaching
until reaching on average a distance of 0.55 (SE =0.04) of
its body length from the body (prosoma or opisthosoma) of
the prey. Then, it stopped and remained motionless with
raised forelegs (Fig. 3a, b). Once the prey made a
movement, Palpimanus immediately lunged to the prey
and placed its forelegs on any part of the prey body, but
significantly more often on the legs (67%, N=63) than on
the prosoma or opisthosoma (X
22
=26.5, P<0.0001). Palpi-
manus grabbed the prey body part with its forelegs and put
it in his chelicera for a bite (Video 2). The time from lunging
to bite took on average 0.2 s (SE=0.01). Palpimanus
0.15.1-
-1.5 1.0
P. cyprius L. blackwalli
R. expers
Z. cyrenaicum A. pallens
B. plumalis
D. lutescens
E. macrochelis
H. pugnans
Hersiliola sp.
Hogna sp.
M. simeonica
Nemesia sp.
N. aussereri
O. judaea
P. conspersa
S. latifasciata
T. fabricii
Thanatus sp.
X.bliteus
X.gymnocephalus
X.lalandei
Xysticus sp.
Fig. 1 Ordination diagram of CCA showing the association of
Palpimanus orientalis and several spider species. The first two
eigenvalues were 0.17 and 0.15. The potential prey species displayed
are (Family: species): Gnaphosidae: Anagraphis pallens Simon, Berlan-
dina plumalis (O. P.-Cambridge), Drassodes lutescens (C. L. Koch),
Haplodrassus pugnans (Simon), Minosia simeonica Levy, Nomisia
aussereri (L. Koch), Pterotricha conspersa (O.P.-Cambridge);Hersilli-
dae: Hersioliola sp.; Lycosidae: Hogna sp.; Nemesiidae: Nemesia sp.;
Philodromidae: Thanatus fabricii (Audouin), Thanatus sp.; Theridiidae:
Enoplognatha macrochelis Levy & Amitai, Steatoda latifasciata
(Simon); Thomisidae: Ozyptila judaea Levy, Xysticus bliteus (Simon),
Xysticus gymnocephalus Strand, Xysticus lalandei (Audouin), Xysticus
sp.; Zodariidae: Lachesana blackwalli (O. P.-Cambridge), Ranops
expers (O. P.-Cambridge), Zodarion cyrenaicum Denis
Naturwissenschaften

delivered a bite significantly more frequently to the legs
(65%, N=63) than to the prosoma or opisthosoma (X
22
=
28.7, P<0.0001). In 29.5% (N=41) of cases, the prey
autotomized its leg and escaped. The average capture
success at the first attack was 65.6% (N=64). If the prey
did not autotomize a leg, Palpimanus grabbed the prosoma
with both forelegs after the bite to a leg and held it tightly for
a few minutes (Fig. 3c). Then Palpimanus released the prey,
laid silk threads around it on the ground and began to feed
on it.
In trials with oecobiid (Oecobius) spiders Palpimanus
captured them in 67.4% (N=43) of the trials using a similar
behaviour as with the other prey. In most cases (60%),
Palpimanus captured the oecobiid outside its web. In 40%
of cases, Palpimanus entered the web and captured the
oecobiid right after entering, or if the oecobiid escaped,
Palpimanus waited inside the web for its return. The size
ratio between the Palpimanus and the oecobiid prey
(ranging from 0.5 to 1) had no effect on the capture
success (GLM-qb, F
1,41
=0.1, P=0.77).
In trials with an araneophagous salticid (C. algerina),
Palpimanus captured the salticid in 90.5% (N=42) of cases
using a similar behaviour as with the other prey. In the
remaining cases, the salticid captured Palpimanus. There
was no effect of size ratio (varying from 0.6 to 1.8, with
43% of values below 1) on the capture success (GLM-qb,
F
1,40
<0.01, P=0.91).
