(ISSN 0161-8202)

Journal of

ARACHNOLOGY

PUBLISHED BY THE AMERICAN ARACHNOLOGICAL SOCIETY

VOLUME 37

2009

NUMBER 3

THE JOURNAL OF ARACHNOLOGY

EDITOR-IN-CHIEF: James E. Carrel, University of Missouri-Columbia

MANAGING EDITOR: Douglass H. Morse, Brown University

SUBJECT EDITORS: Ecology — Stano Pekar, Masaryk University; Systematics — Mark Harvey, Western Aus-
tralian Museum and Ingi Agnarsson, University of Puerto Rico; Behavior — Linden Higgins, University of Vermont;
Morphology and Physiology — Jeffrey Shuitz, University of Maryland

EDITORIAL BOARD: Alan Cady, Miami University (Ohio); Jonathan Coddington, Smithsonian Institution;
William Eberhard, Universidad de Costa Rica; Rosemary Gillespie, University of California, Berkeley; Charles
Griswold, California Academy of Sciences; Marshal Hedin, San Diego State University; Herbert Levi, Harvard
University; Brent Opell, Virginia Polytechnic Institute & State University; Norman Platnick, American Museum
of Natural History; Ann Rypstra, Miami University (Ohio); Paul Selden, University of Kansas; Matthias Schae-
fer, Universitet Goettingen (Germany); William Shear, Hampden-Sydney College; Petra Sierwald, Field Mu-
seum; I-Min Tso, Tunghai University (Taiwan).

The Journal of Arachnology (ISSN 0161-8202), a publication devoted to the study of Arachnida, is published
three times each year by The American Arachnological Society. Memberships (yearly): Membership is open to
all those interested in Arachnida. Subscriptions to The Journal of Arachnology and American Arachnology (the
newsletter), and annual meeting notices, are included with membership in the Society. Regular, $55; Students, $30;
Institutional, $125 . Inquiries should be directed to the Membership Secretary (see below). Back Issues: James
Carrel, 209 Tucker Hall, Missouri University, Columbia, Missouri 65211-7400 USA. Telephone; (573) 882-3037.
Undelivered Issues: Allen Press, Inc., 810 E. 10th Street, P.O. Box 368, Lawrence, Kansas 66044 USA.

THE AMERICAN ARACHNOLOGICAL SOCIETY

PRESIDENT: Rosemary Gillespie (2009-2011), Environmental Science, Policy & Management, Division of
Organisms and Envirironment, University of California, Berkely, CA 94720-3114 USA.

PRESIDENT-ELECT: Jonathan Coddington (2009-2011), Smithsonian Institution, Washington, DC 20013-7012
USA.

MEMBERSHIP SECRETARY: Jeffrey W. Shultz (appointed). Department of Entomology, University of Maryland,
College Park, MD 20742 USA.

TREASURER: Karen Cangialosi, Department of Biology, Keene State College, Keene, NH 03435-2001 USA.
SECRETARY: Alan Cady, Dept, of Zoology, Miami University, Middletown, Ohio 45042 USA.

ARCHIVIST: Lenny Vincent, Fullerton College, Fullerton, California 92634 USA.

DIRECTORS: Matt Persons (2008-2010), Paula Cushing (2009-201 1), Todd Blackledge (2009-2011)

PAST DIRECTOR AND PARLIAMENTARIAN: H. Don Cameron (appointed), Ann Arbor, Michigan 48105 USA.
HONORARY MEMBERS: C.D. Dondale, H.W. Levi, A.F. Millidge.

Cover photo: Mating in Semeiochernes armiger (Psudoscorpiones) collected in the Republic of Panama. Photo by Jeanne A. Zeh.

Publication date: 17 November 2009

©This paper meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper).

NOV :iO ZOU9

s^/BRAR\t§^

2009. The Journal of Arachnology 37:249-253

Frame-web-choice experiments with stingless bees support the prey-attraction hypothesis for silk

decorations in Argiope savignyi

Dumas Galvez: Animal Ecology Group, University of Groningen, Biological Center, PO Box 9750 A A Haren,
the Netherlands. E-mail: galvezd@gmail.com

Abstract. There is controversy about the function of silk stabilimenta, also called silk decorations, on spiders’ webs. Most
of the proposed hypotheses have been tested using indirect methods. Protection against predators, advertisement for
vertebrates to avoid web damage, and increasing prey attraction are the most popular hypotheses. In this study, I tested the
prey attraction hypothesis on the silk decorations of the araneid Argiope savignyi using a trial tunnel built in the field, in
which I exposed stingless bees Tetragonisca angiistula to decorated and undecorated webs placed on wooden frames. I
carried out two experiments: 1) a three-frame choice, consisting of a frame bearing a decorated web, one bearing an
undecorated web and a control frame without web and spider; 2) a two-frame choice, in which the bees were exposed to
only two frames consisting of “decorated web vs. control,” “decorated web vs. undecorated web,” and “undecorated web
vs. control”. In favor of the prey attraction function, I found that decorated webs intercepted more bees than webs deprived
of the decoration or controls with no webs. Argiope savignyi' s decorations might lure prey to the web by UV-reflectance as
it has been suggested for other Argiope species.

Keywords: Decorated, foraging, stabilimenta, undecorated

A wide range of orb-weaving spiders builds silk decorations
or stabilimenta on their webs (Araneae: Araneidae, Tetra-
gnathidae, Uloboridae; Scharff & Coddington 1997). Five
functions have been suggested for these structures: 1)
protection against predators, 2) advertisement to vertebrates
so as to avoid web damage, 3) prey attraction, 4) stabilization
of the web, and 5) a source of shade. Most work has focused
on the first three hypotheses (Herberstein et al. 2000; Bruce
2006). However, after more than 100 years of research, no
consensus about the functionality of decorations has yet been
reached, and a variety of methods have been applied
producing contradictory outcomes. In support of functions
1 and 2, decorations on the web of Argiope aurantia Lucas
1833 reduced predatory attacks by mud-dauber wasps and
web damage by birds; simultaneously, web visibility to prey
was increased and prey capture rates declined. Hence, a cost
associated with decoration construction was suggested
(Blackledge & Wenzel 1999). A different study with A.
aurantia, on the other hand supported Function 3, that
decorated webs attracted more prey although they were
compared with undecorated webs of A. trifasciata (Forsskal
1775) (Tso 1998a). For A. appensa (Walckenaer 1842), no
differences in foraging success were found between decorated
and undecorated webs. In support of Function 3, Bruce et al.
(2001) and Seah & Li (2001) found that decorated webs of A.
keyserlingi Karsch 1878 and A. versicolor (Doleschall 1859)
attracted more prey; however, decorations also attracted
predators, in opposition to Function 1. Researchers have
concluded that there is a trade-off in foraging strategies, since
decorated webs are often smaller than undecorated webs
(Hauber 1998).

The traditional perception that the spider web is an
undetectable trap has changed drastically since the idea that
web decorations might attract prey by UV reflectance was
suggested (Craig & Bernard 1990). The prey-attraction
hypothesis (Function 3) states that the presence of decorations

increases the foraging success of the spiders. Such an outcome
has been proposed for various species of the genus Argiope: A.
aetherea Thorell 1881 (Elgar et al. 1996), A. trifasciata (Tso
1996), A. aurantia (Blackledge & Wenzel 1999), A. versicolor
(Li et al. 2004; Li 2005), A. argentata (Fabricius 1775) (Craig
& Bernard 1990; Craig et al. 2001), A. keyserlingi Karsch 1878
(Herberstein 2000; Bruce et al. 2001), A. aemula (Walckenaer
1842) (Cheng & Tso 2007); as well as for Octonoha sybotkles
(Uloboridae) (Bdsenberg & Strand 1906) (Watanabe 1999),
Araneus eburnus (Keyserling 1886) (Bruce et al. 2004), and
some other species (Herberstein et al. 2000; Bruce 2006).

An important aspect to be considered when testing the prey-
attraction hypothesis is the interference of web-size: decorated
webs, usually smaller than undecorated ones, might attract
more prey due to their decoration. Undecorated webs,
however, are usually bigger and hence prey-capturing success
might be increased due to the larger area. Therefore, the
suggested trade-off in foraging strategies and energetic costs
remains speculative. For that reason, an appropriate tech-
nique to eliminate the influence of web size, when decorated
and undecorated webs are compared, has been manual
removal of the decorations (Bruce et al. 2001, 2004).

I tested the prey-attraction hypothesis for the poorly studied
Neotropical spider Argiope savignyi Levi 1968 using a new
method that consisted of a trial tunnel combined with
decoration removal and prey manipulation. The tunnel is
placed in the field, which can mimic natural visual conditions
in which spiders and preys are found. Many studies have
tested the hypothesis in laboratory conditions (e.g., Y-choice
experiments), which might not reproduce natural conditions.
As well, the influence of web size can be eliminated while the
prey capture history of the spiders, which has an essential
effect on the decoration behavior (Craig et al. 2001), can be
controlled. If the web decoration functions to attract prey,
then I expected that decorated webs would intercept more bees
than the undecorated webs and empty control frames.

249

250

THE JOURNAL OF ARACHNOLOGY

METHODS

Site. — This study was carried out from 18 July to 4 August
2007 at La Selva Biological Station, Heredia, Costa Rica
(10°26'N, 83°59'W), a 1550-ha reserve in the Atlantic
lowlands with an annual average rainfall of 4000 mm. See
Sanford et al. (1994) for more details about the station.

Animals. — Argiope savignyi is an aerial web weaving spider
that decorates its web with zigzags of silk laid in a variety of
designs that include silk discs (juveniles) or one to four arms of
a cross (adults). Some webs lack decorations (Nentwig & Rogg
1988). This species is common at La Selva (Rovner 1989;
Timm & Losilla 2007). 1 confirmed the species identity using
the taxonomic key for Argiope by Levi (2004). No voucher
specimens were collected but some collected from La Selva are
available at the National Museum of Natural History,
Smithsonian Institution, Washington, D.C. (Levi 2004).

Experimental Design. — Collected individuals of A. savignyi
were placed in a large screened cage (7X3X2 m). This cage
contained herbaceous vegetation with insects such as Homop-
tera, Hymenoptera, Orthoptera, on which the spiders were
allowed to forage. In addition, each spider was fed several
stingless bees to guarantee that they were satiated, an
important factor for inducing the construction of decorations
(Craig et al. 2001).

A 300 X 120 X 80 cm tunnel, open at both exits, was
constructed (Fig. 1). The different web treatments were set up
on wooden frames at one end, and a wooden box (40 X 30 X
20 cm) with a nest of the stingless bee Tetragonisca angustula
Latreille 1811 was placed in the other end. The frames were
put on a 2 X 35 X 120 cm wooden board placed at the exit of
the tunnel so that the frames were not in contact with the
ground. The exit of the bee nest faced that of the tunnel for the
web treatments. Bees could leave the tunnel through the exit
containing the frames, which they usually did, or by the other
exit. The nest was placed in the tunnel with both exits opened
for 48 h before the beginning of the experiment in order to get
the bees used to the tunnel and the new nest location. The
reason for placing the nest entrance near the tunnel exit was to
reduce the stress on the bees, which probably occurs when they
are individually manipulated, for instance with CO2 anesthesia
(Bruce 2006). With the intention of comparing the two web
treatments, I used spiders of similar sizes, and the control
frame never contained a spider. The exit of the tunnel where
the frames were placed was in front of herbaceous vegetation,
and a dark green mesh placed one m from it.

1 performed two experiments with A. savignyi: 1 ) A “Three-
frame choice,” consisting of three frames (34.5 X 45.0 cm, or 20
X 20 cm for smaller webs) placed next to each other at the same
time and at the same end of the tunnel with different web
treatments; one bearing a decorated web, one bearing an
undecorated web, and a control without web and spider (Fig. 1)
and 2) a two-frame-choice experiment in which the bees were
exposed to only two frames placed at the same end of the tunnel
and consisting of the following: “decorated web vs. control,”
“decorated web vs. undecorated web,” and “undecorated web
vs. control.” Small frames (20 X 20 cm) did not cover the entire
area of the tunnel’s exit, so I covered the remaining space with
cardboard sheets. For the three-frame-choice experiment, I used
two spiders per replicate {n = 8, 1 55 bees): one for the decorated
web and one for the undecorated web. For the two-frame-choice

Figure 1. — Trial tunnel in which the stingless bees were exposed to
the different web treatments. The walls and roof of the tunnel are
removed in order to reveal the inside. Solid arrows show the two
possible trajectories of bees to fly out of the tunnel from the nest (N).
The exit bearing the web treatments is represented by A and the
opposite exit by B.

experiments, I compared “decorated webs versus control”
frames for 17 repetitions (175 bees), “undecorated webs versus
control” frames for 9 repetitions (86 bees), and “decorated webs
versus undecorated webs” for 10 repetitions (100 bees). The
three-frame-choice experiment trials lasted approximately 5 to
1 5 min. Bees were allowed to return to the nest except for those
that were collected in order to feed the spiders (or caught and
consumed by the spider itselQ- Only one trial was carried out per
day, which greatly reduces the possibility of avoidance learning
by stingless bees (Craig 1994b). Craig (1994b) also proposed that
even if bees learn to avoid decorated webs (e.g., in one location),
they are unable to generalize a similar response to other
decorated webs. The two- and three-frame-choice experiments
were carried out in random order. The three sets within the two-
frame-choice experiments were randomly assigned as well.

All decorations were either cross or linear patterns.
Decorations were removed by burning the fine silk lines with
heated fine-point forceps while the spider was on the web
except on a few occasions when the spider was removed first.
The spider was then placed back on the web after the
decoration was removed. In some cases, a little damage was
done to the web during burning, and in these instances, I used
the forceps to produce similar damage to the decorated web.

I counted the numbers of bees either being intercepted
(including bees caught by spiders) or flying through each
frame, and determined the number of bees intercepted per
frame. I switched the positions of the frames each time two
bees had exited the tunnel or were intercepted in order to
avoid any possible bias due to frame position. The frames were
placed at the exit of the tunnel only when no bee was either
leaving the nest or flying in the tunnel. In cases in which three
or more bees accumulated in the web because the spider did
not attack them, I removed the three frames and used forceps
to remove the bees in order to avoid the possibility that bees
caught there would deter more bees from flying into the web. I
did not remove the bees if they were captured by the spider or

GALVEZ— PREY ATTRACTION FUNCTION OF SILK DECORATIONS

251

4 -

•o

S 3.5 - T

g

£ 3 - »-

C C

is 2.5- ^

8 § "

« O j,

■s I 2

I I 1.5- i

(/) X.

<u

M 1 -

o

» 0.5 -

0 - — __

0 Decorated Undecorated Control

Web type

Figure 2. — Number of Tetragonisca angustula bees intercepted for
the three-frame-choice experiment set for Argiope argentata. Mean
± SD.

wrapped with silk by the spider. After this, I put the frames
back at the exit to continue the experiment. I used 10-30 bees
per repetition, which required a new set of webs made by
spiders not previously used. I did not register numbers of bees
wrapped or consumed. Spiders wrapped bees few times, but
they usually kept consuming the first bee that was caught. This
apparently did not discourage bees from flying into the web.
All the trials were carried out at 09:30-12:00 and 13:00-15:00
h, when the light conditions were relatively constant.

The counts from the three- and the two-frame-choice
experiments were square root transformed. The transformed
data from the three-frame choices was tested for normality
and analyzed using a single factor ANOVA. Finally, the
transformed data from the two-frame-choice experiments were
analyzed with a Atest for paired samples. Effects were
accepted as statistically significant if P < 0.05, and all
analyses were done using STATISTICA 7.0 (StatSoft 2001).

RESULTS

Bees were intercepted by decorated webs significantly more
often than by the undecorated webs or the control frames (48,
30 and 22% of the bees respectively, F2, 21 = 4.65, P = 0.02,
Fig. 2). I did not find significant differences between
undecorated webs and the control frames (Tukey HSD test,
P = 0.362). In the “decorated webs versus control”
experiment, 64% of the bees chose the frames bearing
decorated webs {t = 2.84, df = 16, P = 0.006, Table 1).
Decorated webs also intercepted more bees (58%) than
undecorated webs (42%), t = \.9\, df = 9, P = 0.044).

DISCUSSION

The prey attraction function of silk decorations for A.
savignyi is supported by both the two- and the three-frame-
choice experiments (Fig. 2, Table 1). Decorated webs inter-
cepted significantly more bees than those webs from which the
decoration was removed. Webs deprived of decorations showed
no significant differences from the control frame that lacked
either web or spider. The results from these experiments are also
reinforced by the fact that bees did not show an avoidance-
learning process which would have decreased the interception
of the decorated web (Craig 1994a). The literature covering this
hypothesis is controversial; many studies have revealed that
decorated webs intercept more prey than undecorated webs
(Craig & Bernard 1990; Elgar et al. 1996; Tso 1996, 1998a,
1998b; Watanabe 1999; Herberstein 2000; Bruce et al. 2001;
Craig et al. 2001; Li et al. 2004; Li 2005; Bruce & Herberstein
2005; Cheng & Tso 2007) but some studies found no evidence in
favor of the hypothesis (Blackledge 1998; Blackledge & Wenzel
1999; Hoese et al. 2006; Jaffe et al. 2006; Bush et al. 2008;
Eberhard 2008; Gawryszewski & Motta 2008). This hypothesis
has been previously supported for one of the closest relatives of
A. savignyi, A. argentata by Craig (1991) and Craig et al. (2001),
but no manipulative experiments (e.g., decoration removal)
were performed. Craig et al. (2001) correlated the increase in
decoration frequency with the increase in the number of
stingless bees. Craig (1991) also calculated an index of
predator-prey encounter rates based on the damage found on
the web. Such damage is not necessarily caused by prey,
however. She also assumed that the prey damage or destroy
part of the web, even when they are not captured. I saw several
cases in which a bee was intercepted in the web and later
escaped without damaging the web.

There is one result from the set “undecorated web versus
control” that clearly merits further study. It could be
anticipated that the undecorated webs (bearing a spider)
would intercept more bees than the control, considering the
UV reflective properties of the spider’s dorsal surface that is
thought also to play an important role for attracting prey
(Craig & Ebert 1994; Cheng & Tso 2007; Bush et al. 2008).
This was not observed but partially supported by the three-
choice-frame experiments; 30% and 22% bees intercepted
undecorated and control frames, respectively. Yet the bright
coloration of Argiope spiders may have no relation as a prey
attraction function, serving more as camouflage for the spiders
in A. bruennichi (Vaclav & Prokop 2006) and A. keyserlingi
(Hoese et al. 2006). The functional significance of body
coloration of Argiope spiders remains unresolved.

One of the advantages of the method in this study was that
stress on prey was reduced, since the experiments were
performed in the field. The prey attraction hypothesis can be

Table 1 . — Statistical summary and preferences for the two-frame-choice experiments set for Argiope argentata. dec: decorated webs; undec:
undecorated webs; and control.

Treatment

t

df

P

Total number
of bees

dec

% of bees intercepted

control

undec

dec vs control

2.84

16

0.006

175

64

36

undec vs control

0.22

8

0.42

86

—

49

51

dec vs undec

1.91

9

0.04

100

58

—

42

252

THE JOURNAL OF ARACHNOLOGY

directly tested in a number of ways, all of which have their
advantages and disadvantages. The field correlation technique
involves correlating the presence of web decorations with prey
capture rates, but this method has produced contradictory
results for different species (Hauber 1998; Tso 1998b;
Herberstein 2000; Bruce et al. 2001; Craig et al. 2001; Bruce
et al. 2004). A negative aspect of field correlations is that the
prey capture history of the spiders is unknown, and satiated
individuals can construct more decorations (Blackledge 1998;
Tso 1999; Seah & Li 2002). Consequently, decorated webs
may just be in sites where prey are abundant.

Another method is the Y-choice experiment, which has been
used in laboratory experiments to show that flies are attracted to
decorations (Craig & Bernard 1990; Watanabe 1999; Bruce et al.
2001; Li et al. 2004). These studies have been carried out in
laboratory conditions using artificial lights. Moreover, decorat-
ed and undecorated webs have been contrasted without the
presence of the spider on the web, which might reduce the
similarity to a natural prey-spider encounter. The third
technique is the experimental manipulation of webs by
decoration removal to compare decorated and undecorated
webs. This allows investigating the effects of these structures on
prey capture and predator response without the cause and effect
problem as in the field correlation method (see Bruce 2006).
Some studies in which decoration removal was used in the field
found opposite results for the prey-attraction function. Black-
ledge & Wenzel (1999) suggested that the decoration in Argiope
aiirantia reduced foraging success, but Tso (1998a) found that
the decoration in fact increased it. The possible reason of this
difference is that the former study did not control for web size
and the latter one used webs of similar size, as 1 did.

Even though the prey-attraction function is supported by
these experiments, other hypotheses, such as advertisement to
avoid web damage by vertebrates (Blackledge & Wenzel 1999)
and the anti-predator function (Bruce et al. 2001; Schoener &
Spiller 1992), are not necessarily discardable. Scharff &
Coddington (1997), in their phylogenetic analysis of the
family Araneidae, proposed that web decorations evolved
nine times independently in the 15 genera in which they are
known to occur. Although they reasoned that the widespread
convergent evolution of this trait only in diurnal species
suggests a search for a common cause, it might be possible to
find a wide range of function across the different groups of
spiders that evolved this trait. Is a multifunction role possible
for Argiope' s web decorations? For instance, Argiope trifas-
ciata's decorations increase foraging success (Tso 1996, 1998a)
and also provide protection against predators (Blackledge &
Wenzel 2001). For A. aiirantia, the prey attraction, predator
avoidance, and web advertisement functions have found
support (Tso 1998a; Blackledge & Wenzel 1999). However,
some studies that addressed a multifunction role only found
evidence for one function (e.g., Blackledge & Wenzel 1999;
Bruce et al. 2001). Bruce & Herberstein (2005) found
differences in the decorating behavior of three Australian
Argiope species that were apparently related to the pattern of
decoration that each species built. These dissimilarities suggest
that those decoration patterns perform different functions,
although with different costs and benefits associated. The
other two important visual functions suggested for decora-
tions, the anti-predator and the web advertisement hypotheses

can be tested using a similar approach, employing manual
decoration removal in a more natural visual condition similar
to this study. In this way, the different web types and their
spiders can be exposed to either predator or vertebrates (e.g.,
birds) in order to quantify their behavioral responses.

ACKNOWLEDGMENTS

I thank Carlos Garcia-Robledo, Orlando Vargas, Johel
Chaves, Arietta Fleming-Davies, Claudia Lizana, Mirjam
Knornschild, loana Chiver, Diomedes Quintero and Jose Vidal
for help during the experiments and data collection in the field. I
am thankful to Birgit Kdhler for improving the English of the
manuscript and William Eberhard for comments on it. I am
grateful with Alejandro Farji-Brener and Federico Chinchilla
for the comments on the proposal of this research. Thanks to my
friends from the course “Ecologia Tropical y Conservacion
2007,” who helped me in many ways to develop this research
idea. I also thank David W. Roubik for his help with the
identification of the bees. This research was funded by the
Organization for Tropical Studies (GRANT 4151).

LITERATURE CITED

Blackledge, T.A. 1998. Stabilimentum variation and foraging success
in Argiope aurantki and Argiope trifasciata (Araneae, Araneidae).
Journal of Zoology, London 246:21-27.

Blackledge, T.A. & J.W. Wenzel. 1999. Do stabilimenta in orb webs
attract prey or defend spiders? Behavioral Ecology 10:372-376.
Bruce, M.J. 2006. Silk decorations: controversy and consensus.

Journal of Zoology, London 269:89-97.

Bruce, M.J., M.E. Herberstein «fe M.A. Elgar. 2001. Signaling conflict
between predator and prey attraction. Journal of Evolutionary
Biology 14:786-794.

Bruce, M.J., A.M. Heiling & M.E. Herberstein. 2004. Web
decorations and foraging success in "Araneus' ebitrnus (Araneae:
Araneidae). Annales Zoologici Fennici 41:563-575.

Bruce, M.J. & M.E. Herberstein. 2005. Web decoration polymor-
phism in Argiope Audouin, 1826 (Araneidae) spiders: ontogenetic
and interspecific variation. Journal of Natural History 39:3833-
3845.

Bush, A. A., D.W. Yu & M.E. Herberstein. 2008. Function of bright
coloration in the wasp spider Argiope hnieimichi (Araneae:
Araneidae). Proceedings of the Royal Society B 275:1337-1342.
Cheng, R-C. & I.T. Tso. 2007. Signaling by decorating webs: luring
prey or deterring predators? Behavioral Ecology 18:1-7.

Craig, C.L. 1991. Physical constraints on group foraging and social
evolution: observations on web-spinning spiders. Functional
Ecology 5:649-654.

Craig, C.L. 1994a. Limits to learning: effects of predator pattern and
colour on perception and avoidance-learning by prey. Animal
Behaviour 47:1087-1099.

Craig, C.L. 1994b. Predator foraging behavior in response to
perception and learning by its prey: interactions between orb-
spinning spiders and stingless bees. Behavioral Ecology and
Sociobiology 35:45-52.

Craig, C.L. & G.D. Bernard. 1990. Insect attraction to ultraviolet-
reflecting spider webs and web decorations. Ecology 71:616-623.
Craig, C.L. & K. Ebert. 1994. Colour and pattern in predator-prey
interactions; the bright body colours and patterns of a tropical orb-
spinning spider attract tlower-seeking prey. Functional Ecology
8:616-620.

Craig, C.L., S.G. Wolf, J.L.D. Davis, M.E. Hauber & J.L. Maas.
2001. Signal polymorphism in the web-decorating spider Argiope
argentcita is correlated with reduced survivorship and the presence
of stingless bees, its primary prey. Evolution 55:986-993.

GALVEZ— PREY ATTRACTION FUNCTION OF SILK DECORATIONS

253

Eberhard, W.G. 2008. Araneus expletus (Araneae, Araneidae):
another stabilimentum that does not function to attract prey.
Journal of Arachnology 36:191-194.

Elgar, M..A., R.A. Allan & T.A. Evans. 1996. Foraging strategies in
orb-spinning spiders: Ambient light and silk decorations in Argiope
aelherea Walckenaer (Araneae:Araneoidea). Australian Journal of
Ecology 21:464-467.

Gawryszewski, F.M. & P.C. Motta. 2008. The silk tuft web
decorations of the orb-weaver Gasteracantha cancriformis: testing
the prey attraction and the web advertisement hypotheses.
Behaviour 145:277-295.

Hauber, M.E. 1998. Web decorations and alternative foraging tactics
of the spider Argiope appensa. Ethology, Ecology & Evolution
10:47-54.

Herberstein, M.E. 2000. Foraging behaviour in orb-web spiders
(Araneidae): do web decorations increase prey capture success in
Argiope keyserlingi Karsch, 1878? Australian Journal of Zoology
48:217-223.

Herberstein, M.E., C.L. Craig, J.A. Coddington & M.A. Elgar. 2000.
The functional significance of silk decorations of orb-web spiders:
a critical review of the empirical evidence. Biological Reviews
75:649-669.

Hoese, F.J., E.A.J. Law, D. Rao & M.E. Herberstein. 2006.
Distinctive yellow bands on a sit-and-wait predator: prey
attractant or camouflage? Behaviour 143:763-781.

Jaffe, R., W.G. Eberhard, C.D. Angelo, D. Eusse, A. Gutierrez, S.
Quijas, A. Rodriguez & M. Rodriguez. 2006. Caution, webs in the
way! Possible functions of silk stabilimenta in Gasteracantha
cancriformis (Araneae, Araneidae). Journal of Arachnology
34:448-455.

Levi, H.W. 2004. Comments and new records for the American
genera Gea and Argiope with the description of a new species
(Araneae: Araneidade). Bulletin Museum of Comparative Zoology
158:47-65.

Li, D. 2005. Spiders that decorate their webs at higher frequency
intercept more prey and grow faster. Proceedings of the Royal
Society B 272:1753-1757.

Li, D., M.L.M. Lim, W.K. Seah & S.L. Tay. 2004. Prey attraction as
a possible function of discoid stabilimenta of juvenile orb-spinning
spiders. Animal Behaviour 68:629-635.

Nentwig, W. & H. Rogg. 1988. The cross stabilimentum of Argiope
argentata (Araneae: Araneidae) - nonfunctional or a nonspecific
stress reaction? Zoologischer Anzeiger 221:248-266.

Rovner, J.S. 1989. Submersion survival in aerial web-weaving spiders
from a tropical wet forest. Journal of Arachnology 17:242-245.

Scharff, N. & J.A. Coddington. 1997. A phylogenetic analysis of the
orb-weaving spider family Araneidae (Arachnida, Araneae).
Zoological Journal of the Linnean Society 120:355-M24.

Sanford, R.L. Jr., P. Paaby, J.C. Luvall & E. Phillips. 1994. Climate,
geomorphology, and aquatic systems. Pp. 19-33. In La Selva.
Ecology and Natural History of a Neotropical Rainforest. (L.A.
McDade, K.S. Bawa, H.A. Hespenheide & G.S. Hartshorn, eds.).
University of Chicago Press, Chicago, USA.

Schoener, T.W. & D.A. Spiller. 1992. Stabilimenta characteristics of
the spider Argiope argentata on small islands: support of the
predator-defense hypothesis. Behavioral Ecology and Sociobiology
31:309-318.

Seah, W.K. & D. Li. 2001. Stabilimenta attract unwelcome predators
to orb-webs. Proceedings of the Royal Society of London B
268:1553-1558.

Seah, W.K. & D. Li. 2002. Stabilimentum variations in Argiope
versicolor (Araneae: Araneidae) from Singapore. Journal of
Zoology, London 258:531-540.

StatSoft, Inc. 2004. STATISTICA (data analysis software system).
Version 7. Online at www.statsoft.com

Timm, R.M. & M. Losilla. 2007. Orb-weaving spider, Argiope
savignyi (Araneidae), predation on the proboscis bat Rhynchonyc-
teris naso (Emballonuridae). Caribbean Journal of Science
43:282-284.

Tso, I-M. 1996. Stabilimentum of the garden spider Argiope
trifasciata: a possible prey attractant. Animal Behaviour 52:
183-191.

Tso, I-M. 1998a. Isolated spider web stabilimentum attracts insects.
Behaviour 135:311-319.

Tso, I-M. 1998b. Stabilimentum-decorated webs spun by Cyclosa
conica (Araneae, Araneidae) trapped more insects than undeco-
rated webs. Journal of Arachnology 26:101-105.

Tso, I-M. 1999. Behavioral response of Argiope trifasciata to recent
foraging gain: a manipulative study. American Midland Naturalist
141:238-246.

Vaclav, R. & P. Prokop. 2006. Does the appearance of orb-weaving
spiders attract prey? Annali Zoologici Fennici 43:65-71.

Watanabe, T. 1999. Prey attraction as a possible function of the
decoration of the uloborid spider Octonoba sybotides. Behavioral
Ecology 10:607-611.

Manuscript received 18 March 2008, revised 1 March 2009.

2009. The Journal of Arachnology 37:254-260

Conflict or cooperation in the courtship display of the white widow spider, Latmdectus pallidus

Ally R. Harari: Department of Entomology, The Volcani Center, Israel. E-mail: ally@int.gov.il

Merav Ziv: (Deceased, 1999) Mitrani Department of Desert Ecology, Blaustein Institutes for Desert Research, Ben-
Gurion University, Sede Boqer Campus, Israel

Yael Lubin: Mitrani Department of Desert Ecology, Blaustein Institutes for Desert Research, Ben-Gurion University,
Sede Boqer Campus, Israel

Abstract. We used experimental manipulations to test adaptive explanations for the courtship display of the male widow
spider, Latrodectus paUidus O. Pickard-Cambridge 1872. Two hypotheses have been suggested to explain a long and
complex male display: a) Cooperation of males and females in the effort to physically stimulate the female. As the time of
male arrival is not predictable, females may delay sexual readiness until the appearance of a courting male, b) Conflict
between males and females regarding the display cost. Females impose on the males an energetically costly display that may
last several hours as a test of their quality. To test both hypotheses, we manipulated the previous experience of either the
male or the female. We presented naive or experienced males (males that had courted and were accepted by females but
were prevented from copulating) to females that were either naive or experienced (had been courted by a male but
prevented from copulating). We also presented naive males to mated females. Following the stimulation hypothesis,
courted females were presumed to have been stimulated to mate and thus were expected to accept non-courting males as
mates. Both naive and mated females, however, were expected to await male stimulation before allowing copulation. In
contrast, the conllict of interest hypothesis predicts that the female tests each male for quality indicators and therefore a
non-courting male should not be accepted as a mate. Mated females, however, should apply a less stringent test to courting
males. Our results show that 1) naive females prevented males that did not perform a full courtship display from entering
the nest and mounting; 2) naive males courted virgin females with the full display, independent of the female previous
courting history; and 3) naive males shortened their courtship when presented with mated females. The results are
consistent with the conllict of interest hypothesis.

Keywords: Sexual stimulation, male quality, sexual conllict, Theridiidae

Elaborate, conspicuous and time-consuming precopulatory
displays are known in many animal taxa, including insects
(Thornhill 1976; Svensson et al. 1990), fishes (Milinski &
Bakker 1990), birds (Borgia 1995), and mammals (Behr & von
Helversen 2004), with several hypotheses suggested for their
function. These hypotheses, which are not necessarily mutu-
ally exclusive, postulate either a mutual interest of both male
and female in courtship or a conflict between the two. Mutual
interest may occur if the courtship enhances reproductive
isolation (Mayr 1963; Dobzhansky 1970) by providing cues
for species recognition (Ryan 1985; Andersson 1994) or if a
long courtship is necessary to stimulate the female into mate
(see reviews in Platnick 1971; Robinson 1982). A conflict may
arise between males wanting to increase their fitness by mating
as often and with as many females as possible, and females
who increase their fitness by choosing the best male available
(Trivers 1972; Parker 1979; Eberhard 1996). Conflict occurs
when females choose males based on their courtship display,
and subsequent escalation of the display imposes an increasing
cost on the males, which reduces the males’ fitness and the
potential to mate with additional females (Andersson 1994;
Eberhard 1996). Nevertheless, the male and female usually
share an interest in mating, particularly when the chance of
encountering mates is small (Segoli et al. 2006).

Some species of spiders have elaborate and costly precop-
ulatory displays, including cutting the female’s web, vibrating
on the female lines, and drumming vigorously on the substrate
(e.g., Robinson & Robinson 1980; Suter & Renkes 1984;
Forster 1995; Parri et al. 1997). Understanding the role of

males and females in shaping the courtship display in spiders is
challenging in light of the cannibalistic behavior of females in
many of these species (Elgar & Schneider 2004), since cues
allowing for species recognition to avoid predation may be
similar to those of mate assessment (Robinson & Robinson
1980; Andrade 1996; Schneider & Lubin 1998; Herberstein et
al. 2002).

In order to test these two general explanations for the male
courtship display, we investigated the courtship behavior of
the white widow spider, Latrodectus pallidus O. Pickard-
Cambridge 1872 (Theridiidae), inhabiting the Negev Desert,
Israel, which belongs to a genus known for its sexually
cannibalistic behavior (Ross & Smith 1979; Breene & Sweet
1985; Forster 1995; Andrade 1996; Segoli et al. 2006). Females
of L. pallidus are large, sedentary predators, while adult males
are less than a third of the female’s size and, as in other
Latrodectus species, actively search for females (Segoli et al.
2006). The female’s web consists of a nest located in a shrub
and connected by strong threads to a capture web consisting
of a loosely woven platform and thin, prey-capture threads
stretching from the platform to the ground (Lubin et al. 1991).
In a similar species found in the same habitat, L. revivensis
Shulov 1948, the male is attracted to the female’s web by
means of a female sex pheromone associated with the web silk
(Anava & Lubin 1993). On the web, males engage in a
vibratory display while cutting and removing sections of the
capture web before approaching the female’s nest and
engaging in tactile courtship (Segoli et al. 2006, 2008). The
courtship behavior has been described in several species of

254

HARARI ET AL.— WHITE WIDOW SPIDER COURTSHIP

Latrodectiis (Kaston 1970; Ross & Smith 1979; Lubin &
Anava 1993; Forster 1995), but not in L. paUidus.