Scopulae
Palpimanus spiders possess massive forelegs. We observed
that during the attack, the forelegs were used first to grab a
body part of the prey such as a leg segment and then the
whole body. The tibia, metatarsi and tarsi of the first pair of
legs of Palpimanus bear a row of scopulae, with 15.7 (SE=
5.01) hairs per 0.01 square millimetre on the prolateral side
(Fig. 4). Each seta is terminated with a pad-like structure
bearing a dense cover of 6.4 (SE=2.1) tiny hairs per 1 μm.
The scopulae were clearly used to attach to the prey body.
The adhesive power of the scopulae was seen in nine cases,
when a leg remained attached to the scopulae after an attack
(Video 2).
Experimental sealing of the pad-like scopulae led to
decreased capture efficiency. While normal Palpimanus
spiders used on average 1.4 attacks to catch its prey,
spiders with sealed scopulae needed a significantly higher
number of attacks (5.9) pre prey catch (GLM-qp, F
1,32
=
12.8, P=0.001).
Fig. 2 Capture sequence on a
thomisid. aPalpimanus slowly
approaches the thomisid with
raised forelegs and gently
touches the prey. bPalpimanus
positions itself as near as possi-
ble to the thomisid with raised
outstretched forelegs and open
chelicera before the sudden at-
tack. cGrab of prey followed by
bite into the thomisid prosoma
Fig. 3 Detailed events of the predatory behaviour from the high-
speed camera. aAs the prey approaches, Palpimanus raises its
forelegs. bPalpimanus grabs the second leg (tibia) of the prey with its
scopulae on the metatarsi of the forelegs (arrow). cPalpimanus brings
the prey leg to its chelicera and bites the tibia of the prey leg (arrow).
dWith the Palpimanus still holding the leg in its chelicera, the prey
amputates the leg and escapes
Naturwissenschaften

Cuticle
We recorded 43 attacks of spider prey on Palpimanus
(excluding the trials with C. algerina). Of these only five
attacks were successful (0.9%, N=554)—resulting in the
death of Palpimanus. This happened only when the prey
spider was longer by more than 50%, specifically when the
prey was Stegodyphus (Eresidae), Tegenaria (Agelenidae),
Pterotricha (Gnaphosidae) or Steatoda (Theridiidae). While
the former three genera attacked Palpimanus directly on the
opisthosoma, the latter two first restrained Palpimanus with
threads or bands of silk.
The cuticle thickness of the prosoma and femora of
Palpimanus was more than twice that of its prey spiders
and the total cuticle layer of the opisthosoma of Palpima-
nus was nearly five times thicker than that of its prey
(Table 3). The cuticle is composed of three layers, the
epicuticle, exocuticle and endocuticle. In the prosoma, the
largest difference was in the thickness of the endocuticle:
that of Palpimanus was four times thicker than those of
prey (Fig. 5). On the femora, the exocuticle of Palpima-
nus was thinner than those of prey, but the endocuticle of
Palpimanus was three times thicker. On the opisthosoma,
the exocuticle of Palpimanus was seven times thicker than
that of its prey and the endocuticle was three times
thicker.
Discussion
The fundamental trophic niche of two Palpimanus species,
identified by means of acceptance experiments, includes
spiders of various families and guilds and even a few
insects orders. However, prey other than spiders were
captured at a significantly lower frequency, suggesting that
the two Mediterranean Palpimanus species feed primarily
on spiders. This is consistent with results of natural prey of
the African Palpimanus sp. in which 95% of prey were
spiders (Cerveira and Jackson 2005). We expected that the
spiders found together with Palpimanus under the same
stone are potential prey species. Among the potential prey
occurring on the ground, there were both diurnal (e.g.
thomisids, salticids and philodromids) and nocturnal (e.g.
gnaphosids, zodariids, sparassids) spider species not only
Fig. 4 a Scopulae on the meta-
tarsi and tarsi of forelegs of P.
gibbulus.bDetail of the
pad-like scopular setae
Palpimanus P. agrestis E. arcuata ANOVA
Prosoma
Epicuticle 0.39± 0.09 0.16±.01
a
0.29± .04 0.002
Exocuticle 6.53± .20 4.70±.16
a
4.97± .26
a
<0.0001
Endocuticle 46.70±1.60 12.55 ±.53
a
11.21± 1.26
a
<0.0001
Total 53.61± 1.62 17.41±.49
a
16.47± 1.44
a
<0.0001
Femur
Epicuticle 0.15± .02 0.15±.01 0.22 ±.03 0.13
Exocuticle 1.41± .07 2.34±.06
a
3.50± .33
a
<0.0001
Endocuticle 21.25 ±1.37 7.46± .16
a
6.12± .11
a
<0.0001
Total 22.80 ±1.35 9.95±.18
a
9.83± .33
a
<0.0001
Opisthosoma
Epicuticle 0.02± .01 0.11±.01
a
0.17± .01
a
<0.0001
Exocuticle 39.73± 2.31 5.97±.74
a
8.28± .41
a
<0.0001
Endocuticle 12.93± .41 4.49±.21
a
4.29± .20
a
<0.0001
Total 52.67± 2.52 10.56±.88
a
12.73± .34
a
<0.0001
Tab l e 3 Comparison of the
thickness (mean ± SE) of cuti-
cle layers in P. gibbulus and its
prey (Pardosa agrestis,Evarcha
arcuata) spiders and the results
of ANOVA. Post hoc compar-
isons were based upon treatment
contrasts, i.e., difference in the
thickness from P. gibbulus.
a
Assign values significantly dif-
ferent from those of Palpimanus
Naturwissenschaften

from the cursorial guild but also a few web-building species
(theridiids, oecobiids). We expect that Palpimanus may
feed on a wide range of spider species, as all these species
were captured in acceptance experiments. Palpimanus
moves slowly and could hardly catch up with some spiders,
such as salticids, gnaphosids or lycosids, which move about
ten times faster (Michalková and Pekár 2009). The high-
capture efficiency in acceptance trials was likely due to
limited space in a dish, which prevented prey from hiding
or escaping from the predator. In nature, these prey spiders
may be captured when inactive, i.e., during the night or
duringmoultingwhenhidingincrevicesorretreats.
Gnaphosids, salticids and sparassids are known to build
silken retreats, and zodariids build burrow or igloo-shaped
retreats (Jocqué and Dippenaar-Schoeman 2006). The
capture tactic used to catch Oecobius spiders hidden within
two parallel sheets of silk suggests that Palpimanus can
invade web retreats. Indeed, in Israel, P. orientalis was
found in nests of the eresid spider Stedodyphus lineatus
(Lubin, pers. observ.).
Similar to the gnaphosid spiders, Ta ier i a ere bus (L. Koch,
1873) (Jarman and Jackson 1986)andPoecilochroa senilis
(O. P.-Cambridge, 1872) (Whitehouse and Lubin 1998),
Palpimanus may specialise on spiders hiding in burrows or
silken retreats. Retreat-dwelling species with a single escape
opening blocked by the penetrating Palpimanus would
Fig. 5 Cross-section of the cu-
ticle of P. gibbulus prosoma (a)
and femur (b), E. arcuata pro-
soma (c) and femur (d), P.
agrestis prosoma (e) and femur
(f). Each cuticle is composed of
three layers: the epicuticle (Ep),
exocuticle (Ex) and endocuticle
(En). Scales 5μm
Naturwissenschaften

become an easy victim as it could not run away. Anecdotal
reports support this hypothesis: Palpimanus gibbulus was
observed to catch a salticid in a retreat and P. orientalis
captured an agelenid on the web (Murphy 1991); the
palpimanid Palpimanus stridulator Lawrence was observed
capturing sand-burrowing eresid and heteropodid species in
the Namib desert (Henschel 1997; Henschel and Lubin
1997); Cerveira and Jackson (2005) observed that Palpima-
nus invades silken retreats of salticid spiders.