To determine what factors shape the courtship display in L.
pallidus, we tested two hypotheses concerning the function of
the display and the context in which it is given: 1) cooperation
between the partners aimed at stimulating the female for
mating, and 2) a signal of male quality used by the female when
male and female interests over mating potentially conflict. Un-
der the first hypothesis, sedentary, virgin females that wait in
their nest for the arrival of conspecific males must be
physiologically stimulated before they are ready to mate
(Robinson & Robinson 1980; Suter & Renkes 1984). The
courtship display may provide the trigger that sexually primes
the female (Platnick 1971; Robinson 1982; Anava & Lubin

1993) . This hypothesis predicts that the male courts the female
until she signals her willingness to mate, and only upon
receiving this message will the male enter the female’s nest and
attempt to copulate. According to the alternative hypothesis of
conflict over mating, information provided by the male during
his display enables the female to arrive at a decision whether to
allow him to continue courting and later to copulate (e.g.,
Bukowski & Christenson 1997). This hypothesis predicts that a
choosy virgin female should demand a lengthy courtship, while
the male will attempt to reduce the effort he puts into courtship
in order to lower the energetic cost of the display. The courtship
display of many spider species consists of lengthy vibratory
signaling before contact is made with the female (Robinson &
Robinson 1980; Barth 1990; Amqvist 1992; Robertson & Adler

1994) , and male spiders may advertise their quality using these
vibrations (Coyle & O’Shields 1990; Mappes et al. 1996; Parri et
al. 1997). Vibrations of the substrate and the vigor of the
display may indicate to the female the size of the courting male
and his physical condition (e.g., Kotiaho et al. 1996; Rivero et
al. 2000; Singer et al. 2000; Maklakov et al. 2003). We
conducted experiments in which courtship was interrupted
before copulation, and previously courting males or females
were then paired with naive (non-courting) mates. By this
method, we could compare the behavior of males to naive,
virgin females and to females presumed to have been sexually
stimulated. The specific predictions for each of the two
hypotheses are described below.

METHODS

We collected juvenile and subadult widow spiders, L.
pallidus from sites around Beer Sheva in July-October, 1998;
March-April and June-August, 1999; and June-August, 2000
and 2001. Spiders were taken into the laboratory at the Sede
Boqer Campus of Ben-Gurion University, Israel, and kept at
26-28° C, approximately 30% relative humidity and a 10:14
light:dark cycle, similar to the prevailing hours of light and
dark for the time of year (March-April). Females were housed
in plexiglas cages (15 X 30 X 20 cm) with thin branches placed
in one corner, on which the nest and web were constructed.
We fed the females twice a week with nymphs of either locusts
(Locusta sp.) or crickets (Acheta domestica) and fed the males
once a week with either first instar locusts or adult fruit flies
(Drosophila melanogaster). We conducted all experiments after
the females had rebuilt their nests and molted to adulthood.
Recently molted virgin males and females were selected at
random with respect to body size.

255

In order to determine which of the hypotheses, cooperation
or conflict between the sexes, better explains the courtship
display we first documented courtship in L. pallidus, which
had not been described previously.

Male courtship display. — We placed L. pallidus males (n —
28) individually on the cage inner wall near the capture webs
of conspecific virgin females during the morning hours.
Observations were begun when the male moved onto the
female’s web and lasted for two hours or less if the male
entered the female’s nest, climbed onto the dorsal side of her
abdomen, and then moved to the ventral side into a mating
position. We defined the behavioral patterns of the display,
and recorded the sequence and starting time of each pattern,
for each of the males as well as the response of the courted
female. We noted the starting time of each behavioral pattern,
rather than its duration, because the behaviors were often
performed intermittently and short bouts of different compo-
nents alternated with one another. The time when a male
climbed onto the dorsal side of the female’s abdomen we
designated as the “commitment step”; after this step, 96.4% of
the males copulated successfully.

Testing the hypotheses: cooperation versus conflict of interest. —
According to the cooperation hypothesis, the male’s display is
aimed at stimulating the female. Thus, the female is expected to
signal to the male (e.g., by behavioral or pheromonal cues) when
she is ready to mate, and the courting male is expected to
respond by entering the nest, mounting the female and
copulating. The conjlict of interest hypothesis suggests that the
display provides the female with information about male
quality. Thus, the female imposes an energetically costly display
on the male as a test of his quality or physical condition. Under
this hypothesis, the female is expected to reject males that do not
display or whose display is in some way inadequate.

Rejection of a male by the female involves plucking or
jerking the web and even chasing the male from the nest
entrance. Acceptance of a male, however, does not usually
involve an overt behavior. Therefore, we used the response of
the male as an indication of a female signal to go on to the
next behavioral pattern in the display.

In the following (i-iii) experiments we used males and females
collected as juveniles during 1998-1999. In experiment (iv) we
used males and females collected during 2000-2001. In all
experiments, a virgin male was placed on the inner wall of the
cage containing an adult female with her nest and capture web.
We observed the pair for three hours or until the male mounted
the female’s abdomen, and we recorded the times from the start
of the courtship until the male a) entered the female nest and b)
reached the female abdomen. We conducted three experiments
(i-iii) with virgin males and females, in which males and females
were either naive, or had already engaged in courtship
(experienced). The fourth experiment (iv) compared the
behavior of virgin males to virgin or mated females.

We compared durations until the male entered the female’s
nest (duration of courtship on the web) and until he mounted
the female’s abdomen (total courtship duration) for the first
three tests (i-iii) described below, and used Tukey’s post hoc
test for pairwise comparisons of mean durations. The data
were tested for normal distribution (Liliefors test of the
residuals, Systat 10, 2000). ANOVA was used to test for
differences among the means for data that were distributed

256

THE JOURNAL OF ARACHNOLOGY

normally, and Kruskal-Wallis test for data that were not
normally distributed. In the fourth experiment (iv) we used a
Mann-Whitney U test to compare the response of males to
virgin and mated females.

(/) Naive males and naive females: The courtship of this
species is undescribed, so we observed the courtship
duration of naive males placed with naive virgin
females {n = 20). These data serve as a baseline
against which we compare the duration of male
displays in the subsequent experiments.

(//) Naive males and experienced females: In order to test
the likelihood that a signal is transferred from a
stimulated female to a courting male, we placed a
naive L. pallidas male onto a web of a female who was
courted, but not mated, by a previous male (“stimu-
lated” female) {n =11). The time between removal of
the first male at the commitment step and introducing
the second, naive male to the female was < 2 min.
Following the cooperation hypothesis, the naive
courting male is expected to receive an acceptance
signal from the already-stimulated female. He should
thus reduce his courtship effort and enter the nest after
only a short display. Following the conflict hypothesis,
the female should accept the new male only after a full
display on the web; the naive male should be unaware
that the female was courted previously and therefore
will perform the full display. However, since this
imposed scenario is not likely to occur often in nature,
a female may mistakenly perceive the second male
courtship as an extension of the first male’s display. In
this case the female is expected to be less aggressive,
and the male may subsequently reduce his display,
which may result in an intermediate display time.

{Hi) Experienced males and naive females: In order to test
the two hypotheses further, we observed the display
of a previously courting male presented with a naive
female {n = 10). As in the previous experiment, <
2 min lapsed between removing the male and present-
ing him to a naive female. The cooperation hypothesis
predicts that a male encountering a non-stimulated
female will start a new, lengthy courting display. The
conflict hypothesis predicts that a male that has
already courted a female, and thus provided informa-
tion regarding his quality, will cut short his display.
The naive female, however, is expected to reject the
male until he performs a full display on her web.

{iv) Naive males and mated females: We compared the
courtship duration of naive males presented with
mated {n = 20) and with virgin {n = 21) females. The
spiders were observed for three hours. Males were
removed from the cage after climbing on the female
abdomen, and the time was recorded. A week before
the experiment, males were left with females for 24 h
in order to obtain mated females. We regarded the
female as having mated if, two months later, at least

one egg sac was constructed and the spiderlings
hatched. The cooperation hypothesis predicts that a
male will display similarly to a mated or a virgin
female, since physical stimulation is required by the
female, independent of her mating status, before
mating can take place. Assuming first-male sperm
priority (Austad 1984; Segev et al. 2003), the conflict
hypothesis predicts that virgin females should be
choosier than mated females, forcing the males to
perform a lengthy display.

The predictions of the two hypotheses for each experiment
are summarized in Table 1.

Table 1 . — Predicted behavior of males in four experiments for each
of the two hypotheses proposed: I. Cooperation (mutual stimulation),
II. Conflict and female choice for male quality.

Test

I. Cooperation

II. Conflict

i)

Naive male and naive female

Display

Display

ii)

Naive male and experienced
female

Reduce display Display

iii)

Experienced male and
naive female

Display

Reduced display

iv)

Naive male and mated
female

Display

Reduced display

RESULTS

The courtship display. — The courtship display of L. palUdus
males is similar to other Latrodectus species (Kaston 1970;
Anava & Lubin 1993). The entire courtship is exceedingly
long: of 28 L. paUidus males observed courting virgin females,
only 19 completed the courtship display on the web and
entered the nest in less than two hours (mean ± SD, 109.21 ±
19.07 min).

Most of the display took place on the web, before entering
the nest. The final display was performed inside the nest before
copulation. The courtship on the female’s web consisted of a
series of complex movements performed in a specific order
(Fig. 1). The male entered the female’s web via the frame
threads and proceeded to walk on the web, while laying his
own dragline threads and disconnecting the lines of the
female’s web. He cut the thick attachment lines to the
substrate, the sticky threads attached to the ground, as well
as the threads of the platform and barrier web, and wrapped
the web silk into small bundles, which he suspended from the
female’s threads, usually near her nest (web-reduction
behavior: Watson 1986; Anava & Lubin 1993). Finally, he
performed small and rapid movements (jerking and abdomen
vibrations) on the female’s web before entering the nest,
followed by climbing on the dorsal side of the female’s
abdomen and then moving to her ventral side. Jerking
movements were sometimes performed inside the nest as well,
just before the male climbed on the female’s abdomen to
attempt copulation. While on the female’s ventral side, the
male drummed with his pedipalps on and near the female’s
genital openings (epigyeum), and finally inserted a pedipalp
into the female genital opening. A female might chase the male

HARARI ET AL.— WHITE WIDOW SPIDER COURTSHIP

Figure 1 .—The sequences of behaviors constituting the courtship
display of male widow spiders Latrodectus paUidus {n = 19). Shown
are mean ± SD of the first occurrence (in minutes from the start of
the display) of each behavior component. Note that some compo-
nents were performed in alternation (e.g., disconnecting capture-web
threads, cutting of frame threads and bundling of threads). “On
abdomen” refers to climbing onto the dorsal side of the
female’s abdomen.

away at various stages, and the male would then resume courting
on the web, outside the nest. Once the male was on the female’s
abdomen, however, copulation usually followed. We referred to
the male mounting the female’s abdomen as the commitment step,
and in the following analyses we have taken this step as an
indication of the female’s acceptance of the courting male.

Cooperation versus conflict of interest hypotheses. — Duration
of courtship on the web: The durations of male courtship prior
to entering the female’s nest were not normally distributed.
The display time before entering the female’s nest differed
significantly among the four tests (Kruskal-Wallis test: H =
21.601, P < 0.001, n = 38). In each of the different
experiments, some males did not enter the nest within three
hours or did not reach the commitment step during the
observation time. These males, fewer than 20% of each
experiment, were excluded from the statistical analyses.

Naive males that courted naive females (i) (n = 18),
displayed for 77.67 ± 39.91 min {mean ± SD) before entering
the nest. Naive males that were placed in cages of virgin
females that had been courted previously by another male (ii)
{n = 10), displayed for 68.80 ± 26.26 min before first entering
the nest. Ail males {n = 10) that were transferred to cages of
naive females after reaching the commitment step in the nest
of another female (iii) immediately attempted to enter the
second female’s nest (display duration 3.0 ± 1.94 min), and all
were chased away by the female.

Total courtship duration: There was no significant difference
among the different experiments in overall courtship time until
reaching the commitment step on the female’s abdomen
(ANOVA: ^2,28 = 2.862, P = 0.08). Naive males that courted
naive females (i) (n = 16), reached the commitment step after a
total display duration of 101.73 ± 23.55 min. When paired
with previously courted virgin females (h) (« = 8), the display
duration of naive L. palUdus males until the commitment step,
was 72.5 ± 16.66 min. Virgin males that had previously
courted an experienced female and were then placed in a cage
with a naive female {in) were chased out of the nest by the

257

Naive males & Naive males & Experienced Naive males &
naive females experienced males & naive mated females
females females

Figure 2. — Durations (mean ± SD) of male courtship on the web
until entering the female’s nest (black bars) and of the total courtship
display until the commitment step (white bars) for each of the
four tests.

female when they first tried to enter the female’s nest. One of
these rejected males did not resume courting, and an
additional male did not reach the commitment step within
three hours. All other males (« = 8) resumed courting on the
web and repeated all behavioral patterns in the normal order
(Fig. 1). For these males, courtship duration until the
commitment step was 86.57 ± 24.91 min.

Virgin males tested with mated females (iv) did not cut the
thick attachment threads of the female’s web or bundle them
in front of her nest. Instead they briefly laid dragline silk on
the female’s web, cut some of the thin web threads and rapidly
entered the female’s nest and climbed on the her abdomen,
jerking briefly just before entering the nest (duration until
entering the nest 10.41 ± 17.38 min, until commitment step
23.33 ± 14.62, n = 15). Naive males courting virgin females (n
= 21) performed the typical courtship sequence until entering
the female nest (60.05 ± 33.1 min) and reached the female’s
abdomen (90.7 ± 35.6 min) after a significantly longer time
(Mann-Whitney U test: U = 284.5, n = 36, P < 0.001 and U =
308.0, n = 36, P < 0.001, respectively). The duration of the
courtship display of males in all of the experiments are
summarized in Fig. 2.

DISCUSSION

The results of the experimental manipulation of male and
female experience suggest that the male’s courtship display is
better explained by the hypothesis of conflict over mating
interests between the sexes than that of male and female
cooperation to sexually stimulate the female. We discuss the
two hypotheses suggested in the light of our results.

Cooperation. — In some spider species, the female signals her
acceptance by adopting a receptive posture (Robinson &
Robinson 1980), whereas in others females indicate receptivity
by remaining stationary (Forster 1982). In the widow spider L.
pcdlidus, the females draw away from the nest wall towards its
center, allowing the male the space needed to approach the
ventral side of her abdomen. Testing the hypothesis of
cooperation aimed at stimulating the female, we predicted
that females that were previously stimulated would signal their
readiness to mate. Upon perceiving the signal, a male should
respond by ceasing his display and entering the nest. However,
we found that males did not change their display when they
encountered previously courted females (test ii). Apparently,

258

THE JOURNAL OF ARACHNOLOGY

they did not identify these females as receptive, and their
courtship duration was not statistically different from that of
males courting naive females (test i). In addition, a male that
was interrupted during his display at the commitment step and
then transferred to a cage with a naive female (iii) immediately
attempted to enter the female’s nest and proceed with his
display from the point at which it was interrupted. Thus, we
suggest that the male failed to receive a signal from the female
indicating that she was not stimulated to mate. Finally, if
courting functions to stimulate the female physically before
mating, previously mated females presented with a male
several days after their first mating should not differ from
virgins in requiring a stimulating courtship display. Contrary
to this prediction, however, we found that males shortened
their display when they encountered mated females.

Conflict over mating interests. — The courtship display of the
male widow spider L. pallidus is lengthy and consists of
behaviors such as web vibration (jerking) and cutting and
bundling of silk that are likely to be energetically costly. The
entire process may take more than two hours before the male
enters the nest and a total of four or more hours before he
copulates with the female (A.R. Harari, personal observation).
At each stage of the display, females may gain information
concerning the male’s quality.

Our experiments revealed that virgin females prevented males
from mounting and copulating if the males had not performed
all parts of the display, and naive males courted virgin females
regardless the female’s previous experience. These results suggest
that males expect to be accepted by females only after displaying
the full courtship, and that a female accepts a male only after he
completes a full display on her web. We tested both sides of the
coin by observing whether a naive male courted a female
immediately after she had been courted by a previous male up to
the commitment step (ii). In this experiment, males engaged in
the full courtship display, suggesting that they did not receive
any cue from the female. In the reciprocal experiment (iii), a
naive female was courted by a male whose courtship with a
different female was interrupted at the point of entering her nest.
The male’s response, to continue from the point that he had
stopped, suggests, again, that he initially received no cue from
the female. The female’s aggressive response, however, suggests
that she had not had an opportunity to assess the male’s quality
and therefore rejected his attempt to enter the nest. Overall, we
interpret our results as indicating that the male’s display is under
female control, such that males that attempt to shorten the
display are prevented by virgin females from proceeding with
close-range courtship inside the nest, mounting and copulation.

Additional support for the hypothesis of a conflict over
mating interests and female control over the length of the
male’s display comes from the results comparing the male’s
display to a mated or virgin female. Female widow spiders may
mate with more than one male (Anava & Lubin 1993; Andrade
1996; Segev et al. 2003; Segoli et al. 2006). We found that males
of L. pallidus shortened their display significantly when
courting mated females, a behavior that is expected if there is
first-male sperm priority, and thus, a lower probability of
fathering the offspring from a second or later mating. Segev et
al. (2003) showed evidence for first-male sperm priority in L.
revivensis, as is the case in many other entelegyne spider species
that have been tested (Christenson & Cohn 1988; Watson &

Lighton 1994; Singer & Reichert 1995; Snow & Andrade 2005,
but see e.g., Schneider at al. 2000 for a case of mixed paternity).
In L. pallidus (Segoli et al. 2006) as well as in other Latrodectus
species (Levi 1959; Bhatnagar & Rempel 1962; Kaston 1970;
Foelix 1996; Berendonck and Greven 2002; Snow et al. 2006;
Segoli et al. 2008), the tip of the male’s embolus is often broken
during copulation and becomes lodged in the insemination duct
or inside the spermatheca. The presence of the broken embolus
tip may act as a mating plug and may reduce the likelihood that
a second male will successfully inseminate the female (Beren-
donck & Greven 2002; Segoli et al. 2008). Thus, if the first male
is accepted only after a stringent test and the contribution of the
second male to the clutch is limited, a mated female may be less
choosy with subsequent males. Accepting a second male with
little courtship could be a bet-hedging strategy, as suggested for
Neriene litigiosa (Keyserling 1886) (Linyphiidae) by Watson
(1991b).

Our results indicate that males are able to distinguish between
virgin and mated females (as shown in L. hasselti, Stoltz et al.
2007) and reduce their display effort to the latter. Pheromones
produced by the female have been shown to play an important
role in mate attraction (e.g., L. revivensis: Anava & Lubin 1993)
and may also provide a means of assessing female reproductive
state on the web (Papke et al. 2001). However, it will always be
advantageous for males to shorten the display duration and
reduce its cost if the female will allow it. In courting virgin
females, males frequently attempt to enter the nest and are
repeatedly chased off by the female to continue their display on
the web. This behavior suggests that males repeatedly test the
female’s aggressive intentions during courtship; virgin females
reject males that have not met a criterion, whereas mated
females accept a second male more readily.

In conclusion, our experiments suggest that the costly
display of the widow spider L. pallidus is unlikely to function
as physical stimulation of the female to mate. Rather, the
courtship display in this species is likely a result of a conflict of
interests, with the female imposing a long, vigorous and
energetically costly display in order to test the male’s quality.

In recent years, the view of reproduction as a cooperative
effort has been challenged by increasing evidence for conflicting
interests between the sexes (Dawkins 1976; Parker 1979;
Holland and Rice 1998; Zeh and Zeh 2003). Although both
parents share an interest in maximizing the fitness of their
offspring, they often have conflicting interests in the amount of
reproductive effort (Parker 1979). This conflict begins with the
investment in the size of male and female gametes (anisogamy,
Trivers 1972) and continues with the conflict over the number of
matings (Bateman 1948). As a consequence, females are
expected to be choosy, selecting some males and rejecting others
based on their phenotypic traits (Andersson 1994; Amqvist and
Rowe 2005). This scenario may lead to the complex and lengthy
male display in various spider species, including L. pallidus
(Kaston 1970; Ross & Smith 1979, Lubin and Anava 1993;
Forster 1995). The fixed components observed in the display of
L. pallidus and other Latrodectus species (e.g., Lubin and Anava
1993) can be viewed in the light of the known predatory and
cannibalistic nature of the genus (Elgar & Schneider 2004) and
may be aimed at appeasing the females by providing cues for
species recognition (Ryan 1985; Andersson 1994). The length of
the display and its energetic cost, however, may have evolved as

HARARI ET AL.— WHITE WIDOW SPIDER COURTSHIP

a consequence of female choice for energetically displaying
males. The results of our experiments support the latter view.

ACKNOWLEDGMENTS

Merav Ziv died tragically in 1999. We thank O. Eitan for
assistance with many aspects of the study and B. Berendonck
for discussion of the research. We also thank M. Segoli, T.
Bilde, A. Maklakov, M. Salomon, M. Elgar, J. Sandidge, J.
Kotiaho and an unknown referee for comments. The study
was funded by grant no. 717-96 from the Israel Science
Foundation (to YL) and a fellowship (to ARH) of the Israel
Council for Higher Education (VATAT) from Ben-Gurion
University. This is publication 640 of the Mitrani Department
of Desert Ecology.

LITERATURE CITED

Anava, A. & Y. Lubin. 1993. Presence of gender cues in the web of a
widow spider, Latrodectiis revivensis, and a description of
courtship behaviour. Bulletin of the British Arachnological Society
9:119-122.

Andersson, M. 1994. Sexual Selection. Princeton University Press,
Princeton, New Jersey.

Andrade, M.C.B. 1996. Sexual selection for male sacrifice in the
Australian redback spider. Science 271:70-72.

Arnqvist, G. 1992. Courtship behavior and sexual cannibalism in
semi-aquatic fishing spider, Dolmedes fimhriatus (Clerck) (Ara-
neae: Pisauridae). Journal of Arachnology 20:222-226.

Arnqvist, G. & L. Rowe. 2005. Sexual Conflict. Princeton University
Press, Princeton, New Jersey.

Austad, S.N. 1982. First male sperm priority in the bwl and doily
spider Frontinella pyramitela (Walckenaer). Evolution 36:777-785.
Austad, S.N. 1984. Evolution of sperm priority patterns in spiders.
Pp. 223-250. In Sperm Competition and the Evolution of Animal
Mating Systems. (R.L. Smith, ed.). Academic Press, Orlando,
Florida.

Barth, F.G. 1990. Spider courtship: male vibrations, female respon-
siveness and reproductive isolation. Pp. 161-166. In Sensory
Systems and Communication in Arthropods: Advances in Life
Sciences. (F.G. Gribakin, K. Wiese & A.V. Popov, eds.).
Birkhauser Verlag, Basel, Switzerland.

Behr, O. & O. von Helversen. 2004. Bat serenades - complex courtship
songs of the sac-winged bat (Sac'copteryx hilineata). Behavioral
Ecology and Sociobiology 56:106-115.

Berendonck, B. & H. Greven. 2002. Morphology of female and male
genitalia in Latrodectiis revivensis Shulov, 1948 with regard to
sperm priority patterns. Pp. 157-167. In European Arachnology
2000. (S. Toft & N. Scharff, eds.). Aarhus University Press,
Aarhus, Denmark.

Bhatnagar, R.D.S. & J.G. Rempel. 1962. The structure, function, and
postembryonic development of the male and female copulatory
organs of the black widow spider Latrodectiis citracaviensis
(Muller). Canadian Journal of Zoology 40:465-510.

Breene, R.G. & M.H. Sweet. 1985. Evidence of insemination of
multiple females by the male black widow spider, Latrodectiis
mactans (Araneae, Theridiidae). Journal of Arachnology
13:331-335.

Borgia, G. 1995. Complex male display and female choice in the
spotted bowerbird: specialized function for different bower.
Animal Behaviour 49:1291-1301.

Bukowski, T.C. & T.E. Christenson. 1997. Natural history and
copulatory behavior of the spiny orbweaver spider Micrathena
gracilis (Araneae, Araneidae). Journal of Arachnology 25:307-320.
Bukowski, T.C. & T.E. Christenson. 2000. Determinants of mating
frequency in the spiny orbweaving spider, Micrathena gracilis
(Araneae, Araneidae). Journal of Insect Behavior 13:331-352.

259

Christenson, T.E. & J. Cohn. 1988. Male advantage for egg
fertilization in the golden orbweaving spider. Journal of Compar-
ative Psychology 102:312-318.

Coyle, F.A. & T.C. O’Shields. 1990. Courtship and mating behavior
of Thelechoris kaeschi (Araneae, Dipluridae), an African funnel-
web spider. Journal of Arachnology 18:281-296.

Dawkins, R. 1976. The Selfish Gene. Oxford University Press,
Oxford, UK.

Dobzhansky, T. 1970. Genetics of the Evolutionary Process.
Columbia University Press, New York.

Eberhard, W.G. 1996. Female Control: Sexual Selection by Cryptic
Female Choice. Princeton University Press, Princeton, New Jersey.

Elgar, M.A. & J.M. Schneider. 2004. The evolutionary significance of
sexual cannibalism. Advances in the Study of Behavior 34:135-163.

Foelix, R. 1996. Biology of Spiders. Second edition. Oxford
University Press, New York.

Forster, L.M. 1982. Visual communication in jumping spiders
(Salticidae). Pp. 161-212. In Spider Communication: Mechanisms
and Ecological Significance. (P.N. Witt & J.S. Rovner, eds.).
Princeton University Press, Princeton, New Jersey.

Forster, L.M. 1995. The behavioural ecology of Latrodectiis hasselti
(Thorell), the Australian redback spider (Araneae: Theridiidae): a
review. Records of the Western Australian Museum Supplement
52:13-24.

Herberstein, M.E., J.M. Schneider & M.A. Elgar. 2002. Costs of
courtship and mating in a sexually-cannibalistic spider: female
mating strategies and their consequences for males. Behavioral
Ecology and Sociobiology 51:440^46.

Holland, B. & W.R. Rice. 1998. Perspective: Chase-away sexual
selection: antagonistic seduction versus resistance. Evolution
52:1-7.

Kaston, B.J. 1970. Comparative biology of American black widow
spiders. Transactions of the San Diego Society of Natural History
16:33-82.

Kotiaho, J., R.V. Alatalo, J. Mappes & S. Parri. 1996. Sexual
selection in a wolf spider: male drumming activity, body size and
viability. Evolution 50:1977-1981.

Levi, H.W. 1959. The spider genus Latrodectiis (Araneae, Theridi-
idae). Transactions of the American Microscopical Society

78:7-43.

Lubin, Y., M. Kotzman & S. EUner. 1991. Ontogenetic and seasonal
changes in webs and websites of a desert widow spider. Journal of
Arachnology 19:40-48.

Maklakov, A. A., T. Bilde & Y. Lubin. 2003. Vibratory courtship in a
web-building spider: signaling quality or stimulating the female?
Animal Behaviour 66:623-630.

Mappes, J., R.V. Alatalo, J. Kotiaho & S. Parri. 1996. Viability costs
of condition-dependent sexual male display in a drumming wolf
spider. Proceedings of the Royal Society, London, Series B
263:785-789.

Mayr, E. 1963. Animal Species and Evolution. Harvard University
Press, Cambridge, Massachussetts.

Milinski, M. & T.C.M. Bakker. 1990. Female sticklebacks use male
coloration in male choice and hence avoid parasitized males.
Nature 344:330-333.

Papke, M.D., S.E. Reichert & S. Schultz. 2001. An airborne female
pheromone associated with male attraction and courtship in a
desert spider. Animal Behaviour 61:877-886.

Parker, G.A. 1979. Sexual selection and sexual conlJict. Pp. 123-166.
In Sexual Selection and Reproduction Competition in Insects.
(M.S. Blum & N. Blum, eds.). Academic Press, New York.

Parri, S., R.V. Alatalo, J. Kotiaho & J. Mappes. 1997. Female choice
for male drumming in the wolf spider Hygrolycosa riibrofasciata.
Animal Behaviour 53:305-312.

Platnick, N. 1971. The evolution of courtship behaviour in spiders.
Bulletin of the British Arachnology Society 2:40-47.

260

THE JOURNAL OF ARACHNOLOGY

Rivero, A., R.V. Alatalo, J.S. FCotiaho, J. Mappes & S. Parri. 2000.
Acoustic signaling in a wolf spider: can signal characteristics
predict male quality? Animal Behaviour 60:187-194.

Robertson, M.H. & P.H. Adler. 1994. Mating behavior of Fiorina
coccinae (Hentz) (Araneae: Linyphiidae). Journal of Insect
Behavior 7:313-326.

Robinson, M.H. 1982. Courtship and mating behavior in spiders.
Annual Review of Entomology 27:1-20.

Robinson, M.H. & B. Robinson. 1980. Comparative studies on the
courtship and mating behavior of tropical araneid spiders. Pacific
Insects Monograph 36:1-218.

Ross, K. & R.L. Smith. 1979. Aspects of courtship behavior of the
black widow spider, Latrodectus hespenis (Araneae, Theridiidae),
with evidence for the existence of a contact sex pheromone. Journal
of Arachnology 7:69-77.

Ryan, M.J. 1985. The Tungara Frog. University of Chicago Press,
Chicago.

Schneider, J.M. & Y. Lubin. 1998. Intersexual conllict in spiders.
Oikos 83:496-506.

Schneider, J.M., M.E. Herberstein, F.C. De Crespigny, S. Rama-
murthy & M.A. Elgar. 2002. Sperm competition and small size
advantage for males of the golden orb-web spider Nephila edidis.
Journal of Evolutionary Biology 13:939-946.

Segev, O., M. Ziv & Y. Lubin. 2003. The male mating system in a
desert widow spider. Journal of Arachnology 31:379-393.

Segoli, M., A.R. Harari & Y. Lubin. 2006. Limited mating
opportunities and male monogamy: a field study of white widow
spiders, Latrodectus pallidus (Theridiidae). Animal Behaviour
72:635-642.

Segoli, M., Y. Lubin & A.R. Harari. 2008. Frequency and
consequences of damage to male copulatory organs in a widow
spider. Journal of Arachnology 36:533-537.

Segoli, M., R. Arieli, P. Sierwald, A.R. Harari & Y. Lubin. 2008.
Sexual cannibalism in the brown widow spider (Latrodectus
geometricus). Ethology 114:279-286.

Singer, F. & S.E. Riechert. 1995. Mating system and mating success
of the desert spider Agelenopsis aperta. Behavioral Ecology and
Sociobiology 36:313-322.

Singer, F., S.E. Riechert, H.F. Xu, A.W. Morris, E. Becker, J.A. Hale
& M.A. Noureddine. 2000. Analysis of courtship success in the
funnelweb spider, Agelenopsis aperta. Behaviour 137:93-117.

Snow, L.S.E. & M.C.B. Andrade. 2004. Pattern of sperm transfer in
redback spiders: implications for sperm competition and male
sacrifice. Behavioral Ecology 15:785-792.

Snow, L.S.E. & M.C.B. Andrade. 2005. Multiple sperm storage
organs facilitate control of paternity. Proceedings of the Royal
Society of London, Series B 272:1139-1144.

Snow, L.S.E., A. Abdel-Mesih & M.C.B. Andrade. 2006. Broken
copulatory organs are low-cost adaptations to sperm competition
in redback spiders. Ethology 112:379-389.

Stoltz, J.A., J.N. McNeil & M.C.B. Andrade. 2007. Males assess
chemical signals to discriminate just-mated females from virgins in
redback spiders. Animal Behaviour 74:1669-1674.

Suter, R.B. & G. Renkes. 1984. The courtship of Front inella
(Araneae, Linyphiidae): patterns, vibrations and function. Journal
of Arachnology 12:37-54.

Svensson, B.G., E. Petersson & M. Frisk. 1990. Nuptial gift size
prolongs copulation duration in the dance fly, Etnpis borealis.
Ecological Entomology 15:225-229.

SYSTAT. 2000. Version 10, SPSS Inc. Prentice-Hall, Englewood
Cliffs, New Jersey.

Thornhill, R. 1976. Sexual selection and nuptial feeding behaviour in
Bitttacus apicalis (Insecta: Mecoptera). American Naturalist
110:529-548.

Trivers, R.L. 1972. Parental investment and sexual selection.
Pp. 136-179. In Sexual Selection and the Descent of Man, 1871-
1971. (B. Campbell, ed.). Heinemann, London.

Watson, P.J. 1986. Transmission of a female sex pheromone thwarted
by males in the spider Linvphia liiigiosa (Linyphiidae). Science
233:219-221.

Watson, P.J. 1991a. Multiple paternity and first mate sperm
precedence in the Sierra dome spider, Linyphia litigiosa. Animal
Behaviour 41:135-148.

Watson, P.J. 1991b. Multple paternity as genetic bet hedging in
female Sierra dome spiders, Linyphia litigiosa (Linyphiidae).
Animal Behaviour 41:342-360.

Watson, P.J. & J.R.B. Lighton. 1994. Sexual selection and the
energetics of copulatory courtship in the Sierra dome spider,
Linyphia litigiosa. Animal Behaviour 48:615-626.

Zeh, J.A. & D.W. Zeh. 2003. Toward a new sexual selection paradigm:
Polyandry, conflict and incompatibility. Ethology 109:929-950.

Manuscript received 26 November 2008, revised 14 March 2009.

2009. The Journal of Arachnology 37:261-265

Redescriptioo and transfer of Geolycosa gmndis (Araneae, Lycosidae) to the genus Hogna

Jozef Slowik: University of Alaska Museum of the North, Department of Entomology, 907 Yukon Drive, Fairbanks,
Alaska 99775, USA. E-mail: ftjas4@uaf.edu

Paula E. Cushing: Denver Museum of Nature and Science, Department of Zoology, 2001 Colorado Blvd., Denver,
Colorado 80205, USA

Abstract. Geolycosa grandis (Banks 1894) (Araneae, Lycosidae) is redescribed, and illustrations are provided for the first
time. We propose that Geolycosa grandis be transferred to the genus Hogna Simon 1885. Hogna penmmda (Chamberlin
1904) is synonymized with H. grandis. Notes are given on distinguishing H. grandis from similar species including H. helluo
(Walckenaer 1837), Allocosa georgicola (Walckenaer 1837), and H. aspersa (Hentz 1844). Information on Lycosa permiana
Scheffer 1904 is examined and the species is declared nomen dubium.

Keywords: Wolf spider, taxonomy, synonymy, new combination, nomen dubium

Geolycosa grandis (Banks 1894) is a large wolf spider found
throughout the north central Great Plains of North America.
Although a large wolf spider, arachnologists have routinely over-
looked or misidentified the animal almost from the time of its original
description. Banks’s original 1894 description provides an adequate
depiction of Lycosa grandis for correct identification. However, likely
due to a lack of illustrations, the species has remained obscure to date.
Chamberlin, who described the spider as Lycosa permunda in 1904,
had apparently either not read or misunderstood Banks’s original
description. Chamberlin included both L. permunda and G. grandis
(as Lycosa grandis) in his lycosid revision of 1908; however,
Chamberlin indicates that he had not examined Banks’s type
specimen of L. grandis from Colorado; he only had studied a
specimen he identified as L. grandis from Baja California. It is not
known what species Chamberlin identified as L. grandis from Baja as
the specimen referred to could not be located. Research conducted for
this paper indicates that the range of L. grandis does not extend that
far south.

We contacted any institution that might have possession of the type
L. permunda, including the American Museum of Natural History,
United States National Museum, Museum of Comparative Zoology at
Harvard University, University of Kansas Natural History Museum,
and Brigham Young University Museum. In his description, Cham-
berlin makes note of a spider having a carapace length over 10 mm and
a “pale narrow median line extending backward from first eye row,
widening abruptly in front of dorsal groove.” He also mentions a clear
abdominal pattern, light venter, and an epigynum as in Hogna helluo
(Walckenaer 1837). The lack of another spider having these characters
in the Colorado/Kansas area indicates it is G. grandis.

The holotype male of Lycosa grandis, deposited at the Museum of
Comparative Zoology, was found in a vial with a male L. coloradensis
Banks 1894 and a female Agelenopsis aperta Gertsch 1934. It is not
clear why these specimens are together in the type vial. Lycosa grandis
was transferred to Geolycosa by Roewer (1955); however, the reason
for this transfer was not made clear.

The proposed transfer of Geolycosa grandis into the genus Hogna is
based on comparisons with the generotype Hogna radiata (Latreille
1817) from Montpellier, France, provided by Charles Dondale (the
holotype is presumed lost), as well as comparisons with other North
American representatives of the genus Hogna. The species is placed
into Hogna based on similarities of the palpal structure, specifically
the similar median apophysis shape, the two-part terminal
apophysis, and the palea shape. The species also shares similarities
in the internal and external epigynal structures, specifically the
spermathecal shape, as well as similarities in the carapace shape, eye

location, and leg length. Illustrations of H. grandis are provided
here for the first time.

Justification for the removal of H. grandis from Geolycosa is also
supported by the lack of characteristics diagnostic for Geolycosa,
specifically the lack of a sloped carapace, a defining character for the
genus Geolycosa (Wallace 1942; Dondale & Redner 1990). In
addition, the species differs behaviorally. According to Wallace
(1942), spiders of the genus Geolycosa spend practically their whole
existence in a burrow and females rarely leave the burrow. In
contrast, G. grandis is a non-obligate burrower and the females spend
considerable time roaming outside the burrow.

METHODS

Illustrations were made from digital photographs taken with an
Olympus U-CMAD3 digital camera mounted on an Olympus SZX12
stereo-microscope. Scanning electron micrographs were taken with a
Hitachi TM-1000 tabletop microscope. All measurements are in
millimeters. Specimens used for this study are housed in the
arachnological collection at the Denver Museum of Nature &
Science, the University of Nebraska Museum, and the Museum of
Comparative Zoology at Harvard University. Specimens of compar-
ison species are also housed in the DMNS collection.

Abbreviations. — MA = median apophysis; TA= terminal apoph-
ysis; MS = median septum; PLE = posterior lateral eyes; PER =
posterior eye row; AME = Anterior median eyes; AMNH =
American Museum of Natural History; BYU = Brigham Young
University; DMNS = Denver Museum of Nature & Science; KU =
Kansas University; MCZ = Museum of Comparative Zoology;
USNM = United States National Museum, Smithsonian Institution.

TAXONOMY

Hogna grandis (Banks 1894) new combination
Figs. 1-6

Lycosa grandis Banks 1894:49; Chamberlin 1908:229.

Lycosa permunda Chamberlin 1904:286; Chamberlin 1908:233, NEW

SYNONYMY.

Hogna permunda Roewer 1955: 259; Platnick 2008.

Geolycosa grandis (Banks 1894) Roewer 1955:244; Platnick 2008.

Type material. — Holotype: USA: COLORADO: Larimer County,
Fort Collins (40.35°N, 105.05°W, elev. 1525 m), male, no date, MCZ.
Examined.

Holotype male and paratype female: Lycosa permunda Cham-
berlin 1904: USA: Kansas, no date. Unable to locate specimens.

261

262

THE JOURNAL OF ARACHNOLOGY

Figures 1-5. — Hogna grcmdis (Banks 1894). Figs. 1-3. Male (DMNS. ZAl 1887): 1. Dorsal view of carapace; 2, 3. Ventral view of palpus; p =
palea, MA = median apophysis. Figs. 4,5. Female (DMNS. ZA.9572): 4. Ventral view of epigynum; 5. Dorsal view of spermathecae; be =basal
chamber, 11 = lateral lobe, tc = terminal chamber, ss = spermathecal stalk. Scale line = 1 m.

SLOWIK & CUSHING— REDESCRIPTION OF GEOLYCOSA GRANDIS

263

Figures 6, 7. — Scanning electron micrographs of male palpus. 6. Hogna grandis (Banks 1894); 7. Hogna helluo (Walckenaer 1837).

Other material examined: 17 males and 25 females; USA:
COLORADO: Arapahoe County, IM, 21309 E. Belleview (40.58°N,
106.23°W), no date, B. Parshley, DMNS ZA. 12744; IF, Aurora
(39.65°N, 104.75°W), 20 August 2001, B. Shipley, DMNS ZA. 12740;
IF, 1-70, 0.5 mi (0.8 km) N of Elbert Co. Line (39.58°N, 104.02°W),
30 August 2004, H. Guarisco, DMNS ZA.9572; Boulder County, IF,
2 mi (3.2 km) E. Marshall (39.96°N, 105.23°W), 16 April 1961, B.
Vogel, DMNS ZA.2128; IM, Boulder (40.02°N, 105.31°W), 31 July
1961, D. Ward, DMNS ZA.2129; Delta County, IM, North entrance
Crawford State Park (38.70°N, I07.60°W), 31 July 1999, C. Swinney,
DMNS ZA. 12738; Denver County, IF, Washington Bay in Denver
near Southwest Plaza (39.83°N, 105.20°W), October 1998, D. M.
Endricks, DMNS ZA. 12739; Douglas County, IM, 10615 Jewelbenny
Trail, Highlands Ranch (39.32°N, 104.93“W), 17 August 2006, no
collector given, DMNS ZA.13376; IF, 6414 E. Dutch Creek St.,
Denver (39.7 1°N, 104.99°W, 12 August 2000, N. & J. Tuchton,
DMNS ZA. 12737; IF, Sharptail Open Space, Louviers (39.45°N,
105.05°W), 27 May 2005, B. Morrison, DMNS ZA. 14252; El Paso
County, IF, Bilerest Terrace (38.88°N, 104.76°W), 16 July 2001, A.
Broughton, DMNS ZA.l 1888; Jefferson County, IM, 9797 West Ohio
Avenue, Lakewood (39.70°N, 106.11°W), August 2001, E. House,
DMNS ZA. 12721; Larimer County, IM, 1756 Haase Court, Berthoud
(40.31°N, lOS.SrW), 18-20 August 1999, P. Phillips, DMNS
ZA.12731; IF, 2120 Bridgefield Ln, Ft. Collins (40.56°N,
105.10°W), 29 August 2004, J. Enstrom, DMNS ZA.l 3770; IF,
24205 Colorado Ave, Loveland (40.37°N, 105.08°W), 30 July 1999,
D. Goldade, DMNS ZA.l 1840; IF, 7894 Little Fox Lane, Wellington
(40.70°N, 105.00°W), 14 September 2000, M. Payew, DMNS
ZA.l 1883; IF, Dixon Reservoir (40.55°N, I05.14°W), 24 May 2000,
D. Chlebown, DMNS ZA.l 1886; IF, Environmental Learning Center
(40.57°N, lOS.OrW), 25 September 1999, J. M. Diez, DMNS
ZA.l 1882; IF, Fort Collins (40.58°N, 105.1 1°W), 30 August 1973,
W. D. Frank, DMNS ZA.l 1875; IM, Same locale, 10 August 1970,
W. D. Frank, DMNS ZA. 11880; IM, Same locale, 24 November
1980, W. D. Frank, DMNS ZA.l 1881; !M, Same locale, 9 August
1987, D. Johnson, DMNS ZA. 11872; IF, Same locale, 13 Jun 1989,
B. Holter, DMNS ZA.l 1797; IM, Same locale, 4 March 1985, W. D.
Frank, DMNS ZA.l 1879; IF, Same locale, 15 July 1980, W. D.
Frank, DMNS ZA.l 1877; IF, Same locale, 4 January 1971, W. D.
Frank, DMNS ZA.l 1876; IM, Same locale, 23 September 2001, L.
Sandner, DMNS ZA.l 2698; IF, Same locale, 11 September 1990, no
collector given, DMNS ZA.l 1874; IM IF, Same locale, 1 September
1977, W. D. Frank, DMNS ZA.l 1873; IF, Same locale, 18 September
1973, W. D. Frank, DMNS ZA.l 1878; IF, Same locale, 22 October
1982, D. Clarkson, DMNS ZA.l 1798; IF, Loveland (40.40°N,
105.1 rW), 10 October 1967, W. D. Frank, DMNS ZA.l 1885; IM,
Same locale, 7 August 1990, Kilburn, DMNS ZA.l 1884; IM,
Loveland (40.40°N, 105.1 TW), no date, A. Randall, DMNS

ZA.l 27 18; Weld County, IM, Bones Galore Paleo Site, Pawnee
National Grassland (40.73°N, 103.80°W), 15 August 2001, T. Hiester,
DMNS ZA. 11887; IF, Bones Galore Paleo Site, Pawnee National
Grassland (40.75°N, 103.80°W), 18 August 2000, T. Hiester, DMNS
ZA. 12722; KANSAS: Montgomery County, IF, 1 mi (1.6 km) E
Havana Reservoir (37.22°N, 95.71°W), 25 July 1974, no collector
given, DMNS ZA.l 1306; Wyandotte County, IF, Mission Grade
School 2 mi (3.2 km) N. Bonner Springs (39.06°N, 94.88°W), 29
September 1977, R. Huggins, DMNS ZA.l 1308; IM, Mission Grade
School, 3 mi (4.8 km) N. Banner Springs (39.16°N, 94.83°W), 29
September 1977, R. Hugging, DMNS ZA.l 1841; WYOMING:
Campbell County, IM IF, Gillette (44.29°N, 105.50°W), June 2005,
no collector given, DMNS ZA. 14260.

Diagnosis. — Hogna grandis can be diagnosed by its size, total length
over 20 mm; carapace length about 10 mm; a clear median band on
the carapace originating from between the AME and expanding just
anterior to and around the fovea, forming a diamond or triangle
shape; undulating sub-marginal bands; light marks just behind the
PLE (Fig. 1); and a patterned abdomen. Male H. grandis specimens
can be separated from H. helluo and A. georgicola by the short
laterally directed MA (compare Figs. 6 and 7), and from H. aspersa
males by the color pattern, and a weak or absent sclerite on the palea
(Figs. 2, 3). Female H. grandis can be separated from H. aspersa by a
narrower inverted T-shaped MS, which is no wider than one of the
epigynal hoods. They can be separated from H. helluo by a shorter
MS stalk and thicker MS base, and from A. georgicola by a less
curved atrium and thicker MS base which forms a gradual curve from
the stem to the base without depressions along the curve. The bilobed
terminal chamber of the spemathecae of H. grandis can be used to
separate it from female A. georgicola and H. aspersa.

Description, — Male (type): Total length 21.80 mm, carapace
length 11.40 mm, carapace width 8.76 mm. Pale median band
originating between the AME and abruptly widening just prior to
the fovea, then narrowing again posteriorly, creating a diamond
shape. Broad pale marginal bands extending from the PER
posteriorly, margins of bands often undulating. Light yellow stripe
just posterior to each of the PLE, just longer than the length of the
PLE, followed by a second light yellow stripe about the same length
just posterior to the first stripe. Chelicerae dark brown, endites dark
brown with lighter tips. Sternum brown, with a lighter median band.
Legs light brown except for tarsus, which is black, covered with hair-
like setae. Numerous macrosetae present. Abdomen with dark heart
mark followed by several chevrons, each ending with a light mark.
Venter light brown, covered with many scattered dark spots.
Cymbium with 14 macrosetae at tip. Embolus originating from the
distal area behind the palea and smoothly curving until terminating
on the retrolateral side below the end of the upper TA sickle. Palea
with a distally located, laterally directed weakly sclerotized area. MA

264

THE JOURNAL OF ARACHNOLOGY

Figure 8. — Collection localities of Hogiui grandis (Banks 1894).

small, triangular, located on the retrolateral side of the palp,
projecting laterally, but not beyond the cymbium border. TA double,
sickle-shaped.

Males (n = 8): Total length: 18.2-22.0 mm; carapace length:
8.90-12.33 mm; carapace width: 7.2-10.12 mm. Carapace median
band (Fig. 1) may terminate in a more triangular shape than
diamond. Second set of light marks just posterior to PLE may be
hard to see or absent. Chelicerae may appear black. Sternum various
degrees of brown from light to dark. Cymbium may have 10-20
thick macrosetae at the tip. Weakly sclerotized area of palea may
appear non-scelerotized (Fig. 3). Size of MA consistent (Fig. 2)
relative to the rest of the palp. TA lower sickle not always visible
(Fig. 3).

Females (n = 11 ): Total length: 18.2-23.6 mm; carapace length:
1 1.2-12.6 mm; carapace width: 8. 7-9. 8 mm; coloration same as male
but darker. Marginal bands of carapace thinner, light marks behind
PLE often lacking second light mark set. Legs also darker than male.
Females have an inverted T-shaped MS with a narrow stem no wider
than one of the epigynal hoods. The base of the MS is thick forming a
gradual curve from the stem to the base without depressions along the
curve (Fig. 4). The spermathecae have a kidney-shaped basal
chamber with a lateral lobe not extending anteriorly beyond the
basal chamber. The spermathecal stalk has a globular lateral lobe and
an oblong and often bilobed terminal chamber (Fig. 5).

Habitat and distribution. — Specimens of H. grandis have been
collected in the grasslands east of the Rocky Mountains in
Wyoming and Colorado and west into the San Luis Valley of
Colorado (Fig. 8). There is some discrepancy in the distribution of
the species. Chamberlin (1904) lists the locality of L. pernninda as

being in Kansas, but provided no further information. Worley &
Pickwell (1931), in their Spiders of Nebraska, had their lycosid
spiders identified by Martin Muma; examination by the author (JS,
pers. obs.) of all voucher specimens in the UNB museum found that
he had mistakenly identified specimens of H. lielluo (Walckenaer
1837) as H. ( Geolycosa) grandis. No specimens of H. grandis were
found in the voucher set; however, not all specimens listed by
Worley & Pickwell could be located. Worley & Pickwell also note
that H. ( Geolyco.sa) grandis is found only in the eastern half of the
state, which is peculiar for a species with a type locality of Fort
Collins, Colorado.

More recently Dondale & Redner (1990) cite both H. helluo and H.
aspersa (Hentz 1844) as being found in Colorado and Wyoming;
however, upon review of specimens from both states held in the
DMNS collection, no H. aspersa have been found, and only a single
H. helluo male was collected in Colorado, indicating that ranges for
both species occur further east. The DMNS collection consists of
specimens collected since 1998, as well as the private collection of Bea
Vogel; therefore, the specimens may not have been available to
Dondale & Redner. These species can be confused with H. grandis,
which may explain the discrepancy.

Spiders have been seen using burrows and roaming at night in
search of prey, a behavior similar to Hogna carolinensis (Walckenaer
1805) (JS, pers. obs.). They also use perches when kept in captivity,
another behavior similar to H. carolinensis. Burrow depth is variable
from 5-10 cm, with adults preferring shallower burrows. Burrow
shape is straight with a small cavity at the base in deeper burrows,
and more lateral in shallower burrows. Burrow turrets were not
observed.

SLOWIK & CUSHING— REDESCRIPTION OF GEOLYCOSA GRANDIS

265

Taxonomic note. — In researching this paper, we examined Schef-
fer’s (1905) description and illustrations of Lycosa permkma, which
indicate that the spider is not a Lycosa, but rather belongs in the
genus Arctosa. The type specimen of L. permkma could not be located
at the KU museum or USNM as indicated in this paper. The type
specimen could also not be located at the AMNH or MCZ collections
and is assumed lost. Based on the illustrations of the dorsum and
epigynum showing the median septum gradually widening posteriorly
and appearing covered with hairs, as well as the description of the
dorsum of the carapace and abdominal pattern with radiating lines
and spots rather than a dark heart mark found in Hogna, L. permiana
may be synonymous with Arctosa emertoni Gertsch 1934 or A.
rubicmda (Keyserling 1877) which both occur in Kansas. Because of
the uncertainty about which Arctosa species it may be, we declare it
nomen dubium.

ACKNOWLEDGMENTS

The authors thank Charles Dondale for helpful comments on an
earlier draft of this manuscript. This study was supported by NSF
grant DBI-0346378 awarded to P.E.C.

LITERATURE CITED

Banks, N. 1894. On the Lycosidae of Colorado. Journal of the New
York Entomological Society 2:49-52.

Chamberlin, R.V. 1904. Three new Lycosidae. Canadian Entomol-
ogist 36:286-288.

Chamberlin, R.V. 1908. Revision of North American spiders of the
family Lycosidae. Proceedings of the Academy of Natural Sciences
of Philadelphia 60:158-318.

Dondale, C.D. & J.H. Redner. 1990. The Insects and Arachnids of
Canada, Part 17. The Wolf Spiders, Nursery web Spiders, and Lynx
Spiders of Canada and Alaska, Araneae: Lycosidae, Pisauridae,
and Oxyopidae. Research Branch, Agriculture Canada, Publishers
1856:1-383.

Platnick, N. 2007. The World Spider Catalog, Version 7. American
Museum of Natural History, New York. Online at http://research.
amnh.org/entomology/spiders/catalog/index.html.

Roewer, C.F. 1955. Katalog der Araneae von 1758 bis 1940, bzw.
1954. Bruxelles 2:1-1751.

Scheffer, T.H. 1905. Additions to the list of Kansas spiders.
Industrialist 31:435-444. (reprinted in Transactions of the Kansas
Academy of Science 20:121-130, 1906).

Wallace, H.K. 1942. A revision of the burrowing spiders of the genus
Geolycoss.. American Midland Naturalist 27:1-62.

Worley, L.G. & G.B. Pickwell. 1931. The Spiders of Nebraska.
Studies from the Department of Zoology. The University of
Nebraska, Lincoln, Nebraska 27:1-129.

Manuscript received 24 September 2008, revised 13 May 2009.

2009. The Journal of Arachnology 37:266-271

Nephilid spider eunuch phenomenon induced by female or rival male aggressiveness

Matjaz Kuntner' -, Ingi Agnarsson* --'* and Matjaz Gregoric'; 'Institute of Biology, Scientific Research Centre, Slovenian
Academy of Sciences and Arts, Novi trg 2, P. O. Box 306, SI-1001 Ljubljana, Slovenia. E-mail: kuntner@gmail.com;
-Department of Entomology, National Museum of Natural History, Smithsonian Institution, NHB-105, PO Box 37012,
Washington, DC 20013-7012, USA; ■''Department of Biology and Integrated Bioscience Program, University of Akron,
Akron, OH 44325-3908, USA

Abstract. Plugging of female genitals via male sexual mutilation is a common sexual repertoire in some nephilid spiders
{Herennia, Nephila, Nephilengys), but the behavioral pathways leading to emasculation are poorly understood. Recent
work suggests that copulating Herennia males damage their reproductive organs during copulation and then voluntarily,
and stereotypically, remove their pedipalps to become eunuchs. Presumably, such emasculation increases agility allowing
the male to better fend off rival males. However, through our observation of male antagonism in Nephilengys horhonica
(Vinson 1863) in La Reunion (Indian Ocean), we discovered that genital severance involving the entire male palp is induced
by a rival eunuch. Additionally, laboratory matings of the same species from Mayotte provide the first observations of
female sexual cannibalism in this species, one such forceful copulation termination leading to emasculation of the entire
palp. These novel behaviors suggest that mate plugging and the eunuch phenomenon are more plastic repertoires than
hitherto thought, and thus our observations add to possible pathways leading to them. Based on our examination of 791
samples of Nephilengys spp. from museum collections and of a freshly collected representative sample of N. horhonica, we
conclude that i) palpal severance is common (50% of males from the wild were eunuchs lacking both palps), but ii) the
females (or perhaps subsequent males) must possess a mechanism for removing severed palps from the epigyna (none had a
whole palpal bulb), leaving behind only partial, embolic plugs, and hi) the disparity between male palpal damage (50%) and
visible mating plugs in females (21%) merits further research as the relative numbers of severed males and plugged females
can offer insight into which sex may have the upper hand in an evolutionary arms race.

Keywords: Sexual contlict, emasculation, plugging, behavioral plasticity, Nephilidae, Nephilengys horhonica

Mate plugging via sexual mutilation and eunuchs, males
with severed sperm- transferring organs (palps), are common
sexual repertoires and outcomes in males of the nephilid
spiders Herennia and Nephilengys (Robinson & Robinson
1978; Robinson & Lubin 1979; Kuntner 2005, 2007). The
sparse behavioral data suggest that copulating males typically
first produce a mating plug consisting of the distal two palpal
sclerites stuck in the female reproductive apparatus (epigy-
num), thereby preventing additional access to the used female
copulatory opening by rival males. Then the males voluntarily
remove their copulatory organ and continue in a sterile, and
more agile state, to mate-guard the female from competing
suitors (Robinson & Robinson 1980). Such a stereotyped
pathway leading to the eunuch phenomenon was partially
supported by our recent work, which i) establishes the
phylogenetic pattern of these behaviors as homologous in
nephilids (Kuntner et al. 2008, 2009a); all nephilids with
eunuchs also produce mating plugs; ii) provides the evidence
that severed organs in Herennia indeed function as mating
plugs (Kuntner et al. 2009b); and iii) confirms that sexual
mutilation and palpal severance (or eunuch behavior) in
Herennia is voluntary as previously reported for Nephilengys
(Robinson & Robinson 1980). Plugging in Herennia, however,
is more likely when the female shows aggression towards the
male during copulation (Kuntner et al. 2009b).

However, our new observations reveal behavioral plasticity
that complicates the apparently clear pattern of voluntary,
post-copulatory self-emasculation. We here report the first
observations of plugging that involve immediate, forced whole
bulb loss in a nephilid. One such case was induced by male-
male antagonistic behavior. We observed eunuch male mate-

guarding and male-male antagonism in Nephilengys horhonica
(Vinson 1863) (Nephilidae) in La Reunion (Indian Ocean)
revealing that palpal severance may be forcefully induced by
the male (eunuch) rival, and that such plugging may involve
the entire male palp. Furthermore, laboratory mating
observations of N. horhonica from Mayotte (Indian Ocean)
revealed that female sexual aggressiveness, which may result in
sexual cannibalism, may also induce male whole palp
severance. Although based only on a few observations, we
report on them because novel behaviors were noted that
elucidate the ethological repertoires leading to plugging and
the eunuch phenomenon, and because behavioral observations
on Nephilengys sexual behavior are so scarce. To investigate
the prevalence of palpal severance and the apparent absence of
whole palpal plugs in preserved museum material, we here also
report examinations of 791 samples of Nephilengys spp. from
museum collections and of a freshly collected representative
sample of N. horhonica from islands in the Indian Ocean.

METHODS

Field observations. — We monitored male-male and male-
female interactions in Nephilengys horhonica on the Indian
Ocean island of La Reunion (France), in a forest at Colorado-
La Montagne (20°54'23.5"S; 55°25'29.4"E; 680 m elev.) on 12
April 2008. Our daytime, non-manipulative observations
utilized digital photography (Canon EOS 20D with a 50 mm
macro lens and macro flash) and voice recording.

Laboratory observations. — We collected juvenile and adult
Nephilengys horhonica in their natural environment on the
Indian Ocean island of Mayotte (France) at Plage Tahiti
(12°51'49.0"S; 45°06'39.4"E; 1 m elev.) on 8 April 2008 and

266

KUNTNER ET AL.— NEPHILID SPIDER EUNUCHS

267

transported them live in foam stopper vials to the Ljubljana
laboratory on 10 April 2008. We kept larger spiders in 60 X 60
X 10 cm perspex frames and smaller spiders in smaller plastic
containers, at a constant 27° C temperature, 70% humidity,
and 12:12 h photoperiod. We misted the webs daily and fed
the adult and subadult spiders houseflies and crickets twice a
week, and the younger spiders fruit flies or small crickets daily.
We used final molting to adulthood as evidence for virginity,
but also mated individuals of unknown mating histories (i.e.,
those collected as adults). Following the experimental mating
protocols of Kuntner et al. (2009b), we placed in contact each
experimental pair for up to four hours. Before experiments
took place, we moved and opened the frames housing the
females and kept them undisturbed for the first 60 min,
followed by the transfer of a male onto the female web using a
fine paintbrush. We used video (Sony HDV 1080i) and voice
recording to document behaviors.

Morphological examination. — All adult spiders from mating
experiments were examined for genital damage under a Leica
MZ16 stereomicroscope using KOH maceration, methyl
salicilate treatment, and spermathecal clipping (see Kuntner
et al. 2009b). Spermathecal clipping reveals male parts
(emboli) hidden within the female genitals (spermathecae
and ducts) and thus provides evidence of male plugging.
Microscope imaging was done using a Leica DFC420C digital
camera. Two of the three spiders (eunuch male and female)
from La Reunion and all those that mated from Mayotte are
available as vouchers and will be deposited in the collections
of the Smithsonian Institution’s National Museum of Natural
History.

In order to assess the prevalence of eunuchs and external
plugs in nature, we examined a representative sample of newly
collected N. borhonica from Madagascar and neighboring
Indian Ocean Islands, including La Reunion. External
examination of these samples with a Leica MZ16 stereomi-
croscope would have easily revealed whole palpal plugs in
females and palpal loss in males.

RESULTS

Field observations. — A web contained the host female and
two males, of which the larger one was a full eunuch
(individual that had severed both palps) and the smaller one
was intact with both palps (Fig. lA). Our observation started
when the female was residing in her retreat above the hub, the
full eunuch (hereafter “eunuch”) resting 10 cm above her and
the palped male (hereafter “male”) 10 cm below her (Fig. 1 A).
Showing no apparent courtship behavior, the male ap-
proached the female, entered her retreat, climbed on her
venter, and attempted copulation. At the same time, the
eunuch aggressively pursued the wooing male into the retreat,
forcing him immediately to jump off the female and out of the
retreat into a free fall (Fig. IB), stopped by his signal line that
was attached to the female web. The eunuch dimed on the
female, and while remaining on her, she moved out of the
retreat (yet showed no aggressiveness to either male), and
remained around her web hub. The male returned on his signal
line, re-approached the female, climbed on her venter, and
resumed copulatory attempts despite continued eunuch
aggression. One copulation attempt was apparently successful
resulting in left palp insertion (Fig. IC-arrow). However,

eunuch antagonism (not female aggressiveness) forced the
male to flee (Fig. ID), this time inducing the copulating male
to leave behind his entire left palpal bulb, stuck in the female
epigynal opening (Fig. ID-inset). The severed palp broke in
the typical position between the tibia and the tarsus (Fig. lE-
inset). The male, now half-eunuch, returned facing the
female’s front (she remained unaggressive) and climbed back
onto her venter in a new copulation attempt, this time with his
remaining right palp (Fig. IE, F). Continued eunuch attacks
(Fig. IF) again forced the male to flee, but only to yet again
approach the female, which had returned to her retreat
(Fig. IG). Walking by the prosoma and chelicerae of the
unaggressive female, the male moved towards the eunuch, now
protecting the female epigynum on her venter (Fig. IH). The
eunuch forced the male to withdraw again, this time onto the
female dorsum (Fig. II), and the eunuch remained by the
female epigynum (plugged by the male severed bulb. Fig. II-
arrow). Apparently irritated by the two males on her
abdomen, the female moved abruptly out of her retreat and
the eunuch attacked the male by aggressively biting with his
chelicerae (Fig. IJ shows them still widely open). The male
jumped again, and the eunuch resumed his mate-guarding,
keeping close to the female abdomen in her retreat (Fig. IK).

Laboratory observations. — The first two male-female en-
counters on 15 May 2008 did not result in copulation. Male 1
(virgin) attempted to mate repeatedly using both palps with
female 1 (unknown history), but failed. As predicted from this
episode, subsequent examination revealed an embolic plug in
each female opening. Male 2 (unknown history) attempted to
copulate with female 2 (previously mated as witnessed by a
clutch of offspring), but his approach resulted in female’s
aggression immediately upon contact and in the first recorded
case of sexual cannibalism, where she consumed her suitor
(video available at www.nephilidae.com). Again, subsequent
examination revealed an embolic plug in each epigynal
opening.

On 28 May 2008 at 10:07 h, male 1 (still virgin after
unsuccessful copulation attempts) and female 3 (virgin, one
week after final molt) were placed together. The male, being
placed about 15 cm diagonally below the female, started
walking away from her for several seconds, then turned
around and headed towards the female at the top of the web.
He stopped 10 cm before the female and cleaned his legs in the
order 1 -right, 1-left, 2-right, 2-left, 4-left, 4-right, then
continued towards the female and paused for 68 s about
5 cm away from her. He continued towards the female shaking
his whole body twice and pausing again with his first legs
0.5 cm away from the female’s first legs. The contact (video
available at www.nephilidae.com) occurred after 10 min of
idleness. The male walked onto the female’s venter, tapped
and probed briefly with his palps on the epigynum, and
inserted his right palp into the right genital opening (ipsilateral
insertion), followed by the male's lateral body twist (abdomen
to the left). After only a brief copulation (about 4 s, at 10:20
h), the female aggressively terminated it by manipulating the
male with her first and second legs, grabbing him in her
chelicerae, and holding on and consuming him. During
seizure, the male’s right palp broke off and remained stuck
in the female’s genital opening (Fig. 2A). Her devouring of the
male continued (video available at www.nephilidae.com) until

268

THE JOURNAL OF ARACHNOLOGY

Figure 1 . — Male-female and (eunuch) male-male interactions of Nephilengys horbonica from La Reunion in chronological order (see text). The
male (m) with both palps approached the female (f) (A), but the palplcss eunuch (e) repeatedly chased him off her (B-I) and, ultimately,
successfully off her web (J-K). A male-male fight during brief copulation (C) resulted in a forceful severance of the male copulating palpal bulb,
that remained stuck in the female genital opening (D-inset, arrow, I-arrow) and rendered the sexually active male a half-eunuch (E-inset,
arrow, H-arrow).

we interrupted it to secure the male voucher needed for
morphological examination. The male lost his right palp at the
tibia-tarsus joint (Fig 2B-arrow), as is common in self-
castrated eunuch nephilids (Kuntner 2005, 2007).

Morphological examination. — We have previously examined
791 samples containing preserved Nephilengys specimens
(Table 1, specimen data examined by Kuntner 2007 available
at www.nephilidae.com). In Nephilengys, all four species are

KUNTNER ET AL.— NEPHILID SPIDER EUNUCHS

269

Figure 2. — Outcome of the laboratory mating experiment with Nephilengys borbonica from Mayotte. Female forcefully terminated copulation
resulting in whole bulb plug in epigynal opening (A) and male becoming eunuch (B) just before being cannibalized by the female. Arrow in (B)
indicates the point of palpal breakage.

uniform in gross morphology and behavior, including
common eunuchs (Kuntner 2007). However, none of these
samples contained females plugged with whole palps.

Table 2 lists the N. borbonica specimens examined external-
ly within the representative sample from the 2008 collected
material (« = 33 females, 30 males; specimens from mating
trials and samples containing only juveniles are omitted).
Again, eunuchs were common (50% of males were both palp
eunuchs), but while 21% of females had mating plugs in the
form of externally visible embolic parts in their copulatory
ducts, none of the females had whole palps in their epigyna.

DISCUSSION

The nephilid eunuch phenomenon is not homologous to the
voluntary emasculation by theridiid spiders Echinotheridion
and Tidarren (Agnarsson 2006; Kuntner et al. 2008), where the
subadult males remove one of their palps before seeking
mates. However, increased male agility may explain both
phenomena. In these tiny theridiid males, also known for
whole palpal plugs and fierce male-male competition (Miller
2007), two large palps reduce the mobility of the male and
subsequent emasculation thus facilitates mate searching

(Ramos et al. 2004). As sexual cannibalism characterizes these
theridiids, males do not have the opportunity to defend mated
females. Nephilid males, on the other hand, always start their
sexually active post-molting period with a fully functional pair
of palps, and become eunuchs after copulation. Hence, unlike
their theridiid counterparts, their emasculation serves for post-
copulatory agility. In the theridiids, emasculation is highly
stereotypical (Knoflach & Harten 2000, 2001). In contrast, our
findings here suggest that emasculation does not always follow
the same behavioral pattern in nephilids, as previously
thought.

Eunuchs are known for post-copulatory mate-guarding and
aggressiveness towards rival males (Robinson & Robinson
1978). Our own work on Hereiinia (Kuntner et al. 2009b) has
recently confirmed that the eunuch phenomenon typically
entails initial sclerite damaging during copulation, followed by
voluntary post-copulation palpal removal, corroborating only
two previous observations on Nephilengys (Robinson &
Robinson 1980). However, even if the males of Herennia
removed their palps themselves after initial mutilation,
mutilation itself (resulting in plugging) was more likely when
the female was aggressive towards the male during copulation

Table 1. — Numbers of samples containing Nephilengys spp. examined morphologically by Kuntner (2007; specimen and depository data
available at www.nephilidae.com).

Species

Samples

containing female(s)

Samples

containing immature(s)

Samples

containing male(s)

Total

N. borbonica

86

10

49

145

N. cruentata

360

49

46

455

N. malabarensis

99

8

45

152

N. papuana

24

5

10

39

Total

569

72

150

791

270

THE JOURNAL OF ARACHNOLOGY

Table 2. — Representative sample of female (0 and male (m) N.
borhonica collected in nature and examined externally for palpal
severance and plugs. Left/right scores are for embolic plug counts in
females per opening (0 = no plug; 1 = one plug) and damaged palp in
male (u = undamaged; p = palpal severance). More detailed
specimen data available from the authors. Note absence of simple
embolic severance in males and no half-eunuchs (see Discussion).

Sex

Left

Right

Number of
specimens

f

0

0

26

f

1

0

3

f

0

1

2

f

1

I

2

m

P

P

15

m

u

u

15

(Kuntner et al. 2009b). This suggests the possibility that
aggressiveness could play a role in inducing palpal severance
at various weak points. Although data are sparse, our mating
observations suggest that sexual aggression/cannibalism may
be common in N. borbonica (two out of three mating
sequences observed) and, in one case, induced the male to
release and leave behind a whole palpal plug in the female
opening. Furthermore, the newly observed behavioral pattern
of rival-male-induced palpal severance suggests that intraspe-
cific aggression from the female or rival males may induce
whole palpal plugging.

Most available observations on nephilid sexual mutilation
rely on very few data points {Nephila being an exception, see
Schneider et al. 2001, 2005; Fromhage & Schneider 2006) and
the sexual behavior of N. borbonica has not been studied in
detail (but see Kuntner 2007). However sparse the data, whole
bulb plugging has not been observed before. Certain behaviors
such as plugging and bulb loss can be deduced via
morphological examination (Kuntner et al. 2008, 2009b). We
have examined numerous samples of male and female
individuals of N. borbonica from its entire range and of
Nephilengys spp. in total (Tables 1 and 2). Presumably, these
examinations would have easily detected any evidence of
whole bulb-plugs, yet did not. Similarly, whole bulb-plugs are
unknown in Herennia or any other nephilid (Kuntner 2005,
2006). The data at hand, if interpreted uncritically, would thus
suggest that mating plugs in Nephilengys are usually formed by
the terminal two sclerites as in most other nephilids (Kuntner
2005, 2007; Kuntner et al. 2008). However, in light of our
study, we propose an alternative explanation: that whole palps
may be frequently severed and left in the female, but that they
subsequently break off at the standard point (the remaining
morphological evidence of females would then fail to detect
such involuntary palp loss). Because a palpal bulb consists of
the entire palpal tarsus sexually modified in adult males to
contain a number of interconnected sclerites and membranes
(Fig. 2A), it is thus a relatively large and cumbersome device
for the female to keep on her. Additionally, the sexual conflict
theory (Arnqvist & Rowe 2005) would predict that it is in the
female’s interest to resist such a male monopolization
mechanism (Kuntner et al. 2009a). We suggest that such
plugged females do remove whole palpal plugs (perhaps using
their legs or rubbing against the substrate), and that such
removal would then result in the common morphological

outcome: palpless (eunuch) males, and females plugged only
by the male’s two terminal sclerites. In fact, the second
breaking point between the two sclerites and the remainder of
the male palp may be another twist in the male-female conflict
story: the male’s response to female palp-plug removal. As our
preliminary data show, even the two-sclerite plugs seem
effective in preventing further matings.