Retreat invasion must have led to the evolution of a variety
of hunting adaptations to increase the searching, capture and
handling of prey. It is not clear how Palpimanus species
locate their prey. Prey-specialised predators often utilise
intraspecific signals produced by their exclusive prey. For
example, myrmecophagous species track the alarm phero-
mone of ants (Allan et al. 1996). Other araneophagous
species use chemical cues (e.g. Jackson et al. 1997;Clark
and Jackson 2000). Palpimanus captures various stages and
sexes of spiders; thus, the use of prey sex pheromones is
unclear. Other types of pheromones produced by spiders
have not been identified yet (Schulz 2004). Palpimanus may
be able to detect and recognise spider silk used in webs,
retreats and burrow lining. These can provide a reliable
chemical or mechanical signal for a nocturnal hunter. Once
the retreat is located, the predator must approach a prey in a
stealthy way to avoid counter-attack. This is achieved by a
combination of behavioural adaptations: very slow approach
followed by a flash-like attack. Araneophagous salticids of
the genera Phaeacius and Holcolaethis adopt special stalking
and ambushing routines that are highly effective at capturing
hersiliid and salticid prey (Li 2000;Lietal.2003). Some
other araneophagous spiders rely on aggressive mimicry.
Portia, for example, imitates prey or the mate of the resident
spider (Tarsitano et al. 2000). Cerveira and Jackson (2005)
suggest that Palpimanus use a different form of cryptic
stalking. Palpimanus moves so gently and slowly that the
prey spider might not be able to recognise it as a predator.
They hypothesise that the slow movement of Palpimanus
might be below the threshold of the prey’s secondary eyes
that are used as motion detectors (Land 1971). This is a
reasonable hypothesis if the capture is taking place during
day (at sufficient light intensity). As the hunting seems to
take place at night (Pekár, pers. observ.) or under stones
where the light intensity must be lower, we rather suspect
that a slow approach produces very weak vibrations that are
below the detection threshold of prey lyriform organs.
Finally, the predator should perform a successful attack.
For this purpose, the following morphological adaptations
were found in Palpimanus: scopulae, massive forelegs and
a thick cuticle all over the body. The large area of scopulae
on the forelegs helps Palpimanus to grab hold of a prey
body part during a very fast strike by the forelegs.
Palpimanus then holds the prey firmly against its body
with its stout forelegs until its venom takes affect, so the
prey is unlikely to be able to inflict injury on the
Palpimanus. Such a capture tactic is very different from
the tactic used by spiders catching dangerous insects.
Specifically, myrmecophagous spiders hold ants in chelic-
era in a position that keeps dangerous mandibles away from
the spider body thus preventing a counter-bite (Li and
Jackson 1996). Palpimanus can hold the prey chelicera next
to its body because the extremely thick cuticle acts as
defensive armour.
There is mainly anecdotal evidence for morphological
adaptations related to prey capture in spiders (Eberhard
1980;Stowe1986). Araneophagous archaeid spiders
possess strangely elongated chelicerae, which keep prey
spiders away from their body (Legendre 1961) thus
preventing counter-attack. Other araneophagous spiders
have modified setosity of the forelegs. Scopula-like hairs
were found at the tips of tarsi of Portia and are assumed to
be used in prey capture (Foelix et al. 1984). The scopulae
of Palpimanus appear to be similar to those of Portia,
though long and strong bristles are also found on the
forelegs of araneophagous Ero and Mimetus (both Mim-
etidae). These might also be used to hold the prey-spider
firmly, thus protecting the predator from counter-attack
(Jackson and Whitehouse 1986). Direct evidence for the
role of morphological adaptations is rare. Oniscophagous
Dysdera (Dysderidae) spiders possess three distinct shapes
of chelicera used in the capture of woodlice (Řezáčet al.
2008). The type of chelicera and attack tactic used
corresponds to the type of defence used by woodlice.