Alternatively, subsequent males, in addition to the female,
may be able to remove previous whole palp plugs. Embolic
plug removal by the next male is known to occur sporadically
in Nephila fenestrata (J. Schneider pers. comm.), which is
phylogenetically only a few nodes away from Nephilengys
(Kuntner et al. 2008), although in that species plugs are
simpler and generally functional (Fromhage & Schneider
2006). A similar mechanism in N. borbonica would then
explain the second breaking point in the palpal bulb as
another result of sperm competition, in addition to the plugs
themselves, rather than a result of sexual conflict between the
male and the female. We find this mechanism possible, but
unlikely, because it would imply some percentage of females
that find no mate subsequent to being plugged, and thus
retain the whole palp plug. In our case, this percentage was
zero.

Examination of over 800 preserved Nephilengys specimens
shows high frequencies of female plugging (21%) and male
palpal severance (50%). Given that a considerable portion of
samples must represent unmated specimens, palpal severance
in mated males is likely universal. Our ongoing work
(Kuntner, unpublished data) suggests that external genital
examination in nephilids underestimates the number of
embolic mating plugs (broken emboli may be lodged deep in
spermathecae), and thus the real number of plugged females is
certainly higher than 21%. Further work is thus necessary to
determine if the percentage of plugged females equals the
percentage of severed males. Such work should lead to
interesting conclusions. If differences in frequencies are
confirmed, this would suggest that females are successful in
removing mating plugs so that even whole palpal sacrifice by
the male may not guarantee his mating success and/or female
monopolization. If, however, the numbers of severed males
and plugged females are approximately equal, this would
imply that male plugs are successful and that males may be
winning an evolutionary arms race.

It is worth noting, that the 15 N. borbonica eunuchs
collected in nature were all full eunuchs, meaning they lacked
both palps. Kuntner (2005, 2007) reported also half-eunuchs
(males with only one functional palp) in Nephilengys and
Herennia, and we have documented half-eunuchs of H.
multipimcta in nature (Kuntner et al. 2009b: fig. 2b-c). It
remains to be seen whether half-eunuchs are rarer in N
borbonica than in Herennia and other Nephilengys species.

In sum, the newly discovered behavior where sexual
mutilation is induced by male or female aggressive behavior
and where the mating plug consists of the whole palpal bulb,
adds to the documented plasticity in the pathways leading to
the nephilid eunuch phenomenon. We conclude that palpal
severance in N. borbonica is common (50% of males are
eunuchs) and that the females are likely able to remove
severed palps from their epigyna, resulting in embolic mating
plugs.

KUNTNER ET AL.— NEPHILID SPIDER EUNUCHS

271

ACKNOWLEDGMENTS

This is contribution number one resulting from the 2008
Indian Ocean expedition, funded by the Slovenian Research
Agency (grant Z I -9799-06 18=07 to I. Agnarsson) and the
National Science Foundation (grant DEB-0516038 to T.
Blackledge). We also acknowledge the funding by the
European Community Framework Programme (a Marie
Curie International Reintegration Grant MIRG-CT-2005
036536 to M. Kuntner). We thank Dominique Strasberg,
Sonia Ribes-Beaudemoulin, and Benoit Lequette for their help
with our research, and Jutta Schneider, Simona Kralj-Fiser,
Gail Stratton, Jeremy Miller and an anonymous reviewer for
comments.

LITERATURE CITED

Agnarsson, I. 2006. Phylogenetic placement of Echinotheridion
(Araneae: Theridiidae) - do male sexual organ removal, emascu-
lation, and sexual cannibalism in Echinotheridion and Tidarren
represent evolutionary replicas? Invertebrate Systematics 20:415-
429.

Arnqvist, G. & L. Rowe. 2005. Sexual Conflict. Princeton University
Press, Princeton, New Jersey.

Fromhage, L. & J.M. Schneider. 2006. Emasculation to plug up
females: the significance of pedipalp damage in Nephila fenestrata.
Behavioral Ecology 17:353-357.

Knoflach, B. & A. van Harten. 2000. Palpal loss, single palp
copulation and obligatory mate consumption in Tidarren cimeola-
tum (Tullgren, 1910) (Araneae, Theridiidae). Journal of Natural
History 34:1639-1659.

Knoflach, B. & A. van Harten. 2001. Tidarren argo sp. nov. (Araneae:
Theridiidae) and its exceptional copulatory behaviour: emascula-
tion, male palpal organ as a mating plug and sexual cannibalism.
Journal of Zoology 254:449M59.

Kuntner, M. 2005. A revision of Herennia (Araneae: Nephilidae:
Nephilinae), the Australasian ‘coin spiders’. Invertebrate System-
atics 19:391-436.

Kuntner, M. 2006. Phylogenetic systematics of the Gondwanan
nephilid spider lineage Clitaetrinae (Araneae, Nephilidae). Zoolo-
gica Scripta 35:19-62.

Kuntner, M. 2007. A monograph of Nephilengys, the pantropical
‘hermit spiders’ (Araneae, Nephilidae, Nephilinae). Systematic
Entomology 32:95-135.

Kuntner, M., J.A. Coddington & G. Hormiga. 2008. Phylogeny of
extant nephilid orb-weaving spiders (Araneae, Nephilidae): testing
morphological and ethological homologies. Cladistics 24:147-217.

Kuntner, M., J.A. Coddington & J.M. Schneider. 2009a. Intersexual
arms race? Genital coevolution in nephilid spiders (Araneae,
Nephilidae). Evolution 63:1451-1463.

Kuntner, M., S. Kralj-Fiser, J.M. Schneider & D. Li. 2009b. Mate
plugging via sexual mutilation in nephilid spiders: an evolutionary
hypothesis. Journal of Zoology 277:257-266.

Miller, J.A. 2007. Repeated evolution of male sacrifice behavior in
spiders correlated with genital mutilation. Evolution 61:1301-1315.

Ramos, M., D.J. Irschick & T.E. Christenson. 2004. Overcoming an
evolutionary conflict: Removal of a reproductive organ greatly
increases locomotor performance. Proceedings of the National
Academy of Sciences 101:4883M887.

Robinson, M.H. & Y.D. Lubin. 1979. Specialists and generalists: the
ecology and behavior of some web-building spiders from Papua
New Guinea. Pacific Insects 21:97-132.

Robinson, M.H. & B. Robinson. 1978. The evolution of courtship
systems in tropical araneid spiders. Symposia of the Zoological
Society of London 42:17-29.

Robinson, M.H. & B. Robinson. 1980. Comparative studies of the
courtship and mating behavior of tropical araneid spiders. Pacific
Insects Monographs 36:35-218.

Schneider, J.M., L. Fromhage & G. Uhl. 2005. Copulation patterns in
the golden orb-web spider Nephila tnadagascariensis. Journal of
Ethology 23:51-55.

Schneider, J.M., M.L. Thomas & M.A. Elgar. 2001. Ectomised
conductors in the golden orb-web spider, Nephila phtmipes
(Araneoidea): a male adaptation to sexual conflict? Behavioral
Ecology and Sociobiology 49:410M15.

Manuscript received 7 August 2008, revised 31 January 2009.

2009. The Journal of Arachnology 37:272-281

On the charitonovi species group of the spider genus Coelotes (Araneae: Amaurobiidae)

Xin-Ping Wang and Ming-Sheng Zhu: College of Life Sciences, Hebei University, Baoding 071002, Hebei, China. E-mail;
wang@amaurobiidae.com, xinping@ufl.edu

Abstract. Eight spider species of the genus Coelotes from Central Asia and the Near East currently assigned to the
charitonovi species group are described. Four species, Coelotes charitonovi Spassky 1939, C. juglandicola Ovtchinnikov
1984, C nenilini Ovtchinnikov 1999, and C. tiirkestanicus Ovtchinnikov 1999 are previously known from both sexes, and
four others, C. caiiclatus de Blauwe 1973, C. arganoi Brignoli 1978, C. coenohita Brignoli 1978, and C vignai Brignoli 1978
are known only from females. The dorsal views of the epigynum of C. juglandicola, C. caiidatiis, and C. vignai are illustrated
and described for the first time.

Keywords: Central Asia, Near East, Coelotinae, epigynum, taxonomy

Spassky (1939) described Coelotes charitonovi Spassky 1939
from Uzbekistan based only on females. The “group” remained
untouched until Charitonov (1969) described Agelena buchar-
ensis Charitonov 1969 based only on a male, which has proven
to be a junior synonym of C. charitonovi (Ovtsharenko & Fet
1980; Ovtchinnikov 1988, 1999). Coelotes charitonovi is wide-
spread in Central Asia (Turkmenistan, Uzbekistan, Tajikistan,
and Kyrgyzstan) (Ovtchinnikov 1999). de Blauwe (1973)
described C. candatus de Blauwe 1973 from Lebanon and
Brignoli ( 1978a) described C. arganoi Brignoli 1978, C. coenohita
Brignoli 1978, and C. vignai Brignoli 1978 from Turkey, all
based on females. Coelotes candatus is the only coelotine known
from the Near East. Although Brignoli (1978b) examined more
specimens of C. candatus from this region (2? from Col des
Cedres; 39 from Cedres de Bcharre, and 19 from Baskinta), the
presence of coelotines in the Near East has not been verified by
recent collections, and its occurrence in Lebanon needs
confirmation. Ovtchinnikov (1984) described another species,
C. juglandicola Ovtchinnikov 1984 from Kyrgyzstan, based on
specimens of both sexes; spermathecal morphology, however,
was neither illustrated nor described. Ovtchinnikov (1999)
described two additional species — C. nenilini Ovtchinnikov
1999 from Uzbekistan and C. turkestanicus Ovtchinnikov 1999
from Uzbekistan, Kyrgyzstan, and Kazakhstan and placed them
in the subgenus Brignoliolus, together with C. charitonovi, C.
candatus, and C. vignai. Wang (2002) included the species listed
above, together with C. arganoi and C. coenohita, in the
chctritonovi species-group of Coelotes based on the anterior,
closely-set epigynal teeth, but males were needed in order to
support this placement. Coelotes turkestanicus was also recorded
from the Orenburg Region of Russia, the northernmost limit of
its range (Esyunin et al. 2007).

The phylogenetic relationships of the charitonovi group
species with other coelotines are still unknown. They could be
closely related to the atropos species group from Europe and
Middle Asia in sharing a broad patellar apophysis, a reduced
lateral tibial apophysis, a short cymbial furrow, a rounded
median apophysis, and a prolaterally originating embolus, but
the charitonovi group contains species with the patellar
apophysis strongly modified and the epigynal teeth distinctly
long and closely set.

Within the charitonovi group, the species are geographically
distinctly separated. The four species from eastern Central

Asia (C. charitonivi, C.julandicola, C. nenilini, C. turkestanicus)
might be closely related in sharing the atrium that is situated
anteriorly at level of epigynal hoods or anterior epigynal
hoods, the elongated spermathecae, and other similarities in
the male palp. Unfortunately, data from males were not
available for the other four species from western Central Asia
and Near East. Their females share the medially or posteriorly
situated atrium, the long epigynal teeth, and the more or less
broad spermathecae.

In this paper, all species of the charitonovi group are re-
described and a key is provided to the species, with particular
focus on the genitalic structures. The dorsal views of the
epigynum of C. juglandicola, C. candatus, and C. vigncn are
illustrated and described for the first time. Coelotes nenilini, of
which specimens are not available, is described based on the
illustrations of Ovtchinnikov (1999).

METHODS

All measurements are in millimeters. Scale lines are 0.2 mm.
Terminology used in the text and figures follows Wang (2002).
The distribution map was generated using GIS ArcView
software, and the specimen files of the species studied can be
downloaded from Wang (2009). Due to the limitation of
available specimens from this region, this study is based
mainly on the examination of type specimens, which were
loaned from the following museums: AMNH = American
Museum of Natural History, New York (N.I. Platnick);
MHNG = Museum d’histoire naturelle de Geneve, Switzer-
land (B. Hauser); MNHN = Museum National d’Histoire
Naturelle, Paris (C. Rollard).

Abbreviations used in the text are: AAM-anterior atrial
margin; AME-anterior median eyes; ALE-anterior lateral
eyes; AS-atrial septum; C-conductor; CD-copulatory duct;
CDA-conductor dorsal apophysis; CL-conductor basal la-
mella; CY-cymbial furrow; E-embolus; EB-embolic base;
EH-epigynal hood; ET-epigynal tooth; FD-fertilization duct;
LAM-lateral atrial margin; LTA-lateral tibial apophysis;
MA-median apophysis; PA-patellar apophysis; PAM-poste-
rior atrial margin; PLE-posterior lateral eyes; PME-posterior
median eyes; RTA-retrolateral tibial apophysis; S-spermathe-
cae; SB-spermathecal base; SS-spermathecal stalk; SH-
spermathecal head; ST-subtegulum; T-tegulum; TS-tegular
sclerite.

272

WANG & ZnU—COELOTES CHARITONOV! GROm (COELOTINAE)

273

1 2 3

Figures 1-3. — Coelotes arganoi Brignoli 1978, female holotype from Erzincan, Sakaltutan gecidi, Turkey. 1. Epigynum, ventral view. 2.
Epigynum, dorsal view. 3. Eyes, view between front and dorsal.

SYSTEMATICS

Family Amaurobiidae Thorell 1870
Subfamily Coelotinae F.O. Pickard-Cambridge 1893
Genus Coelotes Blackwall 1841
The charitonovi Group

Diagnosis. — Females with anteriorly situated, closely set
epigynal teeth; males with strongly modified, broad patellar
apophysis, reduced lateral tibial apophysis, short conductor,

round, spoon-shaped median apophysis, and prolaterally
originating embolus.

Distribution. — Central Asia (Turkey, Kazakhstan, Kyrgyz-
stan, Tajikistan, Turkmenistan, Uzbekistan), Near East
(Lebanon), Russia (Orenburg Region) (Fig. 37).

Composition. — Eight species: Coelotes arganoi Brignoli
1978; C caudatus de Blauwe 1973; C. charitonovi Spassky
1939; C. coenobita Brignoli 1978; C. jugkmdkola Owlchlmmkow
1984; C. Ovtchinnikov 1999; C turkestanicus Ovichmm-

kov 1999; C. vignai Brignoli 1978.

KEY TO SPECIES OF THE CHARITONOVI GROUP

1. Male (those of C. caudatus, C. arganoi, C. coenobita and C. vignai unknown) 2

Female 5

2. Median apophysis large, distinctly separated from embolic base; embolus thread originating at level of base of tegular sclerite

(Figs. 21, 22) juglandicola

Median apophysis small, close to embolic base; embolus thread originating at level near middle of tegular sclerite (Figs. 9, 29) 3

3. Conductor dorsal apophysis length subequal to conductor (Fig. 9) charitonovi

Conductor dorsal apophysis longer than conductor (Fig. 29) 4

4. Conductor dorsal apophysis at least twice the size of conductor (Ovtchinnikov 1999:figs. 41, 42) nenilini

Conductor dorsal apophysis slightly larger than conductor (Fig. 29) turkestanicus

5. Atrium situated at level of epigynal hoods (Fig. 26) or anterior of epigynal hoods (Figs. 7, 18) . 6

Atrium situated at level posterior to epigynal hoods (Figs. 1, 4, 15, 34) 9

6. Epigynal teeth contiguous (Fig. 26) 7

Epigynal teeth distinctly separated (Figs. 7, 18) 8

7. Spermathecae abruptly converging, and then slightly diverging anteriorly (Fig. 27) turkestanicus

Spermathecae gradually converging (Ovtchinnikov 1999:rigs. 43, 44) nenilini

8. Epigynal teeth with bases separated by at least 3^ times their width; atrium semi-circular; spermathecal heads extend toward

the spermathecal bases (Figs. 18, 19) juglandicola

Epigynal teeth with bases separated by their width; atrium vase-shaped; spermathecal heads extend toward distal spermathecae
(Figs. 7, 8) charitonovi

9. Atrium with distinct anterior margin; epigynal teeth arising from a transversely extending furrow (Figs. 15, 34) 10

Atrium without anterior margin; epigynal teeth arising from a smooth surface (Figs. 1,4) 11

10. Atrium without distinct septum, with anteriorly protruding posterior margin (Figs. 15, 16) coenobita

Atrium with distinct septum (Figs. 34, 35) ...................................... vignai

11. Epigynal teeth relatively short, extend posteriorly less than halfway to epigastric furrow; spermathecae not convergent

anteriorly (Figs. 1,2) arganoi

Epigynal teeth relatively long, extend posteriorly more than halfway to epigastric furrow, spermathecae distinctly convergent
anteriorly (Figs. 4, 5) caudatus

274

THE JOURNAL OF ARACHNOLOGY

Figures 4—6. — Coelotes cuudutus de Blauwe 1973, female holotype
Eyes, view between front and dorsal.

from Lebanon. 4. Epigynum, ventral view. 5. Epigynum, dorsal view. 6.

Coelotes arganoi Brignoli 1978
Figs. 1-3, 37

Coelotes arganoi Brignoli 1978a:533, Figs. 132, 136 (female

holotype, in MHNG, examined).

Material examined. — TURKEY: ? holotype, Erzincan,
Sakaltutan gecidi, 12 June 1973, R. Argano, L. Boitani & V.
Cottarelli (MHNG).

Diagnosis. — Eemales similar to C. caudatus in having
anteriorly arising, basally contiguous, distally divergent epigynal
teeth, distinct anterior margin of atrium, and broad spermathe-
cae, but can be distinguished by relatively short epigynal teeth
that extend less than halfway to epigastric furrow and
spermathecae that extend parallel anteriorly (Figs. 1, 2).

Description. — See Brignoli (1978a) for more somatic de-
scriptions.

Female: Medium-sized coelotine, total length 8.65. AME
smallest, PME slightly larger than AME, lateral eyes subequal in
size, slightly larger than PME; AME separated from each other
by slightly less than their diameter, from ALE by about half an
AME diameter; PME separated from each other by their
diameter, from PLE by almost twice a PME diameter (Fig. 3).
Chelicera with 3 promarginal and 3 retromarginal teeth.
Epigynal teeth long, slender, arising from slightly anterior to
atrium, with contiguous bases and widely separated apices;
atrium large, wider than long, situated anteriorly, separated
from epigastric furrow by more than its length, with distinct,
continuous, arch-shaped anterior margin; copulatory ducts
broad, originating anteriorly; spemiathecae broad, slightly
extended anteriorly, widely separated by about their width;
spermathecal heads arising distally on spermathecae (Figs. 1, 2).

Male: Unknown.

Distribution. — Turkey (Fig. 37).

Coelotes caudatus de Blauwe 1973
Figs. 4-6, 37

Coelotes caudatus de Blauwe 1973:31, fig. 29 (female holotype,

in MNHN, examined). -Brignoli 1978b:207, fig. 5.

Material examined. — LEBANON: ? holotype, Liban, E.
Simon (MNHN B 2011, 1.037).

Diagnosis. — Eemales similar to C. arganoi in having the
anteriorly arising, basally contiguous, distally divergent epigynal
teeth, distinct anterior atrial margin, and broad spermathecae
but can be distinguished by relatively long epigynal teeth that
extend more than halfway to epigastric furrow and spermathe-
cae that distinctly converge anteriorly (Figs. 4-5).

Description. — Described by de Blauwe (1973) but dorsal
view of epigynum was neither illustrated nor described.

Female: Medium-sized coelotine, total length 8.70. Eyes
subequal in size, or with ALE slightly larger; AME separated
from each other by less than their diameter, from ALE by
about AME diameter; PME separated from each other by
about 1.5 times their diameter, from PLE by almost twice a
PME diameter (Eig. 6). Chelicera with 3 promarginal and 3
retromarginal teeth. Epigynal teeth long, slender, arising
anterior of anterior atrial margin, with slightly separated
bases and widely separated apices; atrium large, wider than
long, situated anteriorly, separated from epigastric furrow by
its length, with distinct, continuous, arch-shaped anterior
margin; copulatory ducts broad, originating anteriorly;
spermathecae broad, with bases separated by their width,
slightly extending and converging anteriorly; spermathecal
heads arising distally on spermathecae (Figs. 4, 5).

Male: Unknown.

Distribution. — Turkey (Fig. 37).

Coelotes charitonovi Spassky 1939
Figs. 7-14, 37

Coelotes charitonovi Spassky 1939:141, fig. 4 (types from
Uzbekistan, depository unknown, not examined). -Ovt-
sharenko and Fet 1980:446; Ovtchinnikov 1999:74, figs. 34,
35; Wang 2002:48, figs. 113-126.

Agelena bucharensis Charitonov 1969:81, fig. 3.

Coelotes hucharensis: Ovtchinnikov 1988:141.

Material examined. — TAJIKISTAN: l?lcJ, near Khovaling
town, 1000 m, 7 November 1987, S. Ovtchinnikov (AMNH-
donation of S. Ovtchnnikov).

Diagnosis. — This species is similar to C. juglandicola in
having the distinctly separated epigynal teeth, but females can

WANG & ZWJ—COELOTES CHARITONOVI GROUP (COELOTINAE)

275

7

8

CF

Figures 7-10. — Coelotes charitomvi Spassky 1939, female and male from Khovaling town, Tadzhikistan. 7. Epigynum, ventral view. 8.
Epigynum, dorsal view. 9. Palp, ventral view. 10. Palp, retrolateral view.

be distinguished by the closely situated epigynal teeth
(separated by about their width) and the anteriorly extended
spermathecal heads; males by the small conductor (about the
size of the conductor dorsal apophysis) and the small,
posteriorly situated median apophysis (Figs. 7-12).

Description. — Described by Spassky (1939) but the dorsal
view of the epigynum was neither illustrated nor described.

Female: Medium-sized coelotine. ALE largest, other eyes
subequal in size, slightly smaller than ALE; AME separated
from each other by 2/3 of their diameter, from ALE by about
half of AME diameter; posterior eyes equally separated by
about 1.5 times PME diameter (Fig. 13). Chelicera with 5

promarginal and 5 retromarginal teeth. Epigynal teeth short,
broad, arising from anterior atrial margin, slightly separated
by their width; epigynal hoods situated medially; atrium
situated anteriorly, with indistinct anterior margin and
distinct, vase-shaped lateral margins that extend posteriorly
to epigastric furrow; copulatory ducts small, originating
anteriorly; spermathecae broad, extend anteriorly, slightly
separated; spermathecal heads large, arising distally on
spermathecae (Figs. 7, 8).

Male: Medium sized coelotine. Eyes and chelicerae similar
to female. Patellal apophysis broad, dorsally concave; RTA
more than half the tibial length, with distinctly extending

276

THE JOURNAL OF ARACHNOLOGY

11

Figures 1 1-14. — Coelotes charitonovi Spassky 1939, female and male from Khovaling town, Tadzhikistan. 1 1, Palp, prolateral view. 12. Palp,
retrolateral view. 13. Female, eyes, front view. 14. Female, chelicera, ventral view.

distal end; lateral tibial apophysis absent; cymbial furrow less
than 1/3 of cymbial length; conductor short, basal lamella
indistinct, dorsal apophysis about the same size as conductor;
median apophysis small, spoon-shaped, without free standing
anterior edge, arising posteriorly; embolus short, filiform,
prolateral in origin, thread originating near middle of tegular
sclerite (Figs. 9-12).

Distribution. — Tajikistan, Kyrgyzstan, Tajikistan, Turkme-
nistan, and Uzbekistan (Ovtchinnikov 1999) (Fig. 37).

Coelotes coenobita Brignoli 1978
Figs. 15-17, 37

Coelotes coenobita Brignoli 1978a:533, figs. 135, 137 (female
holotype and paratypes, in MHNG, examined).

Figures 15 17. — Coelotes coenobita Brignoli 1978, female holotype from Trabzon, Sumela (Macka), Turkey. 15. Epigynum, ventral view; 16.
Epigynum, dorsal view; 17. eyes, view between front and dorsal.

WANG & Z\\\J—COELOTES CHARITONOVI GROUP (COELOTINAE)

277

Figures 18-25. — Coelotes jugkmdicola Ovtchinnikov 1984, male and female from Fergansky Mt., Kyrgyzstan. 18. Epigynum, ventral view. 19.
Epigynum, dorsal view. 20. Palp, prolateral view. 21. Palp, ventral view. 22, 23. Palp, retrolateral view. 24. Eyes, view between front and dorsal.
25. Chelicera, ventral view.

Material examined, — TURKEY: ? holotype, Trabzon, Sumela
(Macka), June 16, 1968, P. Brignoli (MHNG); 2? paratypes,
Trabzon, Sumela (Macka), 10-11 June 1969, P. Brignoli
(MHNG).

Diagnosis. — Females are similar to C. vignai in having
closely arising and extended epigynal teeth, posteriorly
situated atrium, and broad, round spermathecae but can be
distinguished by the absence of atrial septum, anteriorly
extended posterior atrial margin, and slightly separated
spermathecae (Figs. 15, 16).

Description. — See Brignoli (1978a) for more somatic de-
scriptions.

Female: Medium-sized coelotine, total length 7.50. AME
smallest, about half size of ALE, ALE largest, posterior eyes
subequal in size, slightly smaller than ALE; AME separated
from each other by about their diameter, from ALE by slightly
less than AME diameter; PME separated from each other by less
than their diameter, from PLE by more than a PME diameter
(Fig. 17). Chelicera with 3 promarginal and 3 retromarginal
teeth. Epigynal teeth long, slender, arising anteriorly from a

278

THE JOURNAL OF ARACHNOLOGY

28 29 30

Figures 26-33. — Coelotes twkestauicus Ovtchinnikov 1999, male and female from Kizghizsky Mt., Kyrgyzstan. 26. Epigynum, ventral view.
27. Epigynum, dorsal view. 28. Palp, prolateral view. 29. Palp, ventral view. 30, 31. Palp, retrolateral view. 32. Eyes, view between front and
dorsal; 33. Chelieera, ventral view.

transversely extending furrow, close together; atrium situated
posteriorly, with distinct, anteriorly convexing posterior margin;
copulatory ducts indistinct from dorsal view; spermathecae
broad, round, close together; spermathecal heads arising distally
on spermathecae, barely visible (Figs. 15, 16).

Male: Unknown.

Distribution. — Turkey (Fig. 37).

Coelotes juglandicola Ovtchinnikov 1984
Figs. 18-25, 37

Coelotes juglandicola Ovtchinnikov 1984:126, figs. 1-2 (types
not examined).

Material examined. — KYRGYZSTAN: Ml?, Fergansky
Mt., Baubashata Mts., Arslanbop, 17 July 1991, S. Ovtch-
innikov (AMNH-donation of S. Ovtchinnikov).

Diagnosis. — Males can be recognized by embolus with
thread originating near base of tegular sclerite, and females
by widely separated epigynal teeth and posteriorly extended
spermathecal heads (Figs. 18-23).

Description. — Described by Ovtchinnikov (1984) but the
dorsal view of epigynum was neither illustrated nor described.

Female: Medium-sized coelotine. ALE largest, slightly
larger than other eyes, which are subequal; anterior eyes
equally separated by slightly more than half of AME diameter;

WANG & ZmJ—COELOTES CHARITONOVI GROUP (COELOTINAE)

279

Figures 34-36. — Coelotes vignai Brignoli, 1978, female holotype from Bolu, Abant, Turkey. 34. Epigynum, ventral view. 35. Epigynum, dorsal
view. 36. Eyes, view between front and dorsal.

PEM separated from each other by their diameter, from PLE
by slightly less than 1.5 times PME diameter (Fig. 24).
Chelicerae with 3 promarginal and 3 retromarginal teeth
(Fig. 25). Epigynal teeth short, arising from anterior atrial
margin, widely separated by about three times their width;
epigynal hoods situated medially; atrium situated anteriorly,
with continuous, arch-shaped anterior margin; copulatory
ducts small, originating anteriorly; spermathecae broad,

extending and converging anteriorly, separated by their width
at bases and contiguous at apices; spermathecal heads small,
arising distally and extending toward spermathecal bases
(Figs. 18, 19).

Male: Medium-sized coelotine. Eyes and chelicerae similar
to female. Patellar apophysis broad, dorsally concave; RTA
more than half length of tibia, with distinctly extended distal
end; lateral tibial apophysis absent; cymbial furrow less than

Ukraine

Kazakhstan

Uzbekistan

;erbaijaf

Turkmenistan

Turkey

China

Pakistan

Lebani

® Coeloms arganoi
□ C. paudatus
charitonovi
S C coenohita
'"'70 C. juglandicola

(I'C

9 hiykestanictis

Saudi Ar^a

© C. v^iai

Figure 37. — Records of charitonovi group species (a question mark behind the symbol indicates the presence in the corresponding country, but
without detailed locality).

280

THE JOURNAL OF ARACHNOLOGY

1/3 of cymbial length; conductor short, basal lamella
indistinct, dorsal apophysis about as large as conductor;
median apophysis relatively large, spoon-shaped, without free
standing anterior edge, distinctly separated from embolic base;
embolus short, filiform, prolateral in origin, thread originating
near base of tegular sclerite (Figs. 20-23).

Distribution. — Kyrgyzstan (Fig. 37).

Coelotes nenilini Ovtchinnikov 1999
Fig. 37

Coelotes nenilini Ovtchinnikov 1999:77, figs. 40^4 (types not

examined).

Material examined. — None.

Diagnosis. — Similar to C turkestanicus in having the long,
contiguous epigynal teeth and the elongated spermathecae,
but can be distinguished by the gradually convergent
spermathecae in females and the large conductor dorsal
apophysis in males.

Description. — Based on the illustrations of Ovtchinnikov
(1999:figs. 41-44).

Female: Epigynal teeth long, contiguous, arising anteriorly;
epigynal hoods situated medially; atrium reduced to slit;
copulatory ducts indistinct; spermathecae extended and
slightly convergent anteriorly, separated by about their width;
spermathecal heads large, arising distally on spermathecae.

Male: Patellar apophysis broad, dorsally concave; RTA more
than half length of tibia, with distinctly extended distal end;
lateral tibial apophysis absent; cymbial furrow about 1/4 of
cymbial length; conductor short, basal lamella indistinct, dorsal
apophysis much larger than conductor; median apophysis small,
spoon-shaped, without free standing anterior edge, situated
close to embolic base; embolus short, filiform, prolateral in
origin, thread originating near middle of tegular sclerite.

Distribution. — Uzbekistan (Ugamsky Mt. Range) (Fig. 37).

Coelotes turkestanicus Ovtchnnikov 1999
Figs. 2^33, 37

Coelotes turkestanicus Ovtchinnikov 1999:75, figs. 36-39 (male

and female paratypes, in AMNH, examined). -Esyunin,

Tuneva & Farzalieva 2007:53, figs. 14—16, 25.

Material examined. — KYRGYZSTAN: lcJ2$ paratypes,
Kizghizsky Mt. R., Malinovka Canyon, 1700 m, 21 October
1984, S. Ovtchinnikov (AMNH, donation of S. Ovtchnnikov).

Diagnosis. — Similar to C. nenilini in having the long,
contiguous epigynal teeth and elongated spermathecae, but
can be distinguished by abruptly convergent spermathecae in
female and male conductor dorsal apophysis slightly larger
than conductor (Figs. 26-31).

Description. — See Ovtchinnikov (1999) for more somatic
descriptions.

Female: Large-sized coelotine, total length 12.60. Anterior
eyes subequal in size or with ALE slightly larger, posterior
eyes subequal, slightly smaller than anterior eyes; anterior eyes
separated by about half an AME; PME separated from each
other by about their diameter, from PEE by twice a PME
diameter (Fig. 32). Chelicerae with 4 promarginal and 4
retromarginal teeth (Fig. 33). Epigynal teeth long, contiguous,
arising anteriorly; epigynal hoods situated medially; atrium
reduced to slit; copulatory ducts indistinct; spermathecae
extending and abruptly converging anteriorly, separated by

more than their width at bases and slightly separated at apices;
spermathecal heads large, arising distally on spermathecae
(Figs. 26, 27).

Male: Large-sized coelotine, total length 10.00. Eyes and
chelicerae same as in female. Patellar apophysis broad,
dorsally concave; RTA more than half length of tibia, with
distinctly extended distal end; lateral tibial appphysis absent;
cymbial furrow about 1/4 of cymbial length; conductor short,
basal lamella indistinct, dorsal apophysis slightly larger than
conductor; median apophysis small, spoon-shaped, without
free standing anterior edge, situated posteriorly, close to
embolic base; embolus short, filiform, prolateral in origin,
thread originating near middle of tegular sclerite (Figs.
28-31).

Distribution. — Kazakhstan, Kyrgyzstan, Uzbekistan, Russia
(Fig. 37).

Coelotes vignai Brignoli 1978
Figs. 34—37

Coelotes vignai Brignoli, 1978a:535, fig. 133 (female holotype

and paratypes, in MHNG, examined).

Material examined. — TURKEY: 1? holotype, 3? paratypes,
Bolu, Abant, 12 July 1971, A. Vigna (MHNG); 2? paratypes,
Bolu, Abant, 1400 m, 17 July 1971, P. Brignoli (Coll. Brignoli).

Diagnosis. — Females similar to C. coenobita in having
closely arising and extended epigynal teeth, posteriorly
situated atrium, and the broad, round spermathecae but can
be distinguished by the presence of atrial septum and
separated spermathecae (Figs. 34, 35).

Description. — See Brignoli (1978a) for more somatic de-
scriptions.

Female: Medium-sized coelotine, total length 9.31. AME
smallest, about half the size of ALE, ALE largest, PLE slightly
smaller than ALE, PME slightly larger than AME; AME
separated from each other by less than their diameter, from
ALE by about half of AME diameter; PME separated from
each other by about their diameter, from PLE by about 1.5-2
times a PME diameter (Fig. 36). Chelicera with 3 promarginal
and 3 retromarginal teeth. Epigynal teeth long, slender, arising
anteriorly, close together; atria situated posteriorly, with
distinct septum; copulatory ducts indistinct from dorsal view;
spermathecae broad, round, slightly separated; spermathecal
heads large, arising distally on spermathecae (Figs. 34, 35).

Male: Unknown.

Distribution. — Turkey (Fig. 37).

ACKNOWLEDGMENTS

We would like to thank N. 1. Platnick (AMNH), B. Hauser
(MNHG), C. Rollard (MNHN) for the loan of specimens, and
S. Ovtchinnikov for donating coelotine specimens to the
AMNH. We are particularly grateful to S. Q. Li (Institute of
Zoology, Beijing), R. Raven (Queensland Museum, Brisbane),
and two anonymous reviewers for reading and improving the
manuscript.

LITERATURE CITED

Brignoli, P.M. 1978a. Ergebnisse der Bhutan-Expedition 1972 des

Naturhistorischen Museums in Basel. Araneae: Fam. Oonopidae,

Agelenidae, Hahniidae und Mimetidae. Entomologiea Basiliensia

3:31-56.

WANG & ZWJ—COELOTES CHARITONOVI GROUP (COELOTINAE)

281

Brignoli, P.M. 1978b. Ragni di Turchia. V. Specie niiove o
interessanti, cavernicole ed epigee, di varie famiglie (Araneae).
Revue Suisse de Zoologie 85:461-541.