Adaptations found in Palpimanus are not universal for
araneophagy. This was revealed in trials staged with
another araneophagous spider, C. algerina. This salticid
spider does not possess any apparent morphological
adaptations (it has a thin cuticle in comparison with
Palpimanus) used for the capture of spiders and its
predatory behaviour is similar to other Spartaeinae salticids
(Jackson and Hallas 1986). C. algerina hunts various
spiders, but predominantly Oecobius beneath stones
(Guseinov et al. 2004). It is mainly active at dawn and
dusk, though it is able to catch prey even in darkness. Thus,
the trophic niches of Palpimanus and C. algerina overlap
as they occur at similar sites and prey on similar spider
prey. Clearly, Palpimanus is better adapted for preying on
C. algerina than the reverse.
Morphological and behavioural adaptations generally
increase capture efficiency and allow the capture of larger
prey. Scarce evidence suggests that specialised spider
predators can overcome prey that is markedly larger. Many
euryphagous cursorial spiders catch prey that is similar or
smaller in body size (Nentwig and Wissel 1986).
Specialised myrmecophagous spiders that prey on danger-
ous ants, however, often overcome prey several times larger
Naturwissenschaften

(Pekár 2004). We found here that Palpimanus was able to
overcome prey 200% larger. The capture of such large prey
may minimise the foraging rate and exposure to predators.
Interestingly, the prey specificity of araneophagous spider
species varies and is not restricted only to prey spiders. For
example, 56% of prey of C. algerina were spiders, whereas
95% of prey of P. fimbriata (both Salticidae) were spiders,
the remaining being insects (Jackson and Blest 1982;
Guseinov et al. 2004). Similarly, 53% prey of Mimetus
notius (Mimetidae) were spiders (Kloock 2001). Thus, all
araneophagous spiders investigated so far also feed on prey
other than spiders. This could be because the behavioural
capture trade-offs are not as severe as in strict prey specialists
such as Zodarion or Mastophora that possess effective prey-
capture tactics for their exclusive prey (ants and moths,
respectively) and were unable to capture other prey (Yeargan
1988; Pekár 2004). Nor is there evidence for a physiological
trade-off in araneophagous species. The consumption of
insects only reduced some fitness parameters in comparison
with the superior pure spider diet (Li and Jackson 1997).
Although araneophagous spiders do catch insects, they
have a preference for spiders. We did not perform
preference experiments here, but the fact that insects were
captured at a significantly lower frequency than spiders
suggests a preference of Palpimanus for spider prey. Other
araneophagous species, namely those of the genera Brettus,
Cocalus,Cyrba,Gelotia,Holcolaetis,Phaeacius and
Portia (all Salticidae) showed a pronounced preference for
attacking spiders (Jackson 1990b,c,2000; Li et al. 1997;
Jackson and Li 1998;Li2000).
Behavioural and morphological traits found in Palpimanus
spiders are remarkable adaptations that have evolved to
increase efficiency of prey capture and protection against
dangerous prey. Such adaptations are rare as there are only a
few predators specialising on prey of the same taxon and of
similar size.
Acknowledgements We would like to thank P. Cardoso, R. Jackson,
S. Korenko, J. Král, M. Řezáč, and F. Šťáhlavský for providing
Palpimanus spiders and three anonymous reviewers for giving useful
comments to the manuscript. The study was supported by the E.U.
Specific Support Action programme provided by the Jacob Blaustein
Center for Scientific Cooperation given to SP and by the grant no.
MSM0021622416 provided by the Ministry of Education, Youth and
Sports of the Czech Republic. JŠthanks go to project No. Z4 055
0506 realized at the Institute of Organic Chemistry and Biochemistry,
Academy of Sciences of the Czech Republic, Prague. This is
publication no. 736 of the Mitrami Department of Desert Ecology.