Charitonov, D.E. 1969. Material’i k faune paukov SSR. Uchenye
Zapiski Permskogo Ordena Trudovogo krasnogo Znameni gosu-
darstvennogo Universiteta Imeni Alekseya Maksimovicha Gor-
kogo (5 Memories of Sciences of Perm University. Serial:
Biological) 179:59-133.

de Blauwe, R. 1973. Revision de la famille des Agelenidae (Araneae)
de la region mediterraneenne. Bulletin-Institut royal des sciences
naturelies de Belgique 49(2): 1-1 1 1.

Esyunin, S.L., T.K. Tuneva & G.S. Farzalieva. 2007. Remarks on the
Ural spider fauna (Arachnida, Aranei), 12. Spiders of the steppe
zone of Orenburg Region. Arthropoda Selecta 16:43-63.

Ovtchinnikov, S.V. 1984. A new species of the genus Coelotes
Blackwall (Aranei, Agelenidae) from Kirghizia. Entomologicheskie
issledovaniia v Kirgizii 17:126-131.

Ovtchinnikov, S.V. 1988. Materials on spider fauna of the superfam-
ily Amaurobioidea of Kirghizia. Entomologicheskie issledovaniia v
Kirgizii 19:139-152.

Ovtchinnikov, S.V. 1999. On the supraspecific systematics of the
subfamily Coelotinae (Araneae, Amaurobiidae) in the former
USSR fauna. TETHYS Entomological Research 1:63-80.

Ovtsharenko, V.I. & V.Y. Eet. 1980. Eauna and ecology of spiders
(Aranei) of Badhyz (Turkmenian SSR). Revue d’Entomologie de
rURSS 59:442^7.

Wang, X.P. 2002. A generic-level revision of the spider subfamily
Coelotinae (Araneae, Amaurobiidae). Bulletin of the American
Museum of Natural History 269:1-150.

Wang, X.P. 2009. Online Coelotinae, Version 2.0. Online at http://
www.amaurobiidae.com (accessed: March 3, 2009).

Manuscript received 24 December 2008, revised 19 April 2009.

2009. The Journal of Arachnology 37:282-286

Predation by Misiimenops pallidus (Araneae: Thomisidae) on insect pests of soybean cultures in Buenos

Aires Province, Argentina

Alda Gonzalez', Gerardo Liljesthrom', Elizabet Minervino', Dolores Castro-, Sandra Gonzalez' and Andrea Armendano': 'Centro
de Estudios Parasitologicos y de Vectores (CEPAVE) (CCT-CONICET-La Plata) (UNLP), 2 N° 584, 1 900 La Plata, Argentina;
-Facultad de Ciencias Naturales y Museo, UNLP. Laboratorio 32 calle 64 N° 3 La Plata, Argentina. E-mail:
asgonza lez@cepave.edu. a r

Abstract. This study analyzes predation by adult females of Misiimenops pallidus (Keyserling 1880) on pairs of prey items
representing non-pest insects and potential pests. The phenology of the potential pests was such that each insect guild
peaked sequentially, while non-pest herbivorous and insectivorous insects were present during the entire period. Field
experiments were made in a commercial 50-ha soybean plot during two successive years. Ten cages 1 X 1 X 0.5 m were
placed in peripheral furrows of a soybean commercial plot. The pest species were preyed on differentially, with the order
from the most favored species to the least with respect to non-pest herbivorous and insectivorous insects was as follows;
defoliating lepidopterous larvae, seed feeding pentatomids in their early nymphal instars, stem boring lepidopterous larvae,
and seed feeding pentatomids in older nymphal and adult instars. Adult females of M. pallidus fed on all the insect species
offered, but in the presence of defoliator larvae, they hardly accepted alternative prey, whereas in the presence of other
prey, they maintained a more generalized diet.

Keywords: Agroecosystems, biological control, natural enemies

True predators kill their prey more or less immediately after
attacking them, and during their lifetime they will kill several
or many different prey individuals. Although most true
predators have relatively broad diets, some degree of
preference is almost always present (Begon et al. 1996). There
is evidence that generalist arthropod predators choose to eat
certain prey to balance their amino-acid requirements and
therefore may be affected by previous feeding (Greenstone
1979).

While most ecological studies on spiders as potential
biocontrol agents in agroecosystems have focused on Lycosi-
dae, Linyphiidae, and Araenidae, much less is known about
Thomisidae (Dean et al. 1987; Agnew & Smith 1989; Lang et
al. 1999; Symondson et al. 2002; Vichitbandha & Wise 2002;
Romero & Vasconcello- Neto 2003; Harwood et al. 2004). In
soybean plots in the rolling pampa (Argentina), thomisids
represented 47.2% of all spiders collected in the herbaceous
stratum, and Misiimenops pallidus (Keyserling 1880) is the
most abundant species (Liljesthrom et al. 2002). It is a rather
small (<10 mm) spider that hunts preferentially in the upper
and medium strata of the soybean plants. Following the
classification proposed by Uetz et al. (1999), it belongs to the
guild of hunting ambushers. M. pallidus has a relatively long
developmental period, and presumably only one generation
occurs during the soybean growth period. However, M.
pallidus exhibits some characteristics of “agrobiont” spiders
(Luczak 1979). As a species, it disperses widely, colonizes
crops from the beginning of their development, and the adults
(particularly females) are found throughout the entire soybean
cycle. Apart from this, daily predation rates increase with
increasing age of the spider, so that the total number of prey
killed by an adult female represents 81% of prey consumed
during its lifetime (Liljesthrom et al. 2002; Gonzalez et al.
2009).

The most important examples of potential insect prey in
soybean fields are represented by three guilds (Root 1967) of

primary or secondary pests: intermediate-sized defoliating
lepidopterous larvae, stem boring lepidopterous larvae, and
seed feeding pentatomids, as well as a complex of other non-
pest herbivorous and insectivorous insects (Bimboni 1985; La
Porta & Crouzel 1985; Gamundi 1985; Arag6n & Belloso
1987; Bercellini & Malacalza 1994; Luna & Sanchez 1999b).
Previous results showed that in the laboratory M. pallidus
preyed on all prey types except curculionids and adult stink
bugs (Cheli et al. 2006). In the present study area, the
defoliating guild was represented by Rachiplusia nii Guennee
(Lepidoptera: Noctuidae) (Luna & Sanchez 1999a), the stem-
boring guild by Crocidosema (=Epinotia) aporema (Walsing-
ham) (Lepidoptera: Tortricidae) (Luna et al. 1996; Liljesthrdm
& Rojas 2005), and the seed-feeding stink bugs by Nezara
viridula (Linnaeus) and Piezodonis giiildinii West. (Hemiptera:
Pentatomidae) as the most abundant species (Bimboni 1985;
Bercellini & Malacalza 1994; Liljesthrom & Coviella 1999;
Frana 2008). Among non-pest herbivorous and insectivorous
insects, the following were the most abundant: Colaspis sp.
and Diahrotica speciosa (Germ.) (Coleoptera: Chrysomelidae),
Eriopis conexa (Germ.), Cycloneda sanguinea (Germ.), Cocci-
nella sp. (Coleoptera: Coccinelidae), unidentified species of
Membracidae and Cicadellidae (Homoptera: Auchenor-
rhynca), and a species of Curculionidae (Aragon & Belloso
1987; Bercellini & Malacalza 1994).

Although non-pest insects were recorded during the entire
soybean growth period with only small variations in density,
the phenology of the potential pests is such that each guild
peaks sequentially. R. mi larvae peaked during the early
vegetative stages of soybean growth (V4-V6) in late January
during two consecutive years, mean peak density was 17.6
larvae/ linear meter; stem-boring larvae peaked in numbers
(9.1 individuals/ linear meter) during later vegetative stages
(V7-V8) in mid February. The small instars of seed feeders
peaked during reproductive stages (R3-R6) in mid March
(mean peak density was 11.2 larvae/ linear meter) and large

282

GONZALEZ ET M..—MISUMENOPS PREYS ON SOYBEAN PESTS

283

instars peaked 2-3 wk later at a lower density (Aragon &
Belloso 1987; Bercellini & Malacalza 1994; Gonzalez et al. in
press).

In this work, we analyze daily predation by adult female M
palUdus on pairs of different prey types, each pair consisting of
one pest and one non-pest species. We hypothesized that M.
paliidus would prey indistinctly on all prey except adult
curculionids (a component of non-pest insects in some
treatments) and adult stink bugs, and that they would exhibit
similar preferences toward all pairs of prey types offered,
excluding the former.

METHODS

Predation by adult female M paliidus was estimated in four
field experiments conducted in a commercial 50-ha soybean
plot located in Chivilcoy (34°54'S, 60°02'W), Buenos Aires
province, during two successive years. The area belongs to the
pampean phytogeographic province (Cabrera 1976). The
climate is humid-mesothermal, and apart from cattle, the
main agricultural products are soybean, maize, and wheat.

Two kinds of potential prey (a pest and a non-pest
species) were offered in each of four experiments that were
repeated in both years, and the density of potential prey (4
per cage) represented approximately the mean pest density
(expressed as the number of individuals per linear meter of
crop) during the entire phenological soybean period in the
field (Gonzalez et al. 2009). In each experiment, we used 10
cages, 1 X 1 X 0.5 m, covered with a nylon net (1 X 1 mm).
The cages were placed in peripheral furrows of a commercial
soybean plot. The area where each cage would be placed was
fixst cleared by hand. All litter, soybean plants, weeds, and
insects were eliminated, and the bare ground flattened and
compacted. Then we placed three plastic pots containing
soybean plants in a phenological state similar to those of the
crop and covered them with the cage (which was sunk into
the soil to prevent the entrance of ground-active arthro-
pods). We placed potential prey individuals in each cage, as
follows:

1) Two R. nu larval instars; defoliators) and two

non-pest insects (D. spedosa and Cocdneila sp. in the
first year, and D. spedosa and Colaspis sp. in the
second year). The experiment was carried out in
late January when defoliating iepidopterous larvae
peaked.

2) Two C. aporema (4**’-5**’ larval instars, which were
placed on a leaf near a bud. It was expected that each
larva would protect itself by stitching together silk
threads or boring a stem before a spider and two non-
pest insects (D. spedosa and a membracid in the first
year and D. spedosa and an unidentified curculionid
in the second year) were released into the cage. The
experiment was carried out in mid-February when
stem-boring Iepidopterous larvae peaked.

3) Two N. viridula (3*''^ instar nymphs, seed-feeding) and
two non-pest insects (D. spedosa and a membracid in
the first year and only D. spedosa in the second year).
The experiment was carried out in mid-March when
seed-feeding pentatomid nymphal instars peaked.

4) Two N. viridula (5‘^ instar nymphs) and adult P.
guildinii (both seed-feeders) and two non-pest insects
{Cocdneila sp., which preyed mainly on aphids and
Thysanoptera, and D. spedosa in the first year and
only D. spedosa in the second year). The experiment
was carried out in late March, when adult stink bugs
usually exhibit maximum density.

The prey species and adult spiders used in the experiments
were collected in the field with a 40-cm diameter sweeping net.
The specimens were deposited in the Arachnological Labora-
tory of the Center of Parasitological Studies (University of La
Plata) at the study’s conclusion. The potential prey were
placed in the 10 cages (grouped in pairs of a pest and a non-
pest species) at the beginning of the experiments and left for
24 hours to allow acclimation. Adult female spiders were
collected in the field three to four days before, and after a 48-
hour fasting period were provided with an adult Drosophila
melanogaster for standardization of hunger level. Then one
spider was introduced into each of five cages selected at
random, while the remaining five cages were spider-free
controls. Prey and predators were kept together for 24 hours;
then we disassembled the cages and counted the number of
individuals remaining.

For each of the four combinations of prey items and for
both years, we analyzed predation by adult female M. paliidus
by calculating the predation rate per cage (Kajak et al. 1968).
We calculated the daily predation rate per cage on the i-th prey
type, DPR(0, as:

DPR(/) = [No(/) - Nf (/) - Ns(/)]/No(/)/At

where No(/) and Nf(/) represent, respectively, the initial and
final number of the i-th prey type in the cage with spiders,
Ns(/) the mean number of missing i-th prey type in the control
cages, and At = 1 day, the duration of the experiment. Each
year, possible differences between treatments were analyzed by
MANOVA, and in each treatment DPR(/) values correspond-
ing to the i-th insect pest and non-pest insects in each
experiment was examined with a Ntest. We also calculated the
total number of all prey types preyed on per cage per day:

DPR = [No(/) - Nf (/) - Ns(0]

+ \Ho{NPl) - m{NPI) - Ns( AT/)]

Differences among treatments in both years (prey killed per
cage) were tested by ANOVA, and homogeneity of variances
was tested by Levene’s test. No transformation was necessary.

RESULTS

We recorded the disappearance of only one insect from the
control cages: a R. nu in the second year. In the treatment
cages M paliidus caught prey of all those offered except the
adult curculionid and P. guildinii. However, the predation
rates showed that some pests were preferred over non-pest
insects, while others were less preferable. The same tendency
was observed both years: iA. 30 = 7.154; P = 0.00008 (first
year), and F3 30 = 3.997; P = 0.0047 (second year) (Figs, la,
b).

When we analyzed each treatment separately, the defoliat-
ing Iepidopterous larvae exhibited a higher predation rate than

284

THE JOURNAL OF ARACHNOLOGY

□ Pest species
ONPI

Figure la, b. — Predation rate of an adult female Misumenops
pallkbis when a pest species and non-pest insects were offered, a.
During the first year of studies; b. During the second year of studies.
(DLL = defoliator lepidopterous larvae, SBL = stem borer
lepidopterous larvae, SFN = seed feeding pentatomids in the first
three nymphal instars, SFA = seed feeding pentatomids in older
nymphal instars and in the adult stage NPI = non-pest herbivorous
and insectivorous insects). Confidence intervals at 90% probability.

non-pest insects. Conversely, in both years the stem-boring
lepidopterous larvae were preyed on less than the non-pest
insects. In both years the predation rate on stink bugs did not
differ statistically from that of non-pest insects. In the first
three nymphal instars, the predation rate on the bugs was
similar to that of non-pest insects. When bugs were later instar
nymphs or in the adult stage, the predation rate on later-stage
nymphs and adults was lower than that on non-pest insects,
but not significantly so. In the latter case, only late-instar
nymphs, but not adults, were preyed on.

For the total number of prey killed daily per cage, the daily
predation rates varied from 1 to 1.8 in the first year, but
differences were not significant = 1-85, P < 0.18). In the
second year, minimum and maximum daily predation rates

were 1.2 and 1.8, respectively, and non-significant (Ca.m —
1.02, P < 0.41). Thus, regardless of the prey types offered, the
spiders killed approximately the same number of prey items.

DISCUSSION

Adult female M. palUdiis killed one to almost two prey items
per cage per day and fed on all the offered insect species except
adult stink bugs and adult curculionids. These values are
congruent with field estimates by Edgar (1969) and Nyffeler &
Benz (1987, 1988); however, prey densities could be different
in those situations. M palUdus acted almost as a specialist
predator in the presence of the defoliating R. nu, whereas in
the presence of other prey guilds, it showed a more varied
feeding pattern. Yet the low attack rate of the spider on the
stem s is likely due to a low encounter rate with this prey guild,
similar to small defoliating larvae. The stem-boring larvae
remain hidden from potential predators inside cocoons made
of leaves stitched together with silk threads. However, they are
parasitized by larval parasitoids (Liljesthrom & Rojas 2005).
Other species of herbivores that spend a large proportion of
their life cycles in cryptic locations are not susceptible to
significant predation by carabids, staphylinids, and spiders but
are susceptible to specialist natural enemies (Ramert & Ekbom
1996). On the other hand, the low attack rate on stink bugs is
likely due to a certain degree of invulnerability of this prey
guild, particularly adults because of their relatively large size,
their thick cuticle, and production of toxic compounds
(Aldrich 1988; Staddon 1979; Pareja et al. 2007). These factors
probably explain why most of the predation on this guild was
on fifth-instar nymphs (mainly by N. viridula, which is larger
than P. guildinii); we only saw nymphs preyed on by adult
spiders in the field. The thickness of the adult curculionid
cuticles could also account for our results.

Studies by Turnbull (1960) and Riechert & Lockley (1984)
demonstrated prey size selectivity in spiders, and Nentwig
(1983) showed that most of his web-building spiders caught
and ate almost anything small enough for them to handle that
arrived in their webs, although they rejected prey that was
toxic or whose cuticule was too hard to penetrate.

Spiders are the main natural enemies of pest insects in some
agroecosystems, which makes them agronomically significant
(Riechert & Lockley 1984). Notwithstanding, early biological
control studies focused on specialist predators rather than on
spiders because, as generalists, they were thought to have
relatively little impact on agricultural pests (Savory 1928;
Bristowe 1941; Comstock 1965). This view has changed, and
different studies have demonstrated the capacity of spiders to
reduce the population density of some pest insects and their
consequent damage (Riechert & Lockley 1984; Nyffeler et al.
1992; Riechert & Lawrence 1997; Marshall et al. 2002). Still,
other studies indicate limited pest control potential under
certain conditions (Holland & Thomas 1997; Lang et al. 1999;
Birkhofer et al. 2008). On the other hand, since effective
biological control is the result' of complex interactions at the
community level, the activity of spiders would be comple-
mented by that of other natural enemies (Sunderland 1999),
although it was observed in the field that M pallidas preyed
on the taquinid Trichopoda giacomellii (Blanchard), a parasit-
oid of N. viridula (Liljesthrom, pers. obs.), and coccinelid
predators (this study).

GONZAlEZ ET kh.—MISUMENOPS PREYS ON SOYBEAN PESTS

285

Polyphagy helps to sustain predators when pest density is
low in agroecosystems, usually early in the season. Generalists
may already be present, subsisting on non-pest prey, while
specialist parasitoids may take some time to build up in
numbers (Symondson et al. 2002). In soybean cultures in the
study area, defoliating and stem-boring larvae have different
guilds of parasitoids (Luna & Sanchez 1999a; Liljesthrdm &
Rojas 2005), while late-instar nymphs and adults of N. viridula
are attacked by T. giacomellii (Liljesthrdm & Bernstein 1990;
Liljesthrom & Rabinovich 2004). Yet the only arthropod
natural enemies found on young-instar nymphs of N. viridula,
as well as on adults and immature stages of P. guildinii, are
predators, among which spiders are the most abundant
(Bercellini & Malacalza 1994; Liljesthrom et al. 2002). We
found many other spider species representing seven guilds that
reached relatively high densities (Liljesthrdm et al. 2002) and
could prey on certain potential pest species. Even though their
effect is unknown in the agroecosystem, they are important
mortality factors on the pest species considered in this study,
as well as on other phytophagous insects that rarely attain pest
status.

ACKNOWLEDGMENTS

We are grateful to two anonymous referees for helpful
suggestions on previous versions of this manuscript.

LITERATURE CITED

Aldrich, J.R. 1988. Chemical ecology of the Heteroptera. Annual
Review of Entomology 33:211-238.

Aragon, J.R. & C.A. Belloso. 1987. Control de plagas. In: Produccion
de soja. Cuaderno de Actualizacion Tecnica N° 58. Asociacion
Argentina de Consorcios Regionales de Experimentacion Agricola,
Buenos Aires, Argentina.

Agnew, C.W. & J.W. Smith. 1989. Ecology of spiders (Araneae) in a
peanut agroecosystem. Environmental Entomology 18:30-42.
Begon, M., M. Mortimer & D.J. Thompson. 1996. Population
Ecology: A Unified Study of Plants and Animals. 3rd Edition.
Wiley-Blackwell Science, Oxford, UK.

Bercellini, N. & L. Malacalza. 1994. Plagas y depredadores en soja en
el noroeste de la provincia de Buenos Aires (Argentina). Turrialba
44:244-254.

Bimboni, M.A. 1985. Plagas de la soja. Estacion Experimental
Agropecuaria San Pedro, Instituto Nacional de Tecnologia
Agropecuaria, Buenos Aires, Argentina.

Birkhofer, K., A. Fliessbach, D.H. Wise & S. Scheu. 2008. Generalist
predators in organically and conventionally managed grass-clover
fields: implications for conservation biological control. Annals of
Applied Biology 153:271-280.

Bristowe, W.S. 1941. The Comity of Spiders. Volume II. The Ray
Society, London.

Cabrera, A.L. 1976. Regiones Fitogeograficas Argentinas. Enciclo-
pedia Argentina de Agricultura y Jardinen'a. Tomo II, Fascicule 1.
Editorial Acme S.A.C.I., Buenos Aires, Argentina.

Comstock, J.H. 1965. The Spider Book. Comstock Publishing,
Ithaca, New York.

Cheli, G., A. Armendano & A. Gonzalez. 2006. Preferencia
alimentaria de aranas Misumenops pallidus (Araneae: Thomisidae)
sobre potenciales insectos presa de cultivos de alfalfa. Journal of
Tropical Biology 54:505-513.

Dean, D.A., W.L. Sterling, M. Nyffeler & R.G. Breene. 1987.
Foraging by selected spider predators on the cotton fleahopper
(Hemiptera, Miridae) and other prey. Southwestern Entomologist
12:263-270.

Edgar, W.D. 1969. Prey and predators of the wolf spider Lycosa
lugubris (Walckenaer). Journal of Zoology 159:405-411.

Frana, J.E. 2008. Tendencias de abundancia de chinches en soja en el
centre de Santa Fe. Pp. 21-30. In Chinches fitofagas en soja.
Revision y avances en el estudio de su ecologia y manejo. (E.V.
Trumper & J.D. Edelstein, eds.). Estacion Experimental Agrope-
cuaria, Instituto Nacional de Tecnologia Agropecuaria, Manfredi,
Buenos Aires, Argentina.

Gamundi, J.C. 1985. Control integrado de insectos plaga en el cultivo
de soja. In: Actas de Jornadas de Actualizacion Profesional Sobre
Cultivos de Cosecha Gruesa., Instituto Nacional de Tecnologia
Agropecuaria, Buenos Aires Argentina.

Gonzalez, A., G. Liljesthrom, D. Castro & A. Armendano. 2009.
Development and recruitment of Misumenops pallidus (Keyserling)
(Araneae: Thomisidae), and its synchrony with three potential prey
species in soybean cultures from Argentina. Entomological News
120:41-53.

Greenstone, M.H. 1979. Spider feeding behaviour optimises dietary
essential amino acid composition. Nature 282:501-503.

Harwood, J.D., K.D. Sunderland & W.O.C. Symondson. 2004. Prey
selection by linyphiid spiders: molecular tracking of the effects of
alternative prey on rates of aphid consumption in the field.
Molecular Ecology 13:3549-3560.

Holland, M. & S.R. Thomas. 1997. Quantifying the impact of
olyphagous invertebrate predators in controlling cereal aphids in
preventing wheat yield and quality reductions. Annals of Applied
Biology 131:375-397.

Kajak, A., L. Andrzejewska & Z. Wojcik. 1968. The role of spiders in
the decrease of damages caused by Acridoidea on meadows.
Experimental investigation. Ekologia Polska 16:756-764.

Lang, A., J. Filser & J.R. Henschel. 1999. Predation by ground beetles
and wolf spiders on herbivorous insects in a maize crop.
Agriculture, Ecosystems and Environment 72:189-199.

La Porta, N.C. & I.S. Crouzel. 1985. Estudios basicos para ei control
biologico de Nezara viridula (L., 1758) (Hemiptera: Pentatomidae)
en la Argentina. Revista de la Sociedad Entomologica Argentina
43:119-143.

Liljesthrom, G. & C. Bernstein. 1990. Density dependence and
regulation in the system Nezara viridula (L.) (Hemiptera: Penta-
tomidae), host and Trichopoda giacomellii (Blanchard) (Diptera:
Tachinidae), parasitoid. Oecologia 84:45-52.

Liljesthrom, G. & C. Coviella. 1999. Aspectos de la dinamica
poblacional de las chinches Nezara viridula y Piezodorus guildinii e
implicancias con relacion a su manejo en el cultivo de soja. Revista
de la Sociedad Entomologica Argentina 58:141-149.

Liljesthrom, G., E. Minervino, D. Castro & A. Gonzalez. 2002. La
comunidad de aranas del cultivo de soja en la provincia de Buenos
Aires, Argentina. Neotropical Entomology 31:197-209.

Liljesthrom, G. & J. Rabinovich. 2004. Modeling biological control:
the population regulation of Nezara viridula by Trichopoda
giacomellii. Ecological Applications 14:254—267.

Liljesthrom, G. & G. Rojas. 2005. Parasitismo larval de Crocidosema
{=Epinotia) aporema (Lepidoptera: Tortricidae) en el noreste de la
provincia de Buenos Aires (Argentina). Revista de la Sociedad
Entomologica Argentina 64:37-44.

Luezak, J. 1979. Spiders in agrocoenoses. Polish Ecological Studies
5:151-200.

Luna, M.G., E. Minervino & N. Buselini. 1996. Diagnostico sobre el
manejo de plagas en el cultivo de soja en el partido de Chivilcoy,
Buenos Aires, Argentina. Revista de Agronomia de La Plata 101:1-6.

Luna, M.G. & N.E. Sanchez. 1999a. Composicion especifica y
abundancia de la comunidad de lepidopteros defoliadores de la
soja en el noroeste de Buenos Aires, Argentina. Revista de la
Sociedad Entomologica Argentina 58:67-75.

Luna, M.G. & N.E. Sanchez. 1999b. Parasitoid assemblages of
soybean defoliator Lepidoptera in north-western Buenos Aires

286

THE JOURNAL OF ARACHNOLOGY

province, Argentina. Agricultural and Forest Entomology
1:255-260.

Marshall, S.D., D.M. Pavuk & A.L. Rypstra. 2002. A comparative
study of phenology and daily activity patterns in the wolf spiders
Pardosa milvina and Hogna helluo in soybean agroecosystems in
southwestern Ohio (Araneae, Lycosidae). Journal of Arachnology
30:503-510.

Nentwig, W. 1983. The prey of web-building spiders compared with
feeding experiments (Araneae: Araneidae, Linyphiidae, Pholcidae,
Agelenidae). Oecologia 56:132-139.

Nyffeler, M. & G. Benz. 1987. Spiders in natural pest control: a
review. Journal of Applied Entomology 103:321-339.

Nyffeler, M. & G. Benz. 1988. Feeding ecology and predatory
importance of wolf spiders (Pardosa spp.) (Araneae, Lycosidae) in
winter wheat fields. Journal of Applied Entomology 106:123-134.

Nyffeler, M., D.A. Dean & L. Sterling. 1992. Diets, feeding
specialization, and predatory role of two lynx spiders, Oxyopes
salticus and Peiicetia viridans (Araneae, Oxyopidae) in Texas
cotton agroecosystem. Environmental Entomology 21:1457-1465.

Pareja, M., M. Borges, R.A. Laumann & M.C.B. Moraes. 2007.
Inter- and intraspecific variation in defensive compounds produced
by five neotropical stink bug species (Hemiptera: Pentatomidae).
Journal of Insect Physiology 53:639-648.

Riimert, B. & B. Ekbom. 1996. Intercropping as a management
strategy against carrot rust fly (Diptera: Psilidae): a test of enemies
and resource concentration hypotheses. Environmental Entomol-
ogy 25:1092-1100.

Riechert, S.E. & T. Lockley. 1984. Spiders as biological control
agents. Annual Review of Entomology 29:299-320.

Riechert, S.E. & K. Lawrence. 1997. Test for predation effects of
single versus multiple species of generalist predators: spiders and
their insect prey. Entomologia Experimentalis et Applicata
84:147-155.

Romero, G.Q. & J. Vasconello-Neto. 2003. Natural history of
Misumenops argenteus (Thomisidae): seasonality and diet on
Trichogoniopsis adetuuUha (Asteraceae). Journal of Arachnology
31:297-304.

Root, R.B. 1967. The niche exploitation pattern of the blue-gray
gnatcatcher. Ecological Monographs 37:317-350.

Savory, T.H. 1928. The Biology of Spiders. Macmillan Co., New
York.

Staddon, B.W. 1979. The scent glands of Heteroptera. Advances in
Insect Physiology 14:351^19.

Sunderland, K. 1999. Mechanisms underlying the effects of spiders on
pest populations. Journal of Arachnology 27:308-316.

Symondson, W.O.C., K.D. Sunderland & M.H. Greenstone. 2002.
Can generalist predators be effective biocontrol agents? Annual
Review of Entomology 47:561-594.

Turnbull, A.L. 1960. The prey of the spider Linyphia triangularis
(Clerck) (Araneae, Linyphiidae). Canadian Journal of Zoology
38:859-873.

Uetz, G.W., J. Halaj & A.B. Cady. 1999. Guild structure of spiders in
major crops. Journal of Arachnology 27:270-280.

Vichitbandha, P. & D.H. Wise. 2002. A field experiment on the
effectiveness of spiders and carabid beetles as biocontrol agents in
soybean. Agricultural and Forest Entomology 4:31-38.

Manuscript received 23 July 2008, revised 5 May 2009.

2009. The Journal of Arachnology 37:287-291

Karyotypes of the Neotropical pseudoscorpions Semeiochernes armiger and Cordylochernes scorpioides

(Pseudoscorpiones: Chernetidae)

Frantisek Sf ahlavsky: Department of Zoology, Faculty of Science, Charles University in Prague, Vinicna 7, Prague 2,
CZ - 128 44, Czech Republic. E-mail; stahlf@natur.cuni.cz

Jeanne A. Zeh and David W. Zeh: Department of Biology and Program in Ecology, Evolution and Conservation Biology,
University of Nevada, Reno, NV 89557, USA

Jifi Krai: Laboratory of Arachnid Cytogenetics, Department of Genetics and Microbiology, Faculty of Science,
Charles University in Prague, Vinicna 5, Prague 2, CZ - 128 44, Czech Republic

Abstract. The karyotypes and course of meiosis of two pseudoscorpions, Semeiochernes armiger (Balzan 1892) and
Cordylochernes scorpioides (Linnaeus 1758) (Pseudoscorpiones; Chernetidae), are described for the first time. The diploid
chromosome number of the male is 69 in S. armiger and 47 in C. scorpioides. As in most pseudoscorpions studied to date,
autosomes exhibit predominantly biarmed morphology. Both species possess an XO sex chromosome system. In most
pseudoscorpions with XO system karyotyped so far, including European chernetids, the X chromosome exhibits
metacentric morphology. In contrast, the X chromosome of both neotropical chernetids studied exhibits asymmetric,
submetacentric morphology.

Keywords: Pseudoscorpiones, Chernetidae, karyotype, sex chromosome, XO sex chromosome determination

With more than 3,380 described species, Pseudoscorpiones is the
fourth largest order of the arthropod class Arachnida (Harvey 2008).
Despite this considerable species diversity, the morphology of
pseudoscorpions is very conservative, and it can therefore be difficult
to distinguish between closely related species on the basis of
morphological features alone (Zeh & Zeh 1994; Wilcox et al. 1997;
Zeh et al. 2003). Moreover, in many groups of pseudoscorpions,
species identification is also complicated by the lack of detailed
analysis of intraspecific morphological variability. For example,
researchers originally described the neotropical pseudoscorpion,
Semeiochernes armiger (Balzan 1892), from Central and South
America as three species (S. armiger, S. extraordinarius, and S.
militaris) on the basis of sexually dimorphic traits of the male
pedipalpal chelae (Beier 1933, 1954). However, rearing experiments
have demonstrated that intrapopulation variability encompasses the
full range of Semeiochernes “interspecific” chelal morphology and
that species status cannot be established using these male characters
(Zeh & Zeh 1992).

Recent studies suggest that molecular and cytogenetic characters
in pseudoscorpions diverge more rapidly than morphological traits
and may thus prove particularly useful for identifying cryptic species
and for resolving fine-scale evolutionary relationships. For example,
the harlequin beetle riding pseudoscorpion, Cordylochernes scor-
pioides (Linnaeus 1758), ranging from Costa Rica to southern Brazil,
was described by Beier (1948) as a single species, based on
morphological examination of hundreds of specimens from several
countries in South and Central America. However, mitochondrial
cytochrome oxidase I (COI) gene sequencing has revealed extensive
genetic differentiation, with a maximum likelihood nucleotide
divergence of nearly 33% between C. scorpioides populations from
Panama and northern South America (Trinidad and French Guiana)
(Zeh et al. 2003). This extreme molecular divergence is associated
with complete postzygotic incompatibility between individuals from
central Panama and both French Guiana (Zeh & Zeh 1994) and
Trinidad (J.A. Zeh, unpublished data), indicating that geographic
populations of C. scorpioides constitute a complex of cryptic species.
Interestingly, researchers have documented extensive mitochondrial
COI sequence divergence between individuals from Panama and
Trinidad in S. armiger, suggesting that the pattern exhibited by C

scorpioides may be common and that many neotropical pseudoscor-
pion species may actually represent cryptic species complexes
(Wilcox et al. 1997).

Karyotype data also holds great promise as a tool for differenti-
ating between closely related taxa. There is considerable karyotype
diversity in all genera of pseudoscorpions that have been studied in
detail to date, namely Ronciis (Neobisiidae) (Troiano 1990, 1997),
Chthonius (Chthoniidae) (St’ahlavsky & Krai 2004), Lasiochernes
(Chernetidae) (St’ahlavsky et al. 2005), Geogarypus (Geogarypidae),
and Olpium (Olpiidae) (St’ahlavsky et al. 2006). In this paper, we
present the results of the first cytogenetic study of S. armiger and C.
scorpioides. Our karyotype analyses of these two neotropical
representatives of the family Chernetidae not only contribute valuable
data for comparing patterns of karyotype evolution in neotropical
and European chernetid pseudoscorpions, but also provide a basis for
future investigation of the relationship between karyotype evolution,
molecular divergence, and speciation in Semeiochernes and Cordy-
lochernes.

METHODS

All pseudoscorpions used in this study were derived from
populations inhabiting decaying fig trees (Ficus spp.) in the lowland
rain forest of the former Canal Zone, Republic of Panama (9°N,
79°W). Voucher specimens of C. scorpioides and S. armiger have been
deposited with W.B. Muchmore (University of Rochester, USA), V.
Mahnert (Museum d’histoire naturelle, Geneva, Switzerland), and D.
Quintero (Universidad de Panama, Republic of Panama).

Semeiochernes armiger (Balzan 1892): 6 males collected in January
2006 either as adults (n = 3) or as tritonymphs that molted to the
adult stage in the laboratory (n = 3).

Cordylochernes scorpioides (Linnaeus 1758): 8 males and 8 females
from a large laboratory population established from 35 females
collected in the field in August 2000.

Chromosome preparations were made using the technique de-
scribed by St’ahlavsky & Krai (2004). Briefly, gonads were dissected,
hypotonised in 0.075 M KCl for 15 min, and fixed in a mixture of
methanokglacial acetic acid (3:1) for at least 20 min. We placed a
piece of fixed materia! into a drop of 60% acetic acid on a clean
microscope slide suspended by a pair of tungsten needles. Then we

287

288

THE JOURNAL OF ARACHNOLOGY

1

^ »

:ir

1

2

3

4

5

6

M ^

X

% X

7

8

9

10

11

12

J| %

X

Jt.

*1% JPL

13

14

15

16

17

18

%

n

M >«

»#•<*!#■ 10'

19

20

21

22

23

24

n, A

JL J.

» «

25

26

27

28

29

30

4fc>

m /t

«i%.

31

32

33

34

X

Figure 1. — Semeiochemes armiger, male karyotype. Based on two sister metaphase II plates. Bar = 5 pm.

transferred the slide onto a warm histological plate (surface
temperature of 40-45° C) and moved the drop of dispersed tissue
on the slide with a tungsten needle until it evaporated. The
chromosome preparations were air-dried at room temperature
overnight and stained with 5% Giemsa solution in Sorensen
phosphate buffer (pH = 6.8) for 40 min.

Chromosome morphology was classified according to Levan et al.
(1964). We calculated relative chromosome length as a percentage of
the total length of the diploid set, including the sex chromosome.
Owing to the small number of suitable spermatogonial mitotic
metaphase plates, we used sister metaphase II for analysis of
karyotypes in males. In addition, the centromere positions are much
more obvious in pseudoscorpions at this meiotic stage.

RESULTS

Semeiochemes armiger (Balzan 1892)

The male diploid complement comprises 69 chromosomes. The
karyotype contains 18 pairs of metacentric (Nos. 4, 5, 8, 9, 1 1, 12, 14,

15, 16, 18, 19, 20, 21, 22, 25, 27, 30, 31), seven pairs of submetacentric
(Nos. 1, 2, 7, 10, 13, 28, 29), three pairs of subtelocentric (Nos. 3, 17,
24), and six pairs of acrocentric (Nos. 6, 23, 26, 32, 33, 34) autosomes
(Fig. 1). The first three pairs of autosomes are slightly larger than the
other pairs (Fig. 1), and their relative size decreases from 3.4% to
2.5% of the diploid set. The remaining autosomes decrease gradually
in size from 1.9% to 0.7% of the diploid set.

The sex chromosome system is XO. The X chromosome shows
submetacentric morphology (centromeric index 1.83), constitutes
2.1% of the diploid set, and exhibits more intensive staining than
other chromosomes (i.e., positive heteropycnosis) during some
periods of meiotic division. During meiosis (Figs. 2-A), more
intensive staining revealed overcondensation of the X chromosome
from leptotene to pachytene (Fig. 2) and during metaphase II
(Fig. 4). By contrast, we noted that all chromosomes are isopycnotic
during metaphase I (Fig. 3).

Chiasma frequency is relatively low. In diplotene - metaphase I
plates (n = 19), we observed at least one bivalent with two chiasmata
and maximally four bivalents with two chiasmata. The mean chiasma
frequency was 1 .08 per bivalent.

Figures 2^. — Course of meiosis in Semeiochemes armiger. 2. Pachytene. 3. Metaphase I. 4. Metaphase II cell containing X chromosome.
Arrowheads indicate the X chromosome, arrow points to the bivalents with two chiasmata. Bars = 10 pm.

sTAhlavsky et al.— neotropical pseudoscorpion karyotypes

289

5

X Xtf « «

mm

mm

1

2 3 4

5

6

Jin aa

mm

mm

7

8 9 10 11

12

13

aa

an a mm at

m m

mi

14

15 16 17 18

19

20

mm

mm mm

21

22 23

X

6

u

U » tl

II

if

1

2 3 4

5

6

ti

it 11 il t»

• t

Hi

7

8 9 10 11

12

13

II

11 It ti It

ft

ft

14

15 16 17 18

19

20

is

it tl

ff

21

22 23

X X

Figures 5-6. — Cordylochernes scorpioides. 5. Karyotype of male. Based on two sister metaphase II plates. 6. Karyotype of female. Based on
mitotic metaphase. Bars = 5 pm.

Cordylochernes scorpioides (Linnaeus 1758)

The diploid number is 47 in males (Fig. 5) and 48 in females
(Fig. 6). The karyogram of the species is based on two sister
metaphases II of a male (Fig. 5). The karyotype is composed of 17
metacentric (Nos. 1, 2, 3, 4, 5, 6, 7, 9, 10, 11, 13, 14, 16, 17, 18,20,21),
three submetacentric (Nos. 8, 12, 15), and three subtelocentric
(Nos. 19, 22, 23) pairs of autosomes (Fig. 5). The relative length of
the autosomes decreases gradually from 4.4% to 1 .4% of the diploid
set in male metaphase II and from 4.8% to 1.2% of the diploid set in
female mitotic metaphase. Comparison of the male and female
karyotypes, as well as analysis of male meiosis, revealed an XO sex
chromosome system. The X chromosome is submetacentric (centro-
meric index 1.75) and relatively large, forming 3.6% of the diploid set
in males and 2.7% in females. As in S. armiger, the X chromosome of
C. scorpioides exhibits positive heteropycnosis in germinal cells of
males. However, we detected heteropycnosis only during premeiotic

interphase (Fig. 7). During late pachytene (Fig. 8), metaphase I
(Fig. 9), and metaphase II (Fig. 5), the X chromosome appeared to be
isopycnotic with the autosomes. During male meiosis, C. scorpioides
exhibited lower chiasma frequency than S. armiger (number of
analyzed diplotene plates = 378). We calculated the mean chiasma
frequency as 1.03 per bivalent. More specifically, we observed two
bivalents with two chiasmata in five diplotene nuclei and one bivalent
with three chiasmata in five diplotene nuclei. In the remaining cells,
all bivalents had only one chiasma.

DISCUSSION

Until now, karyotype descriptions of the family Chernetidae have
been limited to five species, all from the European region. Sokolow
(1926) published basic data on the karyotype of Demlrochernes
cyrneus (L. Koch 1873) in his pioneering analysis of spermatogenesis
in pseudoscorpions. However, nearly 80 years elapsed before

290

THE JOURNAL OF ARACHNOLOGY

Figures 7-9. — Course of meiosis in Corclylochernes scorpioides. 7. Premeiotic interphase. 8. Late pachytene. 9. Metaphase I. Arrowhead
indicates the X chromosome, arrows point to the bivalents with two ehiasmata. Bars = 10 pm.

scientists conducted any further karyotype analyses of chernetid
species. In a study aimed at reconstructing the phytogeny of
arthropod telomeric sequences, Vitkova et al. (2005) provided only
data on the diploid number and telomeric sequences of Cliernes hahiiii
(C.L. Koch 1839). St’ahlavsky et al. (2005) provided the first detailed
descriptions of chernetid karyotypes, which were limited to three
species in the genus Lasiociwnies. Despite limited comparative data,
European chernetids appear to be characterized by high diploid
chromosome number (49-73), a predominance of biarmed chromo-
somes, and the presence of an XO sex chromosome system
(St’ahlavsky et al. 2005).

In this study, we present the first karyotype descriptions of
pseudoscorpions from the Neotropics. As with previously karyo-
typed pseudoscorpions (e.g., Troiano 1997; St’ahlavsky & Krai
2004; St’ahlavsky et al. 2005, 2006), neotropical chernetids exhibit
considerable karyotype variability. The diploid chromosome
number of the C. scorpioides male (2n = 47) is the lowest known
diploid number within chernetids. By contrast, the male karyotype
of S. armiger consists of 69 chromosomes. Despite this disparity in
chromosome number, the karyotypes of both neotropical species
have several features in common with European chernetids. Their
karyotypes are characterized by high 2n, as well as by a
predominance of biarmed chromosomes. As with the majority of
pseudoscorpions karyotyped so far, all representatives of the
family Chernetidae possess an XO sex chromosome system.
Remarkably, European and neotropical chernetids differ in the
morphology and relative size of the sex chromosome. In all
karyotyped European species, the sex chromosome is metacentric
and is the largest element of the karyotype (St’ahlavsky et al.
2005). Metacentric morphology of the X chromosome has been
found in the majority of pseudoscorpions exhibiting the XO sex
chromosome system (Troiano 1990, 1997; St’ahlavsky & Krai 2004;
St’ahlavsky et al. 2005, 2006). Interestingly, in both neotropical
chernetids, the X chromosome exhibits submetacentric morpholo-
gy. Moreover, it is not the largest element of the karyotype. These
fundamental differences may be the result of long-term isolation
and divergent evolution of the X chromosome in European and
neotropical chernetids. Clearly, many additional cytogenetic and
molecular systematic studies of species and populations from both
biogeographical regions are needed in order to gain a better
understanding of karyotype evolution and diversity in chernetid
pseudoscorpions.

ACKNOWLEDGMENTS

Authors from the Charles University in Prague were supported by
project MSM 0021620828 (F.S. and J.K.) and by project GA UK
36908 (F.S.). We thank La Autoridad Nacional del Ambiente
(A.N.A.M.) for permission to collect pscudoscorpions in Panama
and the Smithsonian Tropical Research Institute for logistical support

in this country. Funding for field collections was provided by grants
to J.A.Z and D.W.Z. from the National Geographic Society (grant
5333-94) and the USA National Science Foundation (IBN-01 15986).
We are also grateful to two anonymous reviewers for their valuable
comments.

LITERATURE CITED

Beier, M. 1933. Revision der Chernetidae (Pseudoscorp.). Zoolo-
gische Jahrbiicher Systematik 64:509-548.

Beier, M. 1948. Phoresie und Phagophilie bei Pseudoscorpionen.

Osterreichische Zoologische Zeitschrift 1:441^97.

Beier, M. 1954. Eine Pseudoscorpioniden-Ausbeute aus Venezuela.
Memorie del Museo Civico di Storia Naturale di Verona 4:131-
142.

Harvey, M.S. 2008. Pseudoscorpions of the World, version 1.1.
Western Australian Museum, Perth. Online at http://www.museum.
wa.gov.au/arachnids/pseudoscorpions/

Levan, A.K., K. Fredga & A. A. Sandberg. 1964. Nomenclature for
centromeric position on chromosomes. Hereditas 52:201-220.
Sokolow, I. 1926. Untersuchungen iiber die Spermatogenese bei
Arachniden. II: Uber die Spermatogenese der Pseudoscorpione.
Zeitschrift fiir Zellforschung und Mikroskopische Anatomic
3:615-681.

St’ahlavsky, F., H. Henderickx & J. Krai. 2005. Karyotype study on
pseudoscorpions of the genus Lasiochernes Beier (Pseudoscor-
piones: Chernetidae). Folia biologica (Krakow) 53:69-74.
St’ahlavsky, F. & J. Krai. 2004. Karyotype analysis and achiasmatic
meiosis in pseudoscorpions of the family Chthoniidae (Arachnida:
Pseudoscorpiones). Hereditas 140:49-60.

St’ahlavsky, F., J. Krai, M.S. Harvey & C.R. Haddad. 2006. A
karyotype study on the pseudoscorpion families Geogarypidae,
Garypinidae and Olpiidae (Arachnida: Pseudoscorpiones). Euro-
pean Journal of Entomology 103:277-289.

Troiano, G. 1990. Karyotype and male meiosis of four species of
Roncus L. Koch, 1873 (Pseudoscorpionida, Neobisiidae). Bollet-
tino di Zoologia 57:1-9.

Troiano, G. 1997. Further studies on caryology of the pseudoscor-
pions of the gen. Roncus: the karyotype of Roncus gestroi and
Roncus heUucitii. Caryologia 50:271-279.

Vitkova, M., J. Krai, W. Traut, J. Zrzavy & F. Marec. 2005. The
evolutionary origin of insect telomeric repeats, (TTAGG)n.
Chromosome Research 13:145-156.

Wilcox, T.P., L. Hugg, J.A. Zeh & D.W. Zeh. 1997.
Mitochondrial DNA sequencing reveals extreme genetic
differentiation in a cryptic species complex of neotropical
pseudoscorpions. Molecular Phylogenetics and Evolution
7:208-216.

Zeh, D.W. & J.A. Zeh. 1994. When morphology misleads:
interpopulation uniformity in sexual selection masks genetic

STAHLAVSKY ET AL.— NEOTROPICAL PSEUDOSCORPION KARYOTYPES

291

divergence in harlequin beetle-riding pseudoscorpion populations.
Evolution 48: 1 1 68- 1 1 82.

Zeh, J.A. & D.W. Zeh. 1992. Are sexually-selected traits reliable
species characters? : Implications of intra-brood variability in
Semeiochernes armiger (Balzan) (Pseudoscorpionida: Chernetidae).
Bulletin of the British Arachnological Society 9:61-64.

Zeh, J.A., D.W. Zeh & M.M. Bonilla. 2003. Phylogeography of the
harlequin beetle-riding pseudoscorpion and the rise of the Isthmus
of Panama. Molecular Ecology 12:2759-2769.

Manuscript received 12 October 2008, revised 21 April 2009.

2009. The Journal of Arachnology 37:292-298

Palpal urticating hairs in the tarantula Ephebopusi fine structure and mechanism of release

Rainer Foelix, Bastian Rast and Bruno Erb: Neue Kantonsschule Aarau, Biology Department, Electron Microscopy
Unit, Zelgli, CH-5000 Aarau, Switzerland. E-mail: rainer.foelix@hotmaiLde

Abstract. The tarantula genus Ephebopus is exceptional with respect to its urticating hairs: they are located on the palps
rather than on the abdomen, as is the rule for other Neotropical tarantulas. These urticating hairs occupy a small field of 1-
2 mm^ on the medial side of the palpal femora. Each urticating hair measures 500-600 pm in length and 5-6 pm in
diameter. Almost the entire hair shaft is studded with little barbs that point toward the hair tip. Urticating hairs arise from
a slipper-shaped socket in the cuticle, at an angle of 25-30°. When the spider is threatened, it shows a brief palpal flick as a
defensive reaction, whereby many urticating hairs are brushed off and fly through the air. These hairs do not have a
preformed breaking point but become detached at the very base and are then pulled out from their sockets, like an arrow
from a quiver. The actual release behavior occurs too quickly (0.1 s) to be followed by the naked eye. Video film analyses
reveal that a single upward movement of the palps rubbing against the lateral surfaces of the spread chelicerae causes the
dispersal of urticating hairs into the air.

Keywords: Morphology, defensive behavior, Neotropical tarantulas

Many Neotropical tarantulas defend themselves by brush-
ing off special urticating hairs from their bodies (Bates 1863;
Bertani & Marques 1995/96). When these hairs come in
contact with the skin, eyes, or respiratory tract of a
threatening animal, they can cause serious irritations (Cooke
et al. 1972) or allergic reactions (Castro et al. 1995).
Commonly, urticating hairs are located on the abdomen and
are brushed off with the hind legs. However, there is one
exception in the genus Ephebopus, a colorful South American
tarantula, where the urticating hairs occur on the palps rather
than on the abdomen (Raven 1985). These hairs form distinct
fields, so-called “pedipalpal brushes” on the medial surfaces of
the femora; the hairs are released into the air by a flick of the
palps. Marshall & Uetz (1990) provided a first description of
the morphology of the urticating hairs in Ephebobus, their
release after provocation, and their effect on laboratory mice.
However, several points remained unclear: 1) structure, size,
and total number of urticating hairs, 2) attachment and
detachment of these hairs from their sockets, and 3) details
about the actual release mechanism during the defensive
reaction. We addressed the morphological questions using
light and scanning electronmicroscopical techniques. For
analyzing which body parts were involved in releasing the
urticating hairs, we performed frame-by-frame analysis on
several video recordings of defensive reactions.

METHODS

Exuvia of subadult Ephebopus cyanognathus West &
Marshall 2000 were used for all microscopical studies. Isolated
urticating hairs were inspected and measured under the light
microscope. Entire fields of urticating hairs were excised and
embedded in Epon; 1-2 pm sections were stained with
methylene blue and examined in a phase contrast microscope.
For scanning electron microscopy palpal brushes were
sputtered with gold and viewed from various directions in a
Zeiss DSM 950. Digital photographs were taken at magnifi-
cations from 20X to 5000X. For a more three-dimensional
representation of the attachment sites, we cut several palpal
brushes into longitudinal strips with a razor blade and then
mounted them sideways before examination in the scanning

electron microscope. For comparative purposes we also
studied nearby sensory hairs.

In order to observe the defensive reaction, a transparent
glass vial was moved toward the spider from in front. Slightly
touching the front legs often triggered a palpal flick and a
concomitant release of urticating hairs. Since this happened
very fast, we used a video camera (Sony DCR PC107E) with
25-30 pictures/s and frame-by-frame analysis. Overall we
elicited about 20 defensive reactions and captured three of
them on video film.

All six Ephebopus cyanognathus spiders used in this study
were bred in captivity by one of the authors (BR). They were
18 mo old, subadult, and still about two molts away from
maturity. With a body length of 3 cm they were slightly
smaller than the adults.

RESULTS

The palpal brushes are located on the medial sides of the
femora, rather distally and near the patellar joint. They
occupy a patch of 1-2 mm^ that is easily visible with the naked
eye (Fig. 1). The term “pedipalpal brush” is quite descriptive
since the hairs are tightly packed and run parallel to each
other, like a paint brush (Fig. 2). In contrast to the
surrounding sensory hairs, which are dark brown or black,
and the ornamental hairs, which are deep blue, the urticating
hairs have a golden-reddish color. Their orientation is almost
parallel to the axis of the femur, but slightly deflected
ventrally. A single urticating hair is 500-600 um long but
only 5-6 pm thick. The ratio of length/width is thus around
100:1, giving the hair a needle-like appearance. As is typical
for urticating hairs, the hair shaft bears numerous pointed
barbs, each about 5 pm long (Figs. 2, 5). They are lacking at
the very base, but appear as small cuticular extensions, just
after the hair comes out of its socket (Fig. 8). The barbs are
spaced rather regularly at an interval of 10 pm along the entire
hair shaft, which also bears fine longitudinal ridges. The distal
end of the urticating hair is acutely pointed, suggesting that it
represents the penetrating tip. However, since all the barbs are
pointing distally, this would quickly prevent any further
penetration into the skin (see Discussion).

292

FOELIX ET AL.— PALPAL URTICATING HAIRS OF TARANTULA

293

Figure 1. — Medial surface of the femur of Ephebopus cyanognathus, near the patella joint. The two arrows mark the location of the
“pedipalpal brush,” a region that consists of thousands of urticating hairs. A small area (asterisk) is devoid of such hairs, because they were
brushed out during a defensive reaction.

At the proximal end the hair shaft becomes very smooth
and decreases in diameter from 5 to 2-3 pm. It also exhibits a
slight bend where it disappears into the socket (Figs. 4, 8). The
angle at which the hair shaft emerges from the socket lies

between 25 and 30°, which means that the urticating hairs are
lying rather flat on the surface of the femur.

The sockets are slipper-shaped, about 10 pm long and 5-6 pm
wide, rising 3^ pm above the leg cuticle (Figs. 3, 4, 7). The

Figure 2. — The pedipalpal brush cut longitudinally: the barbed urticating hairs are seen from the side. Each hair arises from a slipper-shaped
socket (s) at an angle of 25-30°. Some urticating hairs have become detached from their sockets, and their free hair base (hb) is visible on the left.

294

THE JOURNAL OF ARACHNOLOGY

Figure 3. — A marginal region of the pedipalpal brush showing mostly empty sockets, because the urticating hairs had been brushed off in a
defensive reaction. A few hairs (on the right) are still held within their sockets, but the single hair on the left has been pulled out partially. Note
that there is no breaking point at the hair base.

opening for the hair shaft measures 3-4 pm, allowing a bit more
movement vertically than horizontally. The basal part of the hair
shaft fits snugly into its socket and ends at a basal ring (Figs. 9,
10). Below that hollow, circular structure a canal of about 5 pm
in diameter traverses the cuticle vertically (Figs. 4, 7, 9).

How is the urticating hair attached to its socket? It appears
that only the most proximal rim of the hair base is connected
to the basal ring, via a thin cuticular membrane. This
membrane can be seen in its original position in Fig. 9, and
remnants are often found attached to isolated hairs at the very

Figure 4. — Longitudinal section of a socket (s) and the detached base (arrowheads) of an urticating hair. The two asterisks mark the original
position of the hair base inside the socket. The exocuticle (exo) of an exuvium is only 5-10 pm thick.

Figure 5. — For most of its length, the hair shaft is barbed on the outside and hollow inside. The central lumen is only 1-2 pm wide and
exhibits small struts crossing the lumen.

Figure 6. — Section of the base of a large sensory hair showing fine membranous connections to the socket (black arrowheads). A much
stronger joint membrane (Jm) is responsible for the firm attachment of the hair shaft to the socket (white arrowheads).

FOELIX ET AL.— PALPAL URTICATING HAIRS OF TARANTULA

295

Figure 7. — A sagittal section of an empty socket of an urticating hair. The oblique upper part of the canal, which normally houses the base of
the hair shaft, ends in a horizontally running membrane (arrowhead). That is the spot where the hair base is attached laterally to a ring-like
structure (R); it appears as two small circles, because the ring is seen in cross-section (arrow). Further below, the canal continues in a vertical
direction through the exocuticle.

Figure 8. — Two hair bases of detached urticating hairs. Note the obliquely pointed endings reminiscent of a hypodermic needle. The part of
the hair shaft that is normally concealed within the socket has been shaded. Fragments of the attachment membrane (arrow) often still adhere to
the hair base.

base (Fig. 8). The hair shaft seems to be hollow and thus very
fragile at the attachment site; it quickly becomes more solid
distally, with a wall thickness of 2 pm and a central lumen of
1 pm (Fig. 5). The delicate connection via a membrane is
probably the key factor when the hair becomes dislodged from
its socket. A corresponding membrane connecting the hair
base to the socket is also present in the large sensory hairs
(Fig. 6). However, sensory hairs have an additional and much

stronger membrane, the joint membrane, that anchors the hair
shaft movably yet firmly in the socket.

It is noteworthy that we never found broken urticating
hairs. Any released urticating hair (detached either naturally
by the spider or artificially with a needle) remains in one piece.
We found no stumps of broken hair shafts inside the socket.
An exception is pictured in Fig. 9, but there the hair shaft had
been crushed by the razor blade when we cut the cuticle.

Figure 9. — Parasagittal section of the attachment of an urticating hair. In this case the hair shaft (hs) had been broken inadvertently when the
cuticle was cut with a razor blade. However, the attachment membrane (arrowhead) at the base of the socket is clearly visible. Note the narrow
diameter of the hair shaft inside the socket (2-3 pm), compared to a more distal portion (top of figure; 5-6 pm).

Figure 10. — A similar parasagittal section, but with the hair shaft (hs) still intact. The arrowhead marks the ending of the hair base and the
surrounding ring-like structure (R; cf. Fig. 7).

296

THE JOURNAL OF ARACHNOLOGY

Figure 11. — Three consecutive still pictures from a video film of the defensive reaction in Ephebopus cyanognathiis. Left: Immediately
following a mechanical stimulus, both palps (P) are held close together (arrowheads) and are touching the substrate. Middle: Only
40 milliseconds later, the first legs (1) and the palps (P) are rapidly moved upward (vertical arrow); at the same time, the chelicerae are spread
apart (horizontal arrows), thus exerting more friction upon the pedipalpal brushes on the femora. Right: The palps are now in an Up-Position
and kept much farther apart (arrow heads). It is at this point that a small puff of urticating hairs is released.

In order to determine how many urticating hairs there are
within a palpal brush, it is best to examine areas where the
urticating hairs have been brushed off and only the empty
sockets remain. The sockets are rather evenly spaced at a
distance of about 20 pm, a bit closer in the vertical direction
than in the horizontal (Fig. 3). On scanning electron
micrographs we counted 55 sockets on an area of 0.01 mm“,
which corresponds to a density of 5500 urticating hairs/mm".
Since the entire palpal brush measures around 1.5 mm^, the
total number of urticating hairs on one palp is about 8250.

How does Ephebopus actually release the urticating hairs
from its palpal brushes? Is it an interaction between the two
palps or between the palps and chelicerae? Since the defensive
reaction (palpal flicking) happens so fast, we used a video
camera and analyzed the movements of the involved body
parts in consecutive frames.

At the onset of the defensive reaction, the spider holds both
palps close together, touching the substrate (Fig. 11, left). About
30^0 ms later the first legs and the palps are rapidly thrust
upward and at the same time the chelicerae are spread sideways
(Fig. 11, middle). This apparently creates friction between the
inside of the femora (pedipalpal brush) and the outside of the
chelicerae. After 100 ms the palps are seen in the Up-position
and are held much farther apart than at the beginning (Fig. 1 1,
right). It is at this moment that a little puff of urticating hair is
released into the air. The entire defensive reaction consists of
only one upward stroke of the palps that lasts approximately
0. 1 s. Most of the time, the spiders are reluctant to repeat this
behavior and instead tend to flee.

DISCUSSION

The previous descriptions of the urticating hairs in Ephebopus
sp. (Marshall & Uetz 1990; West et al. 2008) were very brief and
did not give any dimensions. We found that most hairs are 500-
600 pm long but only 5-6 pm in diameter; thus the shape
corresponds more to a long knitting needle, which is in contrast
to the description ‘'short and stout” given by Marshall & Uetz
(1990). Only the rather long and thin urticating hairs, with a
length to diameter ratio of 1 00: 1 or 200: 1 , are considered to float
through the air (Bertani & Marquez 1995/96).

Barbs along the urticating hair shaft pointing distally have
been discussed with regard to other tarantulas, and it was argued
that the acute distal end could not be the penetrating tip (Bertani
et al. 2003; Perez-Miles 1998). This makes sense, since the barbs

would quickly get stuck in the skin if pushed in the wrong
direction. The basal end of the hair is less pointed but very thin
(2 pm) and squared off obliquely like a hypodermic needle. One
could well imagine that the basal end functions as the
penetrating tip, yet experimental proof is lacking so far.

Urticating hairs in tarantulas were classified into four
different types by Cooke et al. (1972), based on morphological
differences. Marshall and Uetz (1990) added a “fifth type” for
Ephebopus, but did not describe the typical features. Our
definition of type V urticating hairs would read as follows:
Straight hairs of 0.5 mm length and 5 pm diameter, slightly
bent at the base (socket region), fine barbs (0.5 pm) along the
entire hair shaft pointing toward the distal end. Overall, type
V hairs are similar to type II urticating hairs, except for the
distinct basal stalk of the latter.

Based on the density of empty sockets we calculated about
5500 urticating hairs for 1 mm^, or around 8250 hairs for one
palpal brush. This seems a modest number when compared to
over 10,000 hairs/mm“, as was claimed for abdominal
urticating hairs (Cooke et al. 1972). Unfortunately, the
authors did not provide any measurements for the diameter
of those hairs (type I), but deducing from their micrographs, it
should be around 7 pm. In order to achieve that high density
the hair sockets would have to be spaced at 10 pm or less.
Since this leaves almost no space between the hair shafts, it
would mean a veritable “solid forest” of urticating hairs. In
Ephebopus, sockets are spaced at 20 pm, which itself makes for
a very dense packing (Fig. 3).

It was surprising to find only small denuded areas (with
empty sockets) within the palpal brushes of the exuvia. It
could be that these spiders had hardly ever defended
themselves, or that they shed only a very limited amount of
urticating hairs during one palpal flick. It seems necessary, of
course, that Ephebopus be thrifty with its urticating hairs, since
they can grow back only between molts. In those animals
filmed repeatedly for their defensive reaction, the palpal
brushes showed large denuded patches. We estimated that
those spiders had reacted with a palpal flick approximately 5-
10 times. In one of these animals we found almost 5000 empty
sockets on the proximal side of each palpal brush. This would
correspond to more than 500 urticating hairs released with a
single palpal flick.

Another interesting aspect is that the palpal flick has to be a
controlled action; otherwise all urticating hairs might get lost

FOELIX ET AL.— PALPAL URTICATING HAIRS OF TARANTULA

297

Figure 12. — Cut-away view of the attachment of an urticating hair in Ephebopus cycmognathus. The hair is only held by a thin membrane (M)
between the hair base and a ring-like structure beneath the socket. This is the site of detachment when the urticating hair is released.

with one vigorous stroke. A mechanical feedback indicating
the number of urticating hairs could be transmitted by sensory
hairs that are interspersed among the palpal brush. These hairs
can hardly be seen in an intact hair field, but stand out in
denuded areas. In contrast to the urticating hairs, they remain
in their sockets after a palpal flick.

The main questions of this study were; how are those
urticating hairs attached, and how can they be detached? The
hair shaft tapers down to 2-3 pm near the base and then
disappears in a tightly fitting socket (Figs. 2, 3). Initially we
assumed that any forcible movement of the hair against the
rim of the socket would break the hair shaft right there and
leave a little stump inside the socket (Foelix et al. 2009).
However, this is not the case. All released hairs remain
practically in one piece and there are no stumps in the sockets.
So, how is the hair base actually connected inside the socket?
We were expecting to get an answer from longitudinal sections
of palpal brushes embedded in hard Epon. Unfortunately this
failed because during sectioning with the microtome the
embedding medium separates from the leg cuticle and
practically all urticating hairs are torn out (Fig. 4). We were
more successful by simply cutting strips out of the pedipalpal
brushes with a razor blade and then inspecting those longitudi-

nal sections under the scanning electron microscope. Again,
most sockets were empty due to the manipulation with the
razor blade, but if the sockets were only grazed by the blade,
the hairs remained in place (Figs. 9, 10). In such cases it was
clearly seen that the hair base is connected by a very thin
cuticular membrane to a ring structure lying horizontally
beneath the socket. This delicate attachment is apparently the
only connection of the urticating hair to the socket (Fig. 12).
In contrast, sensory hairs were found to be held by strong joint
membranes in their sockets; additionally, they also have a fine
connecting membrane at the very base of the hair (Fig. 6). It
thus seems that it is the lack of a joint membrane that makes it
easy for urticating hairs to become detached.

How does Ephebopus actually release its urticating hairs?
The only description available states that spiders ""were
observed to bring the pedipalps down across the basal segments
of the chelicerae in a brief scrubbing motion" (Marshall & Uetz
1990). Our analyses show that it is indeed a single motion;
however, it seems to be the upstroke of the palps against the
spread chelicerae that causes the release. The entire reaction
lasts only 0.1 s.

It is also interesting that there is no obvious counterpart
present on the chelicerae, such as marked ridges or a comb that

298

THE JOURNAL OF ARACHNOLOGY

would scrape the urticating hairs out of the pedipalpal brush. On
the contrary, the lateral surface of the cheliceral basal segment is
surprisingly smooth. Perhaps this is actually an advantage, so
that the barbed little “spears” will not get caught on the spider’s
body but can tly off into the air more easily.

Finally, we must raise the question of why Ephebopus is the
only genus so far known to have urticating hairs on the palps
rather than on the abdomen. For a tube-dwelling tarantula it
certainly makes sense to defend itself frontally toward an
aggressor, because any attack would usually occur from in
front. However, there are many other Neotropical tube-
dwelling tarantulas that first have to turn around and then
defend themselves by brushing off their abdominal urticating
hairs with their hind legs. And apparently they do so quite
successfully. Unfortunately, we do not know how and when
Ephebopus makes use of its urticating hairs under natural
conditions.

ACKNOWLEDGMENTS

Sam Marshall and George Uetz shared their experience with
Ephebopus urticating hairs with us. Gianni Sposato kindly lent
us his video camera. Martin Huber helped us with the
literature search and Jerome Rovner critically read the
manuscript. Benno Wullschleger and Jan de Vries assisted us
in technical matters. We are grateful to all these people.

LITERATURE CITED

Bates, H.W. 1863. The Naturalist on the River Amazons. John

Murray, London.

Bertani, R. & O.A.V. Marques. 1995/96. Defensive behaviors in
mygalomorph spiders: release of urticating hairs by some
Aviculariinae (Araneae, Theraphosidae). Zoologischer Anzeiger
234:161-165.

Bertani, R., T. Boston, Y. Evennou & J.P.L. Guadanucci. 2003.
Release of urticating hairs by Avicularia versicolor (Walckenaer,
1837) (Araneae, Theraphosidae). Bulletin of the British Arachno-
logical Society 12:395-398.

Castro, F.F.M., M.A. Antila & J. Croce. 1995. Occupational allergy
caused by urticating hair of Brazilian spider. Journal of Allergy
and Clinical Immunology 95:1282-1285.

Cooke, J.A.L., V.D. Roth & F.H. Miller. 1972. The urticating hairs of
theraphosid spiders. American Museum Novitates 2498:1-43.

Foelix, R., B. Rast & B. Erb. 2009. Uber die Brennhaare von
Ephebopus cyanognathus West & Marshall 2000. Arachne 14:4-13.

Marshall, S.D. & G.W. Uetz. 1990. The pedipalpal brush of
Ephebopus sp. (Araneae, Theraphosidae): evidence for a new site
for urticating hairs. Bulletin of the British Arachnological Society
8:122-124.

Perez-Miles, F. 1998. Notes on the systematics of the little known
theraphosid spider Hemirrhagus cervinus, with a description of a
new type of urticating hair. Journal of Arachnology 26:120-123.

Raven, R.J. 1985. The spider infraorder Mygalomorphae (Araneae):
cladistics and systematics. Bulletin of the American Museum of
Natural History 182:1-180.

West, R.C., S.D. Marshall, C.S. Fukushima & R. Bertani. 2008.
Review and cladistic analysis of the neotropical tarantula genus
Ephebopus Simon 1892 (Araneae, Theraphosidae) with notes on
the Aviculariinae. Zootaxa 1849:35-58.

Manuscript received 28 December 2008, revised 20 May 2009.

2009. The Journal of Arachnology 37:299-305

Possible niche differentiation of two desert wandering spiders of the genus Syspira (Araneae: Miturgidae)

Irma Gisela Nieto-Castaneda' and Maria Luisa Jimenez- Jimenez: Laboratorio de Aracnologia y Entomologia. Centro de
Investigaciones Biologicas del Noroeste (CIBNOR), Mar Bermejo 195, Col. Playa Palo de Santa Rita, La Paz, B.C.S.
23096, Mexico

Abstract. When species with similar morphological and ecological characteristics occupy the same habitat, selection
should minimize resource competition and promote coexistence by means of spatial partitioning. Competing species might
exploit resources at different times or specialize in distinct resources. From July 2005 through May 2006 we examined the
niche axes of two endemic sympatric desert species, Syspira tigrina Simon 1885 and Syspira longipes Simon 1885 in the State
of Baja California Sur, looking for evidence that coexistence is fostered by differences in choice of microhabitat, temporal
activity, occupation of space, or size. The results show high monthly microhabitat overlap (> 0.9). However, we found
subtle differences in temporal activity and marked differences in juvenile and male body size, as well as some evidence of
mutual spatial segregation. We conclude that body size and spatial segregation appear to be the dominant niche axes that
facilitate coexistence of these species.

Keywords: Ecological segregation, morphological segregation, niche overlap

A central goal of ecology is to understand the forces that
maintain species diversity within communities (Hutchinson
1959; Pacala & Tilman 1993; Chesson 2000). Hardin (1960)
suggested that sympatric similar species competing for the
same resources cannot stably coexist because one species will
always be more efficient than the others and will quickly drive
them to extinction. Consequently, coexistence requires some
form of resource partitioning between co-occurring species to
reduce or prevent interspecific competition (Amarasekare
2003). Hutchinson (1959), Chesson (2000), and Davies et al.
(2007) state that partitioning can occur in three ways. First,
species might differ in where they experience and respond to a
limiting factor (spatial habitat partitioning). Second, different
species may be limited by the same resources, but differ in the
time when they exploit the resource (temporal partitioning).
Third, co-occurring species may specialize in different
resources (resource partitioning). These kinds of partitioning
would be the result of selection for ecological character
divergence among sympatric populations (Brown & Wilson
1956; Dayan & Simberloff 2005; Davies et al. 2007).

Different types of segregation have been reported among
coexisting spiders. Among diurnal sympatric web-builders,
scientists have observed a clear microhabitat segregation
pattern in tetragnathids and linyphiids (Aiken & Coyle 2000;
Wright & Coyle 2000). Henaut et al. (2001) have reported prey
partitioning among araneids. Among nocturnal sympatric
araneids and tetragnathids, researchers have noted temporal
and spatial segregation (Ward & Lubin 1992).