References
Allan RA, Elgar MA, Capon RJ (1996) Exploitation of an ant
chemical alarm signal by the zodariid spider Habronestes
bradleyi Walckenaer. Proc Roy Soc Lond Biol 263:69–73
Cerveira AM, Jackson RR (2005) Specialised predation by Palpima-
nus sp. (Araneae: Palpimanidae) on jumping spiders (Araneae:
Salticidae). J East Afr Nat Hist 94(2):303–317
Clark RJ, Jackson RR (2000) Web use during predatory encounters
between Portia fimbriata, an araneophagic jumping spider, and
its preferred prey, other jumping spiders. N Z J Ecol 27:129–136
Cutler B (1972) Notes on the biology of Mimetus puritanus
Chamberlin (Araneae: Mimetidae). Am Midl Nat 87(2):554–555
Dalingwater JE (1987) Chelicerate cuticle structure. In: Nentwig W
(ed) Ecophysiology of spiders. Springer, Berlin, pp 3–15
Darwin C (1859) On the origin of species by means of natural
selection, or the preservation of favoured races in the struggle for
life. John Murray, London
Dejean A (1997) Distribution of colonies and prey specialization in
the ponerine ant genus Leptogenys (Hymenoptera: Formicidae).
Sociobiology 29:293–299
Eberhard WG (1980) The natural history and behavior of the bolas
spider Mastophora dizzydeani sp. n. (Araneidae). Psyche 87(3–
4):143–169
Foelix RF, Jackson RR, Henksmeyer A, Hallas S (1984) Tarsal hairs
specialized for prey capture in the salticid Portia. In: Krafft B,
Leborgne R, Nancy RC (eds) Comptes Rendens du VIIeme
Colloque d’Arachnologie. Rev Arachnol 5(4):329–334
Guseinov EF (2006) The prey of a lithophilous crab spider Xysticus
loeffleri (Araneae, Thomisidae). J Arachnol 34:37–45
Guseinov EF, Cerveira AM, Jackson RR (2004) The predatory
strategy, natural diet, and life cycle of Cyrba algerina,an
araneophagic jumping spider (Salticidae: Spartaeinae) from
Azerbaijan. N Z J Zool 31:291–303
Hardin JW, Hilbe JM (2003) Generalized estimating equations.
Chapman & Hall/CRC, Boca Raton
Henschel JR (1997) Psammophily in Namib desert spiders. J Arid
Environ 37:695–707
Henschel JR, Lubin YD (1997) A test of habitat selection at two
spatial scales in a sit-and-wait predator: a web spider in the
Namib dunes. J Anim Ecol 66:401–413
Heuts BA, Brunt T (2001) Transitive predatory relationships of spider
species (Arachnida, Araneae) in laboratory tests. Behav Process
53:57–64
Hodek I, Honěk A (1996) Ecology of Coccinellidae.Kluwer,
Dordecht
Jackson RR (1990a) Predatory versatility and intraspecific interactions
in Cyrba algerina and Cyrba ocellata, web-invading spartaeine
jumping spiders (Araneae: Salticidae). N Z J Zool 17:157–168
Jackson RR (1990b) Predatory and silk utilisation behaviour of
Gelotia sp. indet. (Araneae: Salticidae: Spartaeinae), a web-
invading aggressive mimic from Sri Lanka. N Z J Zool 17:475–
482
Jackson RR (1990c) Ambush predatory behaviour of Phaeacius
malayensis and Phaeacius sp. indet., spartaeine jumping spiders
(Araneae: Salticidae) from tropical Asia. N Z J Zool 17:491–498
Jackson RR (1992) Eight-legged tricksters: spiders that specialize at
catching other spiders. Bioscience 42:590–598
Jackson RR (2000) Prey preferences and visual discrimination ability
of Brettus,Cocalus and Cyrba, araneophagic jumping spiders
(Araneae: Salticidae) from Australia, Kenya and Sri Lanka. N Z J
Ecol 27:29–39
Jackson RR, Blest AD (1982) The biology of Portia fimbriata, a web-
building jumping spider (Araneae, Salticidae) from Queensland:
utilization of webs and predatory versatility. J Zool 196:255–293
Jackson RR, Hallas SEA (1986) Predatory versatility and intraspecific
interactions of spartaeine jumping spiders (Araneae: Salticidae):
Brettus adonis, B. cingulatus, Cyrba algerina and Phaeacius sp.