Among diurnal wandering spiders, Uetz (1977) and Turner
& Polis (1979) found seasonal specialization to be the
predominant niche dimension facilitating coexistence of some
gnaphosoids, while spatial (Suwa 1986) and microhabitat
segregation (Moring & Stewart 1994; Carrel 2003) were the
key to coexistence in some lycosids. Cutler & Jennings (1992)

' Current address: Laboratorio de Biologia Comparada, Facultad de
Estudios Superiores Zaragoza, Universidad Nacional Autonoma de
Mexico, Batalla 5 de mayo s/n Esq. Fuerte de Loreto, Col. Ejercito de
Oriente, Iztapalapa, C.P. 09230, Mexico D.F., Mexico. E-mail:
ncig@puma2.zaragoza.unam.mx

found that habitat partitioning is common among congeneric
jumping spiders. Among syntopic, congeneric, nocturnal
wandering spiders, only ctenids have been studied, and the
results suggest that among Ctenus species there is no clear
niche partitioning (Gasnier & Hofer 2001), but that Cuppien-
nus species could be separated by differences in phenology on
a seasonal basis (Schuster et al. 1994).

In the desert of Baja California Sur, there are two sympatric
Syspira species: Syspira tigrina Simon 1885 and 5. longipes
Simon 1885 (Araneae: Miturgidae). These endemics (Olmstead
1975), are medium sized (8-12 mm) ground-dwelling spiders
that are active during the night. Pitfall trap collections indicate
that they represent up to 50% of all wandering spiders where
they are found, so they are an important component of the
desert ground spider assemblage (Nieto-Castaneda 2004). This
genus has not yet been used as a model in ecological studies.

By documenting small-scale patterns of sympatric populations
of S. tigrina and S. longipes, we wished to look for patterns of
microhabitat occupation, microhabitat overlap, spatial segrega-
tion, temporal activity, and size segregation over a span of one
year. Specifically, we asked the following questions: 1) What is the
structural microhabitat occupied by the two species? 2) Is there
evidence of overlap in microhabitat? 3) Are there indications of
mutual spatial exclusion? 4) Is there evidence of temporal
separation in activity? 5) Are there indications of size segregation?
The answers to these questions will help us understand how co-
occurrence of these two congeners takes place.

METHODS

Study area. — We selected four localities in the southern
extension of the Sonoran Desert (Leon de la Luz et al. 2000) to
represent a diversity of ground habitats, including three oases
with varying availability of water and one locality without a
water source. The oases are Presa de la Buena Mujer (24°05'N,
110°11'W, 180 m a.s.L), a reservoir; Laguna San Pedro
(23°56'N, 1 10°09'W, 6 m a.s.L), a lagoon on the Pacific coast;
El Novillo (23°55'N, 110°13'W, 220 m a.s.L), a small pond in
the hills, and El Comitan (24°07'N, 110°25'W, 20 m a.s.L), a
dry area (Fig. 1). The region is a subtropical desert with hot

299

300

THE JOURNAL OF ARACHNOLOGY

117°0'0"W 115°0'0"W 110"0'0"W 100°0'(J'W 90”0'0'W

Figure 1 . — Map of study sites where the two Syspira spider species
were collected. S: Laguna San Pedro, P: Presa de la Buena Mujer, E:
El Novillo, C; El Comitan.

summers and a sporadic rainy season between July and
October, and warm winters with little or no rain between
November and February. The vegetation is subtropical desert
shrub (Leon de la Luz et al. 1996).

Field work. — At each site, we plotted two belt transects
(100 m X 1 m), separated by 50 m. Transects were divided
into 20 quadrants (5 m X 1 m). We only sampled odd-
numbered quadrants to avoid disturbance to the contiguous
ones. We sampled the same quadrants every time because sizes
of monthly catches of Syspira did not decline over a year
(Nieto-Castaneda 2004) in a previous study of Baja Califor-
nia’s wandering ground spiders using pitfall traps. Exhaustive
hand collections with headlamps were made by a three-
member team, which spent one night at each locality every 3
mo from July 2005 through May 2006. This sampling pattern
included two rainy seasons (July and January) and two dry
seasons (October and May), totaling 16 collecting days.
Sampling started at dusk when Syspira spiders become active
and continued for 4—5 h, until spiders were no longer present.
The spiders were preserved in 70% ethanol.

Characterization of microhabitats. — When a spider was
found, we drew a 0.19 m^ circle (based on a 25 cm radius)
around it and measured ten variables to characterize the
microhabitat. The percentage of the following five ground
surfaces was identified: (1) bare soil, (2) fallen logs and
branches, (3) leaf litter and twigs, (4) gravel and pebbles (2-
64 mm dia.), and (5) cobbles (64-256 mm dia.). Next, we

Figure 2. — Dorsal view of carapace pattern of adults of S. tigrina
(A) and S. longipes (B).

immediately estimated the percentage of above-ground
vegetation (6). We then recorded (7) plant life forms (trees,
shrub, herbs), (8) soil texture [measured as categorical variable
1: 100% of sand, 2: <100, and >70% of sand, 3: <70% of
sand) by the wet feel method (Thien 1979)], (9) temperature,
and (10) relative humidity at the bare soil surface with a
thermo-hygrometer (Hl-8564, Hanna).

Species identification and measurements. — Al! spiders were
identified to species level using Olmstead (1975) and then
sexed if they were mature. We included the immature spiders
because they are necessary for any objective community
analysis (Sackett et al. 2008). These spiders, even young
specimens, are easily identified by the markings on the
carapace. S. tigrina has three dark stripe markings: two are
longitudinal and almost parallel, beginning near coxa I and
ending before the posterior edge of the carapace; these stripes
are separated by at least the distance between the anterior
lateral eyes. The third stripe is perpendicular to the other two
and is closer to the posterior edge of the carapace (Fig. 2A). S.
longipes has reticulated markings (Fig. 2B).

The total tibia I length and the carapace width of every
spider was measured as an indicator of body size (Hagstrum

NIETO & JIMENEZ— NICHE DIFFERENTIATION OF WANDERING SPIDERS

301

Table 1. — Number of juveniles (J), adult males (M), and adult females (F) of two Syspira species sampled each month in study.

July

October

January

May

Species

J

M

F

J

M

F

J

M

F

J

M

F

Total

S. tigrina

59

11

18

76

4

1

105

0

4

56

2

3

339

S. longipes

21

8

2

27

0

3

21

0

1

7

1

2

93

1971; Toft 1976). We did not measure total body length
because that can change quickly with alterations in foraging
success. Measurements were performed with a stereoscopic
microscope, using a micrometer. Voucher specimens were
deposited in the Arachnological and Entomological Collection
of the Centro de Investigaciones Biologicas del Noroeste
(CAECIB).

Data analysis. — Characterization of microhabitat: To test
whether species occupied different microhabitats with respect
to the month sampled, we used principal components analysis
(PCA) with varimax rotation of the correlation matrix. This
reduced the ten continuous microhabitat variables to a smaller
number of variables that explained most of the variation in the
raw data. Prior to conducting the PCA, we logio-transformed
temperature and relative humidity, after adding 1 to improve
normality and reduce heteroscedasticity. Then percentages
were converted to proportions and transformed by arcsine
square root (Goodman 2007). PCA was separately conducted
for each season, which generated four sets of PC axes. PC axes
with eigenvalues greater than 1.0 and eigenvectors with scores
greater than 0.7 were considered informative. The first two
principal components were plotted against each other to find
structure in the data that could distinguish S. tigrina
specimens from S. longipes specimens. All analyses were
performed with STATISTICA v. 6.0 software (StatSoft, Inc).

Overlap of microhabitat: Using the most informative
variables in PCAs, we calculated Pianka’s Index of microhab-
itat overlap by sampled month with the ECOSIM v. 7.0
software (Gotelli & Entsminger 2008). We then determined the
statistical significance of the observed microhabitat overlap by
comparing it with the RA3 algorithm, where the niche breadth
was retained and the zero states were reshuffled. In the
monthly presence-absence matrices, each row represented the
two Syspira species and each column represented a different
microhabitat category, in which the observed data on resource
utilization were randomized between the two species in 5000
simulations with proportional representations of the two
Syspira species and resources (Gotelli & Entsminger 2008).

Spatial segregation: With an abundance matrix by sampled
month in place, where rows represented the two Syspira
species, columns represented different quadrants (80 by
month, 320 in total), the co-occurrence module of ECOSIM
V. 7.0 was used to test for non-random patterns of species co-
occurrence. We calculated co-occurrence scores (C-score) as
the numbers of checkerboard units, based on 5000 interactions
with proportional representations of species and quadrant
sites. Species representations (rows) and quadrant location
representations (columns) were kept ‘proportional’ because
these conditions best reflected differences between species in
terms of trapping and spatial heterogeneity in trapping
probability. We calculated the expected C-scores (null models)
and subsequently tested for whether the occurrence of S.

tigrina and S. longipes deviated from random occurrence
(Gotelli & Entsminger 2008).

Temporal activity pattern: We used a Fisher exact test by
species to test for independence of spider abundance between
transects. We then performed a chi-squared goodness of fit
test to determine monthly differences in species abundance.
All analyses were performed with the Stata v. 9.1 software
(StataCorp, College Station, TX).

Size segregation: We also conducted a multivariate two-
group Hotelling’s T-squared test with Stata v. 9.1 software,
which tested for significant monthly size differences in the
carapace width and total tibia length between Syspira species
by developmental stage. Statistical significance was set at F <

Spiders collected. — During this study, we collected 432
Syspira spiders. Immature spiders represented almost 87% of
all Syspira spiders and were more abundant in January, while
adults were more abundant in July. S. tigrina was the most
abundant species in every month (Table 1).

Characterization of microhabitat. — PCAs with microhabitat
variables organized by month were used to characterize the
microhabitats occupied by each species. The first two compo-
nents of the four PCAs accounted for ~ 50% of the variation
(Table 2). S. longipes were always within the microhabitat
conditions occupied by S. tigrina. Positive scores for PCI
correlated with moist areas during July, October, and January
and dry areas during May. Negative scores for PCI were
associated strongly with cool areas in July, October and January
and warm areas during May. According to PCI, S. longipes was
restricted to the areas that were cooler and had higher relative
humidity than areas occupied by S. tigrina during July and
January and to the areas that were warmer and had lower
relative humidity occupied by S. tigrina during October and
May. During May, positive scores were linked with low sand soil
texture. There were slight differences in the structural micro-
habitat variables correlated with PC2. During July, positive
scores showed a relationship with higher bare soil surface,
negative scores were correlated with higher leaf litter and twigs
surface. During October and January, positive scores for PC2
were interconnected with high leaf litter and twig surface, and
negative scores correlated with high bare soil surfaces. During
May, PC2 was positively correlated with vegetation coverage
above ground and life form, but PC2 did not distinguish S.
longipes from S. tigrina (Fig. 3).

Overlap of microhabitat. — Monthly microhabitat overlap
for both species was high, indicating an almost complete
overlap. This result was significantly higher than expected by
chance (F < 0.05) (Table 3).

Spatial segregation. — The two species shared only a small
number of the 320 quadrants in all months, and most

302

THE JOURNAL OF ARACHNOLOGY

Table 2. — Correlations of ten structural microhabitat variables with the first two axes obtained from PC A (PCI, PC2) for each month. Bold
numbers indicate the most important variables. Eigenvalues and total variance explained are provided too.

Variable

July

October

January

May

PCI

PC2

PCI

PC2

PCI

PC2

PCI

PC2

Bare soil

0.38

0.86

-0.08

-0.91

0.26

-0.91

0.00

-0.06

Fallen logs and branches

-0.09

0.00

0.39

-0.28

0.05

0.19

0.02

-0.52

Leaf litter and twigs

-0.02

-0.84

0.01

0.90

0.14

0.95

-0.19

0.16

Gravel and pebbles

-0.48

-0.22

-0.17

-0.06

-0.73

-0.03

0.24

0.34

Cobbles

-0.31

0.01

0.50

0.12

-0.25

-0.07

0.12

-0.49

Above ground vegetation

0.56

-0.33

0.03

0.07

0.10

0.16

-0.29

0.69

Life form

0.21

-0.57

0.09

0.19

0.13

0.22

0.05

0.77

Soil texture

-0.65

-0.09

-0.39

0.21

-0.24

0.23

0.74

-0.19

Temperature

-0.76

0.15

-0.84

0.00

-0.83

0.03

0.95

0.01

Relative humidity

0.84

-0.11

0.88

0.10

0.88

-0.09

-0.91

0.04

Eigenvalue

2.55

1.98

2.38

1.89

2.70

2.17

2.92

1.70

% Total variance

25.50

19.77

23.84

18.93

27.00

21.69

29.16

17.03

Figure 3. — Structural microhabitat occupied by two Syspira species each month in two-dimensional ecological space based on principal
components scores (PCI and PC2).

Table 3. — Observed and expected Pianka’s overlap indices for each
month. Indices are given as the mean ± SD. Expected values are
based on 5000 interactions with proportional representation of
species and resources.

Month

Overlap Index

Observed

Expected

July

0.95

0.79 ± 0.08

October

0.97

0.75 ± 0.09

January

0.94

0.73 ± 0.09

May

0.92

0.87 ± 0.03

quadrants were exclusively occupied by one species. In all
months, C-scores were significantly higher than expected by
chance, which indicated interspecific spatial segregation {P <
0.05) (Table 4).

Temporal activity pattern. — We found that S. tigrina had
significantly different patterns between transects {^"3 = 20.79,
P < 0.05), but S. longipes did not; neither species had the same
monthly abundance pattern (/ ^3 = 4.21, T > 0.05) (Fig. 4).

Size segregation. — Juveniles of neither species had the same
average tibia and carapace size among months (July: F2J1 —
7.38, October: ^2,100 = 40.77, January: ^2,123 = 60.02, May:

NIETO & JIMENEZ— NICHE DIFFERENTIATION OF WANDERING SPIDERS

303

Table 4. — Frequency of quadrants (for each month = 80; total =
320) occupied by one, both, or neither Syspira species. Observed and
expected C-scores are given. Indices are given as the mean ± SD.
Expected values are based on 5000 interactions with proportional
representation of species and trap quadrants.

Species in quadrants

C-score

Month

None

Single

Both

Observed

Expected

July

31

41

8

310

111 ±41

October

32

34

14

208

96 ± 35

January

28

44

8

259

108 ± 46

May

40

36

4

155

59 ± 32

^2,60 = 12.96, P < 0.05). Females of neither species exhibited
significantly different tibia and carapace sizes in most months
(October: F2.1 = 0.02, January: ^2,2 = 0.23, May: F22 = 0.96,
P > 0.05) except in July (FVjy = 6.35, P < 0.05). Males of
both species had significantly different tibia and carapace sizes
in July (^2,16 = 20.10, P < 0.05), but in the other three months
the sample size was too small to test. S. longipes spiders had a
longer tibia I and a wider carapace than S. tigrina in all stages
and months (Fig. 5).

tfi

70 1

c

®

E

60

0

sx

50

m

Ib.

0)

40

■0

a

m

30

©

20

a»

n

E

10

3

z

0

S. tigrina

Jul Oct Jan May

Figure 4. — Monthly numbers of active Syspira species in two
transects (A, B) sampled at each site.

DISCUSSION

Large similarities in microhabitat occupancy indicate that
Syspira species should compete intensely (Holt et al. 1994),
with limited species overlap (Goodman 2007). Additionally,
members of the species should segregate in other dimensions,
as has been observed for other wandering spiders (Schuster et
al. 1994; Gasnier & Hofer 2001). In co-occurring, congeneric,
web-building spiders of the families Araneidae, Tetragnathi-
dae, and Linyphiidae in Nearctic regions, microhabitat
segregation appears to be a main factor allowing coexistence
(Aiken & Coyle 2000; Wright & Coyle 2000). This type of
segregation between congeneric species has been documented
in burrowing wolf spiders of the genus Geolycosa in Florida
(Marshall et al. 2000; Carrel 2003) and ground-wandering
Pardosa species in Japan (Suwa 1986). There is, however, no
evidence of habitat segregation among nocturnal, congeneric,
terrestrial cursorial spiders.

The constrained occupancy of the microhabitat of S. longipes
relative to S. tigrina, with respect to temperature and relative
humidity, suggests possible differences in their metabolism, since
these environmental factors affect spider performance (Huey &
Kingsolver 1989). Spatial arrangements of Syspira species
appear unrelated to leaf litter and twigs or bare surfaces, even
in desert communities that lack much structural complexity,
although these features are considered to be one of the most
critical microhabitat variables affecting community structure
(Melville & Schulte 2001; Goodman 2007).

We were not surprised to find significant spatial
segregation between these Syspira species during all months
since there are several studies suggesting that similar
sympatric spider species differ principally in spatial
distribution. For example, four wandering Ctenus species
in Central Amazonia segregate spatially (Gasnier & Hofer

2001) and four sympatric orb weavers (Araneidae and
Tetragnathidae) that inhabit coffee plantations in Mexico
reduce competition by building webs of varying structures
in different locations (Henaut et al. 2001). Congeneric
species of Pardosa have the same daily and seasonal
pattern, but segregate by vertical or horizontal stratification
(Greenstone 1980; Suwa 1986).

As expected, we found slight differences in temporal
segregation; however, it has not been well documented in
congeneric wandering spiders, although Turner & Polis
(1979) and Uetz (1977) stated that temporal segregation
was an important factor in reducing niche overlap. Yet this
type of segregation has been reported in other co-ocurring
species, namely Pardosa milvina (Hentz) with Hogna liellno
(Walckenaer) on soybean farms in Ohio (Marshall et al.

2002) . Ward & Lubin (1992) found that six nocturnal orb-
weavers (Tetragnathidae and Araneidae) occupied the same
habitat, but had different daily and seasonal activity
patterns.

Competition for food has long been considered a keystone
of community ecology. Therefore, differences in the average
monthly size of juveniles, and sometimes adults, of both
species may enable trophic divergence because body size is a
reliable determinant of prey size, non-web-building spiders
typically consuming prey of similar size to themselves
(Gertsch & Riechert 1976; Nentwig & Wissel 1986). The
significant overlap in average body size between females of

304

THE JOURNAL OF ARACHNOLOGY

o -

MFJ MFJ MFJ MFJ
July October January May
I I S. longipes I I S. tigrina

B

MFJ MFJ MFJ MFJ
July October January May
I I S. longipes I I S. tigrina

Figure 5. — Box plots of tibia I length (A) and carapace width (B) of both Syspira species. M: males, F: females, J: juveniles.

both species from October to May suggests that competition
for prey or other resources may be high. There are many
cases among animals that highlight the important role of
phenotypic divergence between sympatric species as a way to
avoid competition for food: bill size and shape in passerine
birds (Newton 1967); body size among amphibians, reptiles,
insects, and rodents; canine teeth diameter in carnivores
(Pimm & Gittleman 1990); and neck height or incisor arcade
structure in herbivorous mammals (Gordon & Illius 1988; Du
Toit 1990).

Differences in size may also be attributable to character
displacement (Guilleman et al. 2002; Dayan & Simberloff
2005). Such displacement occurs when selection, during
extended periods of sympatry among animals that partition
resources, results in an accumulation of morphological
differences that reduces or resolves competition. However,
such differences in morphology may have arisen before the
species came into sympatry, and these differences may have
been responsible for facilitating coexistence at its initial stages.
In either case, morphological distinctiveness is likely to be a
major contributor to stable coexistence of potential compet-
itors (York & Papes 2007). Few studies actually address
phenotypic divergence between sister sympatric species. Many
researchers assume that closely related species are more likely
to compete than distantly related ones (Dayan & Simberloff
2005). There are no studies in which phenotypic differences
among sympatric spider congeners are correlated as a
consequence of character displacement.

These two Syspira are closely related sympatric species that
have probably evolved similar life histories, and, since they use
the same microhabitats, have similar resource use. The spatial,
temporal (to a lesser degree), and size differences between the
two species may be the key factors permitting their coexistence.

ACKNOWLEDGMENTS

We thank F.A. Coyle, F Alvarez-Padilla, S.T. Alvarez-
Castaneda, Y. Henaut, and C. Blazquez for their suggestions;
I.H. Salgado-Ugarte for reviewing the statistical analyses; and
S. Pekar and two anonymous reviewers for the time and
concern spent on the review of this manuscript. M.M. Correa-
Ramirez and C. Palacios-Cardiel assisted during field collec-
tions, and I. Fogel provided valuable editorial improvements.
This project was supported by Consejo Nacional de Ciencia y
Tecnologia of Mexico (SEMARNAT-CONACYT COl-0052).
I.G. Nieto-Castaneda is a recipient of a graduate fellowship
from CONACYT. Collection, study, and preservation of
specimens comply with Mexican law, including a permit from
SEMARNAT.

LITERATURE CITED

Aiken, M. & F.A. Coyle. 2000. Habitat distribution, life history and
behavior of Tetragmitha spider species in the Great Smoky
Mountains National Park. Journal of Arachnology 28:97-106.
Amarasekare, P. 2003. Competitive coexistence in spatially structured
environments, a synthesis. Ecology Letters 6:1109-1122.

Brown, W.L. & E.O. Wilson. 1956. Character displacement.
Systematic Zoology 5:49-64.

Carrel, J.E. 2003. Ecology of two burrowing wolf spiders (Araneae,
Lycosidae). Journal of the Kansas Entomological Society
76:16-30.

Chesson, P. 2000. Mechanisms of maintenance of species diversity.

Annual Review of Ecology and Systematics 31:343-366.

Cutler, B. & D.J. Jennings. 1992. Habitat segregation by species of
Metaphidippus (Araneae, Salticidae) in Minnesota. Journal of
Arachnology 20:88-93.

Davies, T.J., S. Meiri, T.G. Barraclough & J.L. Gittleman. 2007.
Species co-existence and character divergence across carnivores.
Ecology Letters 10:146-152.

NIETO & JIMENEZ— NICHE DIFFERENTIATION OF WANDERING SPIDERS

305

Dayan, T. & D. Simberloff. 2005. Ecological and community-wide
character displacement, the next generation. Ecology Letters
8:875-894.

Du Toit, J.T. 1990. Feeding height stratification among African
browsing ruminants. African Journal of Ecology 28:55-61.

Gasnier, T.R. & H. Hofer. 2001. Patterns of abundance of four
wandering spiders (Ctenidae, Ctenus) in a forest in Central
Amazonia. Journal of Arachnology 29:95-103.

Gertsch, W.J. & S.E. Riechert. 1976. The spatial and temporal
partitioning of a desert spider community, with descriptions of new
species. American Museum Novitates 2604:1-25.

Goodman, B.A. 2007. Microhabitat separation and niche overlap
among five assemblages of tropical skinks. Australian Journal of
Zoology, Supplementary Series 55:15-27.

Gordon, I.J. & A.W. Illius. 1988. Incisor arcade structure and diet
selection in ruminants. Functional Ecology 2:15-22.

Gotelli, N.J. & G.L. Entsminger. 2008. EcoSim: Null models software
for ecology. Version 7. Acquired Intelligence Inc. & Kesey-Bear.
Jericho, Vermont. Online at http://www.garyentsminger.coni/
ecosim/index.htm

Greenstone, M.H. 1980. Contiguous allotopy of Pardosa ramidosa
and Pardosa tuoba (Araneae, Lycosidae) in the San Francisco Bay
Region, and its implications for patterns of resource partitioning in
the genus. American Midland Naturalist 104:305-311.

Guilleman, M., H. Fritz, N. Guillon & G. Simon. 2002. Ecomorphol-
ogy and coexistence in dabbling ducks, the role of lamellar density
and body length in winter. Oikos 98:547-551.

Hagstrum, D.W. 1971. Carapace width as a tool for evaluating the
rate of development of spiders in the laboratory and field. Annals
of the Entomological Society of America 64:757-760.

Hardin, G. 1960. The competitive exclusion principle. Science
131:1292-1297.

Henaut, Y., J. Pablo, G. Ibarra-Nunez & T. Williams. 2001.
Retention, capture and consumption of experimental prey by
orb-web weaving spiders in coffee plantations of southern Mexico.
Entomologia Experimentalis et Applicata 98:1-8.

Holt, R.D., J. Grover & D. Tilman. 1994. Simple rules for
interspecific dominance in systems with exploitative and apparent
competition. American Naturalist 144:741-771.

Huey, R.B. & J.G. Kingsolver. 1989. Evolution of thermal sensitivity
of ectotherm performance. Trends in Ecology & Evolution
4:131-135.

Hutchinson, G.E. 1959. Homage to Santa Rosalia or why are there so
many kinds of animals. American Naturalist 93:145-159.

Leon de la Luz, J.L., R.B. Coria & M.E. Cruz. 1996. Fenologia
reproductiva de una flora arido-tropical de Baja California Sur,
Mexico. Acta Botanica Mexicana 35:45-64.

Leon de la Luz, J.L., J.J. Perez-Navarro & A. Breceda. 2000. A
transitional xerophytic tropical plant community of the Cape
Region, Baja California. Journal of Vegetation Science
11:555-564.

Marshall, S.D., W.R. Hoeh & M.A. Deyrup. 2000. Biogeography and
conservation biology of Florida’s Geolycosa wolf spiders, threat-
ened spiders in endangered ecosystems. Journal of Insect Conser-
vation 4:11-21.

Marshall, S.D., D.M. Pavuk & A.L. Rypstra. 2002. A comparative
study of phenology and daily activity patterns in the wolf spiders

Pardosa milvina and Hogna helliio in soybean agroecosystems in
southwestern Ohio (Araneae, Lycosidae). Journal of Arachnology
30:503-510.

Melville, J. & J.A. Schulte. 2001. Correlates of active body
temperatures and microhabitat occupation in nine species of
central Australian agamid lizards. Austral Ecology 26:660-669.

Moring, J.B. & K.W. Stewart. 1994. Habitat partitioning by the wolf
spider (Araneae, Lycosidae) guild in streamside and riparian
vegetation zones of the Conejos river, Colorado. Journal of
Arachnology 22:205-217.

Nentwig, W. & C. Wissel. 1986. A comparison of prey lengths among
spiders. Oecologia 68:595-600.

Newton, 1. 1967. The adaptative radiation and feeding ecology of
some British finches. Ibis 10:35-98.

Nieto-Castaneda, LG. 2004. Aranas edaficas (Caponiidae, Gnapho-
sidae, Lycosidae, Miturgidae y Oonopidae (Arachnida: Araneae))
asociadas a humedales de Baja California sur, Mexico. M.Sc.
Thesis, Centro de Investigaciones Biologicas del Noroeste, La Paz,
B.C.S., Mexico.

Olmstead, J.V. 1975. Revision of the genus Syspira (Araneidae,
Clubionidae). M.Sc. Thesis, California State University, Long
Beach.

Pacala, S.W. & D. Tilman. 1993. Limiting similarity in mechanistic
and spatial models of plant competition in heterogeneous
environments. American Naturalist 143:222-257.

Pimm, S.L. & J.L. Gittleman. 1990. Carnivores and ecologists on the
road to Damascus. Trends in Ecology & Evolution 5:70-73.

Sackett, T.E., C.M. Buddie & C. Vincent. 2008. Relevance of
collected juveniles to the analysis of spider communities. Journal of
Arachnology 36:187-190.

Schuster, M., D. Baurecht, E. Mitter, A. Schmitt & F.G. Barth. 1994.
Field observations on the population structure of three ctenid
spiders {Cuppiennius, Araneae, Ctenidae). Journal of Arachnology
22:32-38.

Suwa, M. 1986. Space partitioning among the wolf spider Pardosa
amentata species group in Hokkaido, Japan. Researches on
Population Ecology 28:231-252.

Thien, S.J. 1979. A flow diagram for teaching texture-by-feel analysis.
Journal of Agronomic Education 8:54-55.

Toft, S. 1976. Life histories of spiders in a Danish beech wood.
Natura Jutlandica 19:5-40.

Turner, M. & G.A. Polis. 1979. Pattern of co-existence in a guild of
raptorial spiders. Journal of Animal Ecology 48:509-520.

Uetz, G.W. 1977. Coexistence in a guild of wandering spiders. Journal
of Animal Ecology 4:531-541.

Ward, D. & Y. Lubin. 1992. Temporal and spatial segregation of
web-building in a community of orb-weaving spiders. Journal of
Arachnology 20:73-87.

Wright, R.L. & F.A. Coyle. 2000. Habitat distribution, life history
and behaviour of Neriene species in the Great Smoky Mountains
National Park (Araneae, Linyphiidae). Bulletin of the British
Arachnological Society 1 1 :293-304.

York, H.A. & M. Papes. 2007. Limiting similarity and species
assemblages in the short-tailed fruit bats. Journal of Zoology
273:249-256.

Manuscript received 3 November 2008, revised 29 April 2009.

2009. The Journal of Arachnology 37:306-311

Construction and function of the web of Tidarren sisyphoides (Araneae: Theridiidae)

Ruth Madrigal-Brenes and Gilbert Barrantes: Escuela de Biologia, Universidad de Costa Rica, Ciudad Universitaria
Rodrigo Facio, San Jose, Costa Rica. E-mail: ruthymad@gmail.com

Abstract. In this paper, we describe the construction and function of the double sheet and tangle web of Tidarren
sisyphoides (Walckenaer 1842). Web construction includes several stages: construction of the scaffolding that serves to
support the rest of the web; filling in the dome-shaped and horizontal sheets; and construction of the upper tangle. During
construction of the scaffolding, the spider descends by a pre-existing thread to the substrate, moves a few centimeters and
attaches the dragline, then she ascends by the new thread, doubling the line or attaching it to another thread. The spider
fills in the sheet while walking in an irregular pattern under the sheet, and attaching her dragline using either one or both
legs IV simultaneously to hold pre-existing sheet lines against her spinnerets. During scaffolding construction and filling in
the dome-shaped sheet, the spider returns frequently to the retreat, apparently using the same threads near the retreat each
time. Threads of both the dome-shaped sheet and the horizontal sheet have small drops of viscid material. The dome-
shaped sheet and upper tangle comprise the functional trap of the web, while the horizontal sheet apparently plays only a
little role in prey capture.

Keywords: Web-building behavior, aerial sheet web, web function, viscid threads

Web designs in Theridiidae are strikingly variable (Szlep 1965,
1966; Lamoral 1968; Eberhard 1972, 1981, 1991; Agnarsson
2004, 2005, 2006; Eberhard et al. 2008a), and similar designs
have evolved independently in different genera, and in different
species within a genus (Darchen & Ledoux 1978; Eberhard 1991;
Japyassu & Jotta 2005; Barrantes & Weng 2006a, 2007; Jorger &
Eberhard 2006; Eberhard et al. 2008a). The broad disparity in
theridiid webs is possibly the result of their great flexibility in
microhabitat use, their ability to adjust web design to different
physical spaces, prey types, and prey availability (Turnbull 1964;
Eberhard 1990a; Agnarsson & Coddington 2007; J5rger &
Eberhard 2006; Eberhard et al. 2008b), and their response to
parasitism and predation pressures (Blackledge et al. 2003;
Agnarsson 2004; Barrantes et al. 2008).

Webs of theridiid spiders are sometimes described as an
irregular, three-dimensional structure (Foelix 1996). However,
their webs range from those that are extremely simplified as in
Phoroncidia studo Levi (Eberhard 1981), with a web consisting
of a single sticky line, to extremely complex, three-dimensional
webs with aerial sheets, as in Achaearanea disparata Denis
1965 (Darchen & Ledoux 1978) and Tidarren sisyphoides
(Eberhard et al. 2008a). Despite the diverse array of web
designs and the convergence in some of these designs, the
detailed descriptions of the web-building behavior have begun
to reveal some patterns in the typical behavior used to
manipulate lines and in the sequence of lines laid (Benjamin &
Zschokke 2003; Jorger & Eberhard 2006; Eberhard et al.
2008b). Knowledge of how three-dimensional webs of
theridiids are built is generally fragmentary (Szlep 1965,
1966; Lamoral 1968; Benjamin & Zschokke 2003), and limited
to only a few genera.

All webs described for species within the derived genus
Tidarren are tangles with aerial sheets (Agnarsson 2004;
Benjamin & Zschokke 2003; Eberhard et al. 2008a). Those of
T. sisyphoides (Walckenaer 1842) (Benjamin & Zschokke 2003)
and Tidarren spp. (see Agnarsson 2004) have been described as
lacking viscid threads. The sheet of T. sisyphoides is dome-
shaped, with a relatively dense tangle above it (Eberhard et al.
2008a). The spider hides in a retreat, often a curled leaf.

suspended in the tangle at the peak of the dome, opening onto
the underside of the dome (Eberhard et al. 2008a). Web
construction behavior has never been described in Tidarren. The
only report of construction of an aerial sheet web is for
Achaearanea tesselata (Keyserling 1884) (J5rger & Eberhard
2006). This study describes the web construction behavior of T.
sisyphoides, and the function of the areas of its web.

METHODS

We observed web construction behavior of 15 adult female
T. sisyphoides indoors in wire cubes 30 cm on a side, hanging
2 m above the floor from a thin fishing line. The cubes had a
wire along each of the diagonals at the top, and one along one
of the diagonals at the bottom. We collected the spiders with
their retreats on the campus of the Universidad de Costa Rica,
San Jose province (9°54'N, 84°03'W), Costa Rica. We hung
each retreat individually from the intersection of the two
diagonal wires at the top, using silk threads (4-5 cm long) of
the same web. Two spiders did not use the retreats and
constructed webs without them.

We photographed twelve webs each day for several days
(digital camera Olympus SP-510UZ), until each web was
completed. Webs were sprayed with water just before
photographing them to create a better contrast of silk threads
against the cubes’ backgrounds. We video recorded the
complete construction of three additional webs using a Sony
digital camera DCR-HC 96. Recording distance from the
spider was intentionally changed during web construction to
have either the entire cube in view or close-ups of different
construction behaviors. We searched for sticky droplets on
threads in five webs from the field. Thread samples from
sheets were collected on slides framed with strips of double-
sided adhesive tape, and density of viscid globules was
measured following Barrantes & Weng (2006b). We photo-
graphed viscid globules present in these threads under a
compound microscope (digital camera Nikon Coolpix 4500)
with a relative humidity of 60%. Viscid globules were then
placed in a saturated humidity chamber for 40 min and
observed under the dissecting microscope for changes in size.

306

MADRIGAL-BRENES & BARRANTES— WEB OF TIDARREN SISYPHOIDES

307

Figure 1. — Web of Tidarreti sisyphoides (powdered with talcum) showing the upper tangle, the dome-shaped sheet, and the horizontal sheet.

We videotaped the attack behavior on different prey types
to determine the possible function of the different sections of
the web. Each spider was offered a blow fly, a moth, a
grasshopper, a damselfly, a bee (Trigona spp.), a katydid, or a
leaf hopper every 2 days. Some prey items were placed directly
on the lower, horizontal sheet of some webs. We made
additional observations of the general shape of the web, prey
captured, and attack behavior of spiders in the field. Voucher
specimens of the spiders were deposited in the Museo de
Zoologia, Universidad de Costa Rica.

RESULTS

Webs of adult spiders, — Adult females of T. sisyphoides
constructed their webs mostly on large, solitary individuals or
small groups of Agave sp.. Yucca guatemalensis (Agavaceae) and
Monstera deliciosa (Araceae) plants scattered over the campus.
These plants all have large, relatively rigid leaves. Other plants
were seldom used. The webs consisted of a large tangle in which
there was a dense, upper dome-shaped sheet; a more or less
horizontal, much less dense sheet at the bottom; and a retreat at
the top of the dome-shaped sheet in the midst of the tangle
(Fig. 1). The dome-shaped and the horizontal sheets were very
loosely connected at their borders, and there was an empty space
without threads under the dome in which the spider moved freely
during prey capture. The dense, irregular tangle above the dome-
shaped sheet connected the dome to the substrates or to thick,
multiple threads suspending the retreat. The border of the dome-
shaped sheet was also connected to the substrates nearby (wire
frame, or leaves and twigs in the field). The horizontal sheet was
rarely connected to leaves or other substrates (3 out of 23).