indet. N Z J Zool 13:491–520
Jackson RR, Li D (1998) Prey preferences and visual discrimination
ability of Cyrba algerina, an araneophagic jumping spider
Naturwissenschaften

(Araneae: Salticidae) with primitive retinae. Isr J Zool 44:227–
242
Jackson RR, Li D, Robertson MB (1997) Cues by which suspended-
leaf nests of Euryattus (Araneae: Salticidae) females are
recognized by conspecific males and by aggressive-mimic salt-
icid, Portia fimbriata. J Zool Lond 243:29–46
Jackson RR, Whitehouse MEA (1986) The biology of New Zealand
and Queensland pirate spiders (Araneae, Mimetidae): aggressive
mimicry, araneophagy and prey specialization. J Zool 210:279–
303
Jackson RR, Wilcox RS (1990) Aggressive mimicry, prey-specific
predatory behaviour and predator-recognition in the predator-
prey interactions of Portia fimbriata and Euryattus sp., jumping
spiders from Queensland. Behav Ecol Sociobiol 26:111–119
Jarman EAR, Jackson RR (1986) The biology of Taieria erebus
(Araneae, Gnaphosidae), an araneophagic spider from New
Zealand: silk utilisation and predatory versatility. N Z J Zool
13:521–541
Jocqué R, Dippenaar-Schoeman AS (2006) Spider families of the
world. Royal Museum for Central Africa, Tervuren
Kloock CT (2001) Diet and insectivory in the “araneophagic”spider,
Mimetus notius (Araneae: Mimetidae). Am Midl Nat 146:424–428
Land MF (1971) Orientation by jumping spiders in the absence of
visual feedback. J Exp Biol 54:119–139
Legendre R (1961) Études sur les Archaea (Aranéides). II. La capture
des proies et la prise de nourriture. Bull Soc Zool Fr 86:316–319
Li D (2000) Prey preference of Phaeacius malayensis, a sprtaeine
jumping spider (Araneae: Salticidae) from Singapore. Can J Zool
78:2218–2226
Li D, Jackson RR (1996) Prey-specific capture behaviour and prey
preferences of myrmicophagic and araneophagic jumping spiders
(Araneae: Salticidae). In: Mahnert V (ed) Proceedings of the
XIIIth International Congress of Arachnology. Rev Suisse Zool
hors serie:423–436.
Li D, Jackson RR (1997) Influence of diet on survivorship and growth
in Portia fimbriata, an araneophagic jumping spider (Araneae:
Salticidae). Can J Zool 75:1652–1658
Li D, Jackson RR, Barrion A (1997) Prey preferences of Portia
labiata,P. africana, and P. schultzi, araneophagic jumping
spiders (Araneae: Salticidae) from the Philippines, Sri Lanka,
Kenya, and Uganda. N Z J Zool 24:333–349
Li D, Jackson RR, Barrion A (1999) Parental and predatory behaviour
of Scytodes sp., an araneophagic spitting spider (Araneae:
Scytodidae) from the Philippines. J Zool 247:293–310
Li D, Jackson RR, Lim MLM (2003) Influence of background and
prey orientation on an ambushing predator’s decisions. Behav-
iour 140:739–764
Lloyd JE (1975) Aggressive mimicry in Photuris fireflies: signal
repertoires by femmes fatales. Science 187:452–453
Mead R (1988) The design of experiments. Cambridge University
Press, Cambridge
Michalková V, Pekár S (2009) How glyphosate altered the behaviour of
agrobiont spiders (Araneae: Lycosidae) and beetles (Coleoptera:
Carabidae). Biol Control 51:444–449
Montanucci RR (1989) The relationship of morphology to diet in the
horned lizard genus Phrynosoma. Herpetologica 45:208–216
Murphy FM (1991) The 1989 presidential address –part 2. Some
interesting european spiders. Br J Entomol Nat Hist 4:69–76
Nentwig W (1987) The prey of spiders. In: Nentwig W (ed)
Ecophysiology of spiders. Springer, Berlin, pp 249–263
Nentwig W, Wissel C (1986) A comparison of prey lengths among
spiders. Oecologia 68:595–600
Pekár S (2004) Predatory behavior of two European ant-eating spiders
(Araneae, Zodariidae). J Arachnol 32(1):31–41
Pekár S, Brabec M (2009) Modern analysis of biological data. 1.
Generalised linear models in R. Scientia, Prague [in Czech]
Pekár S, Král J (2002) Mimicry complex in two central European
zodariid spiders (Araneae: Zodariidae): how Zodarion deceives
ants. Biol J Linn Soc 75(4):517–532
Pekár S, Lubin YD (2003) Habitats and interspecific associations of
zodariid spiders in the Negev (Araneae: Zodariidae). Isr J Zool
49(4):255–267
Platnick NI (1981) A review of the spider subfamily Palpimaninae
(Araneae, Palpimanidae), I. Bull Br Arachnol Soc 5(4):169–173
Platnick NI (2011) The world spider catalog, version 11.5. American
Museum of Natural History. http://research.amnh.org/iz/spiders/
catalog. doi: 10.5531/db.iz.0001.
R Development Core Team (2010) R: A language and environment for
statistical computing. R Foundation for Statistical Computing,
Vienna, http://www.R-project.org
Reynolds ES (1963) The use of lead citrate at high pH as an electron
opaque stain in electron microscopy. J Cell Biol 17:208–212
ŘezáčM, Pekár S, Lubin Y (2008) How oniscophagous spiders
overcome woodlouse armour. J Zool 275:64–71
Richardson PRK (1987) Food consumption and seasonal variation in
the diet of the aardwolf Proteles cristatus in southern Africa. Z
Saugetierkunde 52:307–325
Schulz S (2004) Semiochemitry of spiders. In: Cardé RT, Millar JG
(eds) Advances in insect chemical ecology. Cambridge Univer-
sity Press, Cambridge, pp 110–150
Schulz M, Wainer J (1997) Diet of the golden-tipped bat Kerivoula
papuensis (Microchiroptera) from north-eastern New South
Wales, Australia. J Zool 243(4):653–658
Singer MS (2008) Evolutionary ecology of polyphagy. In: Tilmon KJ
(ed) Specialization, speciation, and radiation: the evolutionary
biology of herbivorous insects. University of California, Berkeley,
pp 29–42
Spurr AR (1969) A low-viscosity epoxy resin embedding medium for
electron microscopy. J Ultrastruct Res 26:31–43
Stowe MK (1986) Prey specialization in the Araneidae. In: Shear WA
(ed) Spiders. Webs, behavior, and evolution. Stanford University
Press, Stanford, pp 101–131
Tarsitano M, Jackson RR, Kirchner WH (2000) Signals and signal
choices made by the araneophagic jumping spider Portia
fimbriata while hunting the orb-weaving web spiders Zygiella
x-notata and Zosis geniculatus. Ethology 106:595–615
Ter Braak CJF, Šmilauer P (2002) CANOCO Reference Manual and
CanoDraw User‘s Guide. Software for Canonical Community
Ordination (version 4.5). Biometris, Wageningen and České
Budějovice.
Uetz GW, Halaj J, Cady AB (1999) Guild structure of spiders in major
crops. J Arachnol 27:270–280
Underwood G (1997) An overview of venomous snake evolution. In:
Thorpe RS, Wüster W, Malhotra A (eds) Venomous snakes:
ecology, evolution and snakebite. Symp Zool Soc Lond 70:1–13
Whitehouse MEA, Lubin Y (1998) Relative seasonal abundance of
five spider species in the Negev desert: intraguild interactions
and their implications. Isr J Zool 44:187–200
Wilcox RS, Jackson RR, Gentile K (1996) Spiderweb smokescreens:
spider uses background noise to mask stalking movements. Anim
Behav 51:313–326
Yeargan KV (1988) Ecology of a bolas spider, Mastophora hutchinsoni:
phenology, hunting tactics, and evidence for aggressive chemical
mimicry. Oecologia 74:524–530
Naturwissenschaften