Web construction, — T. sisyphoides {n = 15) began construc-
tion of the web between 1730 to 1830 h and ended the night’s

work at about 530 h next day (n = 5). Spiders took from one
to four nights to construct a complete, functional web,
although some additional threads were certainly added
subsequently. The time spent in building decreased over
successive nights. The first night the spiders were nearly
continually active, spinning different parts of their webs, but
on subsequent nights they began later (between 21 and 23 h),
had longer pauses, and finished earlier (usually at 2 or 3 h).
The spiders’ only construction-oriented diurnal activity was to
secure the retreat to the wire frame soon after the retreat was
first placed in the wire frame.

Web construction can be roughly divided into five different
stages, some of them not being mutually exclusive: explora-
tion, suspension of the retreat, construction of the scaffolding,
construction of the dome-shaped sheet, and construction of
the lower horizontal sheet. The spider walked underneath silk
lines at all times during construction. She held her dragline
with the tarsus of one leg IV, frequently switching the leg IV
that held the dragline.

Exploration: The spider began the construction of the web
by exploring the wire cube. She climbed up to the frame along
the threads that secured the retreat, then walked along the
horizontal and vertical wires of the frame, attaching her
dragline at irregular intervals and occasionally returning to the
retreat. Sometimes the spider descended beyond the wire
frame, from 30 cm to nearly one meter, hanging from her
dragline, and then climbing back up the dragline to the frame.
While ascending, the spider sometimes packed the slack
dragline into a mass, and a small white mass was observed
near the point where she reached the frame. More frequently
she did not reel up the dragline, and attached a loop, or
sagging threads to the wire. Occasionally the spider descended

308

THE JOURNAL OF ARACHNOLOGY

a second time. All spiders but one did not descend in this way
from the frame during the exploration stage. The complete
exploratory phase lasted 15-30 min.

Anchoring the retreat: After exploration, the spider began to
reposition the retreat. First the spider walked up the line
supporting the retreat and along a horizontal wire of the upper
wire frame, away from the retreat, until she reached a vertical
wire. The spider then descended a few centimeters along the
vertical wire and attached the thread from the retreat to the
wire. The spider often reinforced this line by walking back to
the retreat on the same thread, doubling it. Some of these
threads were attached to the wire frame above the retreat, but
others were attached to the vertical wires either at the level of
the retreat or a few centimeters below it. Then the spider broke
threads attached to the upper section of the retreat, causing it
to drop approximately 1 cm. This sequence was repeated
several times until the retreat was moved up to nearly 10 cm
downward from its original position, and was reoriented so
that its opening was directed downward. The broken lines
were occasionally packed. In these cases, the spider moved
along another line while reeling up the cut line. She packed the
loose line with her legs II and III, and then attached the
whitish mass of silk to the wire frame or to another thread.

With the retreat in position, the spider began to spin threads
from the top of the retreat to the upper wires, within the nearest
5 cm of the crisscrossing point of the diagonal wires. These
threads were frequently reinforced by the spider walking back
and forth, up to five times, on the same threads between the
retreat and the furthest attaching point, forming thick cables
that were clearly distinguishable from other threads. During
construction of this cable, the spider was frequently observed
attaching the new threads to those previously made. Construc-
tion of other sections of the web did not begin until the retreat
was securely suspended from the upper section of the wire frame.

Construction of the scaffolding: Immediately after suspend-
ing the retreat, the spider began to construct the scaffolding
for the dome-shaped sheet. She first spun threads that
extended from the retreat opening, or near to it, to the wire
frame. Additionally, she spun threads from some point along
these threads to the wire frame, so that only five to six (n = 2
webs) threads converged at (or near) the retreat opening.
These threads were then interconnected, forming a roughly
conical scaffolding just below the retreat.

To spin the first threads of the conical scaffolding, the
spider walked along one of the threads from which the retreat
was suspended and then descended by one of the vertical
wires. She then either attached her dragline to the vertical wire
or continued to descend to the horizontal, bottom wire frame
where she attached the dragline, touching her spinnerets
repeatedly on the side of the wire facing the web or on the side
away from the web. Once the thread was attached, the spider
ascended by the thread she just had created and attached the
new thread to it, producing a double line, or else she attached
the new thread to another thread she encountered on the way
up, usually a few centimeters away from the retreat opening.
Only rarely, this second thread was attached at the retreat
opening. After some lines were present below the hub, the
spider descended by a previous thread, walked 2-A cm along
the wire, attached her dragline, and ascended by this new
thread. This new thread was sometimes attached to the thread

Figures 2, 3. — Behavioral sequences during web construction. 2.
Placement of threads during the scaffolding construction. The numbers
(la-5a and lb-4b), arrows, and dotted lines mark the sequence and
direction of movements of the spider during lines placement. Dashed
lines indicate the pre-existing threads and solid lines indicate newly
placed threads. 3. Two different paths of the spider as she filled in the
dome-shaped sheet (100 s each traced from video images recorded
looking approximately perpendicular to the plane of the sheet). Black
dots indicate the position of the spider every 5 s.

she ascended, producing a double thread (Fig. 2: Ib^b), or
others to another thread (Fig. 2: la-5a).

When the spider had spun most lines (lines were difficult to
observe and were not counted) forming the conical scaffolding,
she connected these lines, and also connected them to the lines
suspending the retreat. She also spun additional lines connecting
the middle part of the retreat to pre-existing lines. The lines
connecting the scaffolding of the dome-shaped sheet to the
retreat suspension lines and to the upper section of the wire cube
constituted part of the upper tangle. Video recordings showed
that when attaching the dragline to another line, the spider held
the dragline with one leg IV, while ipsilateral legs III and IV
grasped the other line, bringing it toward the spinnerets at the
same time that she bent her abdomen ventrally toward the line to
touch it with her spinnerets.

Construction of the dome-shaped sheet, the horizontal sheet,
and upper tangle: After the spider had constructed the
scaffolding, she filled in the dome-shaped sheet. The process
of filling in this sheet alternated with the construction of the

MADRIGAL-BRENES & BARR ANTES— WEB OF TIDARREN SISYPHOIDES

309

horizontal sheet and with the upper tangle. All spiders
constructed the dome-shaped sheet in two phases: first the
spider wove a complete but sparse dome-shaped sheet; then
she filled in the spaces in this sheet. The sparse dome-shaped
sheet was constructed in the first (« = 4) or second night (n =
!1). During the second and third nights, the spiders increased
the density of the dome-shaped sheet and of the threads of the
upper tangle, which mostly consisted of threads connecting the
dome-shaped sheet with the wires above. Construction of the
horizontal sheet did not begin until the dome-shaped sheet had
been partially built. The horizontal sheet was much less
densely woven (Fig. i).

The spider spent 1-3 min filling in a relatively small section of
the dome-shaped sheet, then moved to a different section,
sometimes on the opposite side of the dome, or sometimes
nearby. After filling in a section, the spider often went up to the
retreat, tapped the egg sac, then moved away to the next web
section to fit! in. The repeated visits to the retreat did not
increase the number of threads converging at its mouth (n = 2
webs). During the filling in activity, the spider walked under the
sheet rapidly forward, and sideways in an irregular pattern
(Fig. 3), while tapping actively with both legs I. We did not see
individual threads in all cases, but based on the spider’s behavior
in video analyses, the spider did not attach her dragline to all
threads she came in contact with, since she walked several
millimeters and frequently several centimeters without attaching
her dragline. During the dragline attachments, the spider
displayed two different movements: in one, she held the dragline
with one leg IV, while ipsilateral legs III and IV grasped the sheet
line, and brought it toward the spinnerets; in the other, the
spider’s two legs IV grasped the sheet simultaneously on either
side of her spinnerets while her abdomen bent ventrally toward
the lines and no leg held the dragline. We clearly observed both
types of attachment behaviors in the construction of both the
dome-shaped and the horizontal sheets.

Most spiders had constructed the scaffolding (14 out of 15)
and part of the dome-shaped sheet by the end of the first night.
Only four spiders constructed a complete web during the first
night. By the end of the third night, all but one spider that
never constructed the horizontal sheet had complete webs. All
spiders added more threads, primarily to the dome-shaped
sheet and to the tangle above it in subsequent nights. Filling in
the dome-shaped sheet consumed most of the construction
time of the spider (about 70%) on subsequent nights.

Viscid balls. — Viscid globules were present on threads of
both the dome-shaped and the horizontal sheets in all webs
examined (Fig. 4). Globules measured 58.5 ± 30.8 x 52.0 ±
29.3 pm (n = 20 globules, 5 webs) on dome-shaped sheet lines
and 100.0 ± 62.0 X 88.3 ± 56.7 pm on lines in the horizontal
sheet (n = 6 globules, 2 webs). Their mean density was lower
in the horizontal sheet (0.94 balls/mm, SD = 0.62; 2 webs;
26 mm of thread sampled) than in the dome-shaped sheet (1.5
balis/mm, SD = 1 .3; 4 webs; 22 mm of thread sampled).
Globules were hydrophilic and increased in size in a humid-
saturated environment.

Dissecting function of the web. — In nature, seven flies in at
least three different families, five treehoppers (Membracidae),
two beetles (one Scarabaeidae, one Chrysomelidae) and one
honey bee were found in webs (n= 22). Most prey that were
dropped on webs were retained for several seconds in the upper

sisyphoides. The globule is on a pair of core fibers.

tangle (27 out of 30) before dropping to the dome-shaped sheet.
The spider sensed the prey as soon as it contacted the upper
tangle, first orienting inside her retreat (this was not possible to
observe in all cases) and then moving to the area of the dome
sheet beneath the struggling prey. There she pulled some
threads, turned a few degrees, and pulled other threads until
the prey contacted the dome-shaped sheet. On one occasion, the
spider broke the threads of the dome-shaped sheet and climbed
up to attack the prey in the upper tangle. Prey that fell to the
dome-shaped sheet were generally constrained until the spider
arrived, but in a few cases, large, strong, struggling insects (e.g.,
katydids) broke free from the upper tangle and dome-shaped
sheet. These prey hit the horizontal sheet, but were not trapped
there long enough for the spider’s attack. Prey that were placed
directly on the upper face of the horizontal sheet (n = 7) were
not restrained long enough to allow the spider attack the prey.

Attack behavior. — Attacks began by applying viscid threads
on to the prey with both simultaneous and alternate
movements of legs IV. If the prey was dangerous (e.g.,
katydids), viscid threads were applied from farther away than
to flies or moths. Wrapping continued until prey was
immobilized, at which point it was bitten. In 83% of 72
spider-prey encounters, the spider retired to her retreat and
returned to the prey after the prey’s movements had subsided.
When the prey was large, the spider cut it free from the dome-
shaped sheet before continuing the wrapping attack as it hung
on a few lines below the level of the dome. If prey’s movements
had not completely subsided when the spider returned from
her retreat, the prey was wrapped and bitten again. Then it
was carried, dangling from one leg IV to near the retreat where
it was wrapped some more and attached to the threads near
the mouth of the retreat.

DISCUSSION

The complex aerial-sheet web of T. sisyphoides seems to be
unusual in several respects among theridiids. Several other

310

THE JOURNAL OF ARACHNOLOGY

theridiids (e.g., Anelosimus spp., Chrosiothes portalensis,
Achaeanmea tesselata, A. disparata, A. japonica) also construct
webs with horizontal or bowl shaped aerial-sheets (Darchen
1968; Eberhard 1972; Darchen & Ledoux 1978; Eberhard et al.
2008a), but never a dome-shaped sheet as in T. sisyphoides
(and also some webs of T. haemorrhoidale Eberhard et al.
2008a). The aerial sheets have most likely evolved indepen-
dently in these theridiid lineages, as indicated by a recent
phylogenetic study that showed an extremely high flexibility in
web-building behaviors and high convergence in web features
among theridiids (Eberhard et al. 2008a). However, the dome-
shaped sheet and the presence of a horizontal sheet connected
to the border of the dome-shaped sheet (Fig. 1) seem to be
unique features of the Tidanen genus; a horizontal sheet
connected to the dome-shaped sheet has only been found in T.
sisyphoides. The absence of similar elements in webs of other
theridiid species (Agnarsson 2004; Eberhard et al. 2008a)
suggests that, at least, some elements of the T. sisyphoides' web
represent an independent and unique event in the evolution of
webs in Theridiidae.

Despite the unusual design of the webs of T. sisyphoides, there
are several general behavioral patterns in the web construction
that resemble those behaviors of other theridiid species that have
quite different webs. T. sisyphoides explored prior to initiating
web construction, constructed its web only at night, constructed
a scaffold that supports the rest of the web, alternated
construction of different sections of the web, held its dragline
with one leg IV, doubled lines during the scaffold construction,
and added new threads and repaired the web over many
subsequent nights. These behavior patterns are similar to those
of other species of Theridiidae: Latrodectus, Steatoda, Theridion,
and Achaearaea (Szlep 1965; Eberhard 1982; Benjamin &
Zschokke 2002, 2003; Jorger & Eberhard 2006; Barrantes &
Weng 2007; Eberhard et al. 2008b), indicating that they are
widespread within theridiids. Similar behaviors occur in other
spider families. For instance, holding the dragline with one leg
IV, alternating construction of different parts of the web, and
doubling threads has also been described for other Orbiculariae
(Eberhard 1990b). Descending from the retreat (or near to it)
along a pre-existing thread while putting out a new line, walking
on the substrate, attaching this new line to the substrate, and
then ascending by this new line to return to the retreat (or near
to it) during the scaffolding construction is another behavior
that has also been described for Steatoda triangulosa and A.
tepidariorum (Benjamin & Zschokke 2002, 2003). This order of
thread placement is similar to, though less stereotypical of,
radius construction in the Nephilidae and Uloboridae (Eberhard
1982; Kuntner et al. 2008). However, further phylogenetic based
studies are necessary to determine whether these behaviors are
homologous between Theridiidae and other Orbiculariae.

The detailed description of web-construction by A. tesselata
(Jorger & Eberhard 2006), allows us to further compare the
construction behavior between this species and T. sisyphoides.
Both species strengthened the lines holding up the retreat prior
to initiating construction of the web. Securing the retreat first
is likely due to the fact that in both species the spiders that
were observed had either egg sacs or spiderlings in their
retreats; in nature, these spiders first construct a web, and then
collect a curled leaf or other plant debris to construct the
retreat. Attaching the anchor and scaffolding lines to the far

side of objects that likely make attachments more secure,
occurs also in A. tesselata and some orb-weaver araneoids
(Jorger & Eberhard 2006; Eberhard 1990b; Eberhard 2001).
Breaking and releasing threads is frequent during some phases
of the web construction of these two species, as well as in S.
triangulosa (Benjamin & Zschokke 2002) and L. geometricus
(Eberhard et al. 2008b). This behavior may be at least partially
explained by an inability of theridiids to digest silk, but it is
also possible that loose threads might increase prey retention
in the web (Kirchner 1986; Blackledge et al. 2008). Break and
reel behavior was not observed in T. sisyphoides, though it
occurs in A. tesselata (Jorger & Eberhard 2006) during
exploration, and in A. tepidariorum (W. Eberhard pers.
comm.), and L. geometricus during gum foot line construction
(Eberhard et al. 2008b).

Sheet construction by T. sisyphoides also resembled that of
A. tesselata. The spider walked under silk lines while
constructing the sheet, filling in different parts of the sheet
in no apparent order (perhaps more detailed observations
might establish some pattern). Attachments of the dragline
were similar in both species: the spider used either one or both
legs IV to hold sheet threads when she attached her dragline
during filling in behavior. Both species filled in the sheet with
apparently erratic wandering movements, although they were
apparently more regular in T. sisyphoides. Both species often
returned to the retreat during filling-in behavior, presumably
using lines previously laid in the near vicinity of the retreat.
This behavior results in only a few lines converging at the
mouth of the retreat, and explains the ability of the spider to
orient inside the retreat toward the prey in the web before
launching an attack (Barrantes & Weng 2006a; Jorger &
Eberhard 2006). Having few threads converging at the retreat
is also a feature of newly constructed webs of several
Latrodectus (Szlep 1965; Eberhard et al. 2008b).

Some general behaviors (e.g., construction of scaffolding,
expansion of web over time) are widely spread within
theridiids. However, some other traits such as the presence
of an aerial sheet in the web have probably evolved
independently several times within Theridiidae (Jorger &
Eberhard 2006; Agnarsson 2004; Eberhard et al. 2008a), and
other families (e.g., Linyphiidae, Pholcidae, and Synotaxidae-
Chileotaxus sans) possibly as a result of using similar habitats,
capturing similar prey types (Wise 1982), and predation and
parasitism pressure (Blackledge et al. 2003; Agnarsson 2004).

The dome-shaped sheet and the tangle above it (upper
tangle) seem to function as the trapping section of the web.
The upper tangle probably functions to knock down jumping
and flying insects that are then restrained by viscid elements in
the dome-shaped sheet, as indicated by the insect types found
in nature. The horizontal sheet at the bottom of the web seems
to have little effect in prey retention; perhaps it serves as a
barrier to reduce attacks of predators and parasitoids (Lubin
1986; Blackledge et al. 2003). Viscid balls have not previously
been reported in webs of species in this genus (Benjamin &
Zschokke 2003; Agnarsson 2004; Eberhard et al. 2008a). The
viscid balls of gum foot lines and the viscous wrapping silk in
theridiids are apparently produced by the aggregate glands
(Kovoor 1977; Coddington 1989). However, until the origin of
the core axial fiber on which T. sisyphoides place the viscid
balls is clearly established, it will be possible to determine

MADRIGAL-BRENES & BARRANTES— WEB OF TIDARREN SISYPHOIDES

311

whether these viscid threads are homologous to the gum foot
lines or other sticky threads of other theridiid webs (Eberhard
et al. 2008a).

ACKNOWLEDGMENTS

We thank William Eberhard for his helpful comments on
the manuscript and Emilia Triana for help feeding the spiders.
This study was supported by the Vicerrectoria de Investiga-
cion, Universidad de Costa Rica.

LITERATURE CITED

Agnarsson, I. 2004. Morphological phylogeny of cobweb spiders and
their relatives (Araneae, Araneoidea, Theridiidae). Zoological
Journal of the Linnean Society 141:447-626.

Agnarsson, 1. 2005. A revision and phylogenetic analysis of the
American ethicus and rupmuni groups of Anelosimus (Araneae,
Theridiidae). Zoologica Scripta 34:189-314.

Agnarsson, 1. 2006. A revision of the New World eximius lineage of
Anelosimus (Araneae, Theridiidae) and a phylogenetic analysis
using worldwide exemplars. Zoological Journal of the Linnean
Society 146:453-593.

Agnarsson, 1. & J.A. Coddington. 2007. Notes on web and web
plasticity and description of the male of Achaearanea heiroglyphica
(Mello-Leitao) (Theridiidae, Araneae). Journal of Arachnology
34:638-641.

Barrantes, G. & J. Weng. 2006a. The attack behavior of Achaearanea
tesselata (Araneae: Theridiidae). Journal of Arachnology
34:45^66.

Barrantes, G. & J. Weng. 2006b. Viscid globules in webs of the spider
Achaearanea tesselata (Araneae: Theridiidae). Journal of Arach-
nology 34:480M82.

Barrantes, G. & J. Weng. 2007. The behavior and natural history of
Theridion evexum (Araneae, Theridiidae). Bulletin of the British
Arachnological Society 14:61-65.

Barrantes, G., W.G. Eberhard & J. Weng. 2008. Seasonal patterns of
parasitism of the tropical spiders Theridion evexum (Araneae,
Theridiidae) and AUocyclosa bifurca (Araneae, Araneidae) by the
wasps Zatypota petronae and Polysphincta gutfreundi (Hymenop-
tera, Ichneumonidae). Revista Biologia Tropical 56:749-754.
Benjamin, S.P. & S. Zschokke. 2002. Untangling the tangle-web: web
construction behavior of the comb-footed spider Steadota trian-
gulosa and comments on the phylogenetic implications (Araneae:
Theridiidae). Journal of Insect Behavior 15:791-809.

Benjamin, S.P. & S. Zschokke. 2003. Webs of theridiid spiders:
construction, structure and evolution. Biological Journal of the
Linnean Society 78:293-305.

Blackledge, T., J.A. Coddington & R.G. Gillespie. 2003. Are three-
dimensional spider webs defensive adaptations? Ecological Letters
6:13-18.

Coddington, J.A. 1989. Spinneret silk spigot morphology: evidence
for the monophyly of orbweaving spiders, and the group
Theridiidae plus Nesticidae. Journal of Arachnology 17:71-95.
Darchen, R. 1968. Ethologie d’ Achaearanea disparata Denis, Aranea,
Theridiidae, Araignee sociale du Gabon. Extrait de la Revue
Biologia Gabonica 4:5-25.

Darchen, R. & C. Ledoux. 1978. Achaearanea disparata, araignee
sociale du Gabon, synonyme ou espece jumelle de Achaearanea
tessellata, solitaire. Revue Arachnologique 1:121-132.

Eberhard, W.G. 1972. Observations on the biology of Achaearanea
tesselata (Araneae: Theridiidae). Psyche 79:209-212.

Eberhard, W.G. 1981. The single line web of Phoroncidia stiido Levi
(Araneae: Theridiidae): a prey attractant? Journal of Arachnology
9:229-232.

Eberhard, W.G. 1982. Behavioral characters for the higher classifi-
cation of orb-weaving spiders. Evolution 36:1067-1095.

Eberhard, W.G. 1990a. Function and phylogeny of spider webs.
Annual Review of Ecology and Systematics 21:341-372.

Eberhard, W.G. 1990b. Early stages of orb construction by
Philoponella vicina, Leiicaiige mariana and Nephila clavipes
(Araneae, Uloboridae and Tetragnathidae) and their phylogenetic
implications. Journal of Arachnology 18:205-234.

Eberhard, W.G. 1991. Chrosiothes tonala (Araneae, Theridiidae): a
web-building spider specializing on termites. Psyche 98:7-19.

Eberhard, W.G. 2001. Under the influence: webs and building
behavior of Plesiometa argyra (Araneae, Tetragnathidae) when
parasitized by Hymenoepimecis argyraphaga (Hymenoptera, Ich-
neumonidae). Journal of Arachnology 29:354-366.

Eberhard, W.G., 1. Agnarsson & H.W. Levi. 2008a. Web forms and
the phylogeny of theridiid spiders (Araneae: Theridiidae): chaos
from order. Systematics and Biodiversity 6:415-475.

Eberhard, W.G., G. Barrantes & R. Madrigal-Brenes. 2008b. Vestiges
of an orb-weaving ancestor? The “biogenic law” and ontogenetic
changes in the webs and building behavior of the black widow
spider Latrodectus geometricus (Araneae: Theridiidae). Ethology
Ecology and Evolution 20:211-244.

Foelix, R.F. 1996. Biology of Spiders. Oxford University Press, New
York.

Japyassii, H.F. & E.G. Jotta. 2005. Forragemento em Achaearanea
cinnabarina Levi 1963 (Araneae, Theridiidae) e evolucao da caga
em aranhas de teia irregular. Biota Neotropica 5:1-15.

Jorger, K. & W.G. Eberhard. 2006. Web construction and
modification by Achaearanea tesselata (Araneae: Theridiidae).
Journal of Arachnology 34:511-523.

Kirchner, W. 1986. Das Netz der Zitterspinne (Pholcus plialangioides
Fuesslin) (Araneae: Pholcidae). Zoologischer Anzeiger
216:151-169.

Kovoor, J. 1977. Donnees histochimiques sur les glandes sericigenes
de la veuve noire Latrodectus mactans Fabr. (Araneae, Theridi-
idae). Annales des Sciences Naturelles, Zoologie 19:63-87.

Kuntner, M., J.A. Coddington & G. Hormiga. 2008. Phylogeny of
extant nephilid orb-weaving spiders (Araneae, Nephilidae): testing
morphological and ethological homologies. Cladistics 24:147-217.

Lamoral, B.H. 1968. On the nest and web structure of Latrodectus
South Africa and some observations on body colouration of
Latrodectus geometricus (Araneae, Theridiidae). Annals of the
Natal Museum, Pietermaritzburg 20:1-14.

Lubin, Y. 1986. Web building and prey capture in the Uloboridae.
Pp. 132-171. In Spiders: Webs, Behavior, and Evolution. (W.A.
Shear, ed.). Stanford University Press, Stanford, California.

Szlep, R. 1965. The web-spinning process and web-structure of
Latrodectus tredecimguttatus, L. pallidus and L. revivensis. Pro-
ceedings of the Zoological Society of London 145:75-89.

Szlep, R. 1966. The web structure of Latrodectus varioliis Walckenaer
and L. bishopi Kaston. Israel Journal of Zoology 15:89-94.

Turnbull, B.A. 1964. The search for prey by web-building spider
Achaearanea tepidariorum (C. L. Koch) (Araneae, Theridiidae).
Canadian Entomologist 96:568-579.

Wise, D.H. 1982. Spiders in Ecological Webs. Cambridge University
Press, Cambridge, UK.

Manuscript received 3 March 2009, revised 21 June 2009.

2009. The Journal of Arachnology 37:312-320

Scorpion taphonomy; criteria for distinguishing fossil scorpion molts and carcasses

Victoria E. McCoy and Danita S. Brandt': Department of Geological Sciences, Michigan State University,

206 Natural Science Building, East Lansing, Michigan 48824, USA

Abstract. The ability to distinguish fossil arthropod carcasses from their molts is necessary for a more complete
understanding of the arthropod fossil record and for more accurately assessing the role of fossil arthropods in
paleoecosystems. Taphonomic characteristics, e.g., recurrent patterns of disarticulation of exoskeletal elements, are the
primary data that have been used to differentiate fossil exuvia and fossil carcasses among arthropods. This study
documents recurrent taphonomic patterns in modern scorpion carcasses and molts and extends these patterns to the fossil
record to define criteria by which fossil scorpion molts might be distinguished from fossil scorpion carcasses. The three
most useful and statistically significant characters in making the scorpion carcass/molt distinction are: position of the
chelicerae (drawn in or extended); position of walking legs (folded or splayed); and body line (straight or curved). Two
other characteristics, the position of pedipalps and presence or absence of telescoped segments, approach statistical
significance and are also potentially useful. Disarticulation data are not as useful for distinguishing fossil scorpion molts
and carcasses, because there are no statistically significant differences in length of time to total disarticulation or in the
sequence of disarticulation between scorpion molts and carcasses. Among extant arthropods, scorpions possess the body
plan most similar to that of the extinct eurypterids. Therefore, the taphonomic criteria developed for distinguishing fossil
scorpion molts and carcasses may have implications for understanding molting among eurypterids.

Keywords: Arthropod, ecdysis, eurypterid

Scorpions are terrestrial chelicerates that have a fossil
record extending back more than 400 million years to the
Silurian Period. Fossil scorpions, although relatively rare, can
be well preserved (Menon 2006). However, the relative
importance of fossil scorpions in ancient ecosystems is difficult
to assess because of the arthropod trait of growth through
ecdysis. An individual scorpion may accumulate a number of
molts over its lifetime; these molts may not be distinguished
easily from carcasses, especially after becoming fossils, thereby
complicating interpretation of the scorpion fossil record. This
is also the case with fossil horseshoe crabs (Chelicerata:
Xiphosura), whose fossil carcasses and molts are often
indistinguishable (Babcock et al. 2000). Therefore, for many
fossil arthropod taxa, simple or apparent abundance may not
be a reliable reflection of actual abundance. Without accurate
abundance estimates, it is difficult to determine the impor-
tance of these arthropods in paleoecosystems.

There are a number of differences between modern scorpion
molts and carcasses, such as the presence of internal organs in
carcasses and the lack of most of these structures in exuvia,
but differences based on internal anatomy are not readily
apparent in fossils; indeed, internal structures usually cannot
be discerned in fossil scorpions (but see: Wills 1925, 1946,
1960; Kjellesvig-Waering 1986). External features of the
exoskeleton, such as the presence or absence of appendages,
opened sutures, and dislocations (separations) between
segments, are more readily accessible characters for making
the potential distinction between carcasses and exuvia among
fossil arthropods. These kinds of taphonomic observations of
the scorpion exoskeleton are the focus of this investigation.
The purpose of this paper is twofold: 1) to document whether
different taphonomic patterns exist between modern scorpion
carcasses and molts; and 2) to use these patterns, if present, to

' Corresponding author. E-mail: brandt@msu.edu

develop criteria for distinguishing fossil scorpion molts from
carcasses.

PREVIOUS WORK ON SCORPION MOLTING
AND TAPHONOMY

Scorpion molting is well understood (Polls 1990; Brownell
& Polls 2001; Gaban & Farley 2002). The steps scorpions
follow during molting do not appear to vary significantly
among taxa (Rosin & Shulov 1962; De Armas 1986): the
anterior suture opens and the animal moves forward to exit
the old exoskeleton, in a manner similar to that of a horseshoe
crab (Shuster 1982). A scorpion exuvium may include the
booklung lamellae, preoral tube, and other internal features
and there may be little distortion of delicate hairs, bristles, and
setae (Gaban & Farley 2002). The exuvium is commonly intact
except for a wedge-shaped gap beneath the carapace (Gaban
& Farley 2002). These observations on modern scorpion
exuvia suggest that, as in the case of horseshoe crabs, a well-
preserved fossil scorpion exuvium may be difficult to
distinguish from a well-preserved fossil scorpion carcass. In
this study, we identify more readily observed external features
of the scorpion exoskeleton as the basis for distinguishing
scorpion molts and carcasses.

METHODS

Six species of living scorpions belonging to six genera were
used in the taphonomy experiments (Table 1). These taxa were
chosen because they were available in sufficient quantity for
replicate experiments. The scorpion molts used in this study
were donated by arachnid hobbyists; live scorpions were
purchased through a commercial supplier (www.swiftinverts.
com) or were donated by arachnid hobbyists. Scorpion
carcasses were obtained through either the natural death of
the animal or mortality through freezing. We collected two
categories of taphonomic data on scorpion molts and
carcasses: 1) the initial post-mortem (for carcasses) or post-

312

MCCOY & BRANDT— SCORPION TAPHONOMY

313

Table 1. — Taphonomic characteristics of scorpion molts and carcasses. Significant p-values are in bold. 'This cell does not total 13 because six
exuvia lack appendages. ^Chi-square with Yates correction, all one degree of freedom.

Character

Description

Molt {n = 13)

Carcass (n = 13)

Chi-square^

P

Chelicerae

Extended

12

0

18.73

1.5X10'-“'

Retracted

1

13

Pedipalps

Extended

11

6

2.71

0.10

Retracted

2

7

Body line

Curved

8

0

8.85

0.003

Straight

5

13

Walking legs

Splayed

7'

3

7.91

0.005

Folded

0

10

Telescoping of segments

Present

4

0

2.66

0.10

Absent

9

13

molt (for exuvia) exoskeletal posture and 2) the order and
timing of disarticulation of the exoskeleton (i.e., the disartic-
ulation sequence) through subsequent tumbling experiments.

Death/molt posture. — Intact scorpion carcasses and exuvia
were photographed, and the following exoskeletal character-
istics were recorded for each specimen: 1) curvature of the
mesosoma and metasoma; 2) presence or absence of telescoped
mesosomal and metasomal segments; 3) position/orientation
of the chelicerae; 4) position/orientation of the walking legs;
and 5) position/orientation of the pedipalps. A total of 13
carcasses and 13 molts were used in this study. Twenty-four
fossil scorpions from the collections of the Yale Peabody
Museum (YPM) were photographed and described following
the same five criteria.

Tumbling. — After documenting the initial condition of the
exoskeleton, scorpions were treated in one of three ways: 1)

tumbling (wet)- 13 carcasses and 1 1 exuvia were tumbled to the
point of complete disarticulation in fresh water inoculated by
the addition of water from an aquarium; 2) tumbling (dry)-one
carcass and two exuvia were tumbled dry until disarticulated; 3)
one carcass and one exuvium were left to decay in inoculated
freshwater with no agitation as controls. Wet tumbling was
done in one of two small (5-cm radius) plastic rock-tumblers
available from hobby stores. One tumbler barrel had a smooth
interior; the other tumbler had molded invaginations of the
barrel wall that acted as interior bails. Dry tumbling of larger
specimens was done in a larger (20-cm radius) tumbler with two
internal bails. The tumblers were checked daily and the extent
of disarticulation of the specimens recorded. Carcasses were
kept frozen until used in the tumbling experiments to minimize
potential differences in disarticulation between specimens due
to differences in the degree of decay (Allison 1986).

Table 2. — Laboratory treatment of scorpion carcasses and molts.

Species

Molt/Carcass

Wet/Dry

Tumbler

Time (days)

Pandimis imperator (C.L. Koch 1841)

molt

dry

bail-lg

2

Leiurus quinquestricitus (Ehrenberg 1928)

molt

dry

smooth

16

P. imperator

molt

wet

smooth

14

P. imperator

molt

wet

smooth

7

Parahuthus transvaalicus

molt

wet

smooth

15

P. transvaalicus

molt

wet

smooth

11

P. imperator

molt

wet

smooth

18

P. imperator

molt

wet

smooth

12

P. imperator

molt

wet

bail

18

P. imperator (juv)

molt

wet

smooth

10

P. imperator

molt

wet

smooth

11

P. imperator

molt

wet

bail

6

P: imperator (juv)

molt

wet

smooth

7

P. imperator (juv)

carcass

dry

bail-lg

4

P. imperator (juv)

carcass

wet

smooth

22

Anuroctonus phaiodactylus (Wood 1863)

carcass

wet

smooth

8

Smeringurits mesaensis (Stanhke 1957)

carcass

wet

smooth

19

S.mesaensis

carcass

wet

bail

18

S. mesaensis

carcass

wet

smooth

11

S. mesaensis

carcass

wet

bail

10

S.mesaensis (juv)

carcass

wet

bail

6

S.mesaensis

carcass

wet

smooth

13

S.mesaensis

carcass

wet

bail

10

S.mesaensis

carcass

wet

smooth

11

S.mesaensis (juv)

carcass

wet

bail

3

S.mesaensis (juv)

carcass

wet

smooth

13

Vaejovis spinigerus (Wood 1863)

carcass

wet

smooth

12

314

THE JOURNAL OF ARACHNOLOGY

Table 3. — Comparison of molt/carcass and tumbler effects on disarticulation. F-values from the /-tests on the mean time to separation of
exoskeletal tergites; significant P-values are in bold. S = smooth tumbler, B = tumbler with bails, PC = pedipalp claws, PA = pedipalp
appendages, DL = distal leg segments, PL = proximal leg segments, M345 = third, fourth and fifth metasomal segments as one unit, M2 =
second metasomal segment. Ml = first metasomal segment, Ch = chelicerae, Ca = carapace, and TMS = total mesosomal separation (= total
exoskeletal disarticulation). The lack of significant P-values in the first row indicates that there are no differences in the timing of disarticulation
between molts and carcasses. In the second row, the majority of the P-values are significant, reflecting differences in the timing of disarticulation
between specimens in the smooth tumbler and specimens in the tumbler with the bail.

PC

PA

DL

PL

M345

M2

Ml

Ch

Ca

TMS

molt vs carcass

S- vs. B-tumbler

0.6817

0.0001

0.7332

0.0001

0.9333

0.0083

0.9858

0.0085

0.65

0.005

0.932

0.0025

0.605

0.939

0.4175

0.0053

0.363

0.089

0.891

0.255

Statistics. — Standard chi-square tests of independence with
Yates correction (Preacher 2001) were used to compare the
frequency of each of the five exoskeletal characteristics of
death/molt postures listed above. To adjust for multiple tests,
we modified the standard alpha level of 0.05 as per the
conservative Bonferroni correction by dividing by 5 (the
number of characteristics tested), so that P-values less than
0.01 were considered statistically significant (Preacher 2001).
The chi-square calculations were carried out on an online
calculator (Preacher 2001).

We used /-tests to compare the mean time-to-separation of
various exoskeletal elements: pedipalp claws, pedipalp
appendages (tibia and brachium), distal leg segments,
proximal leg segments, last three metasomal segments,
second metasomal segment, first metasomal segment, chelic-
erae, carapace, and total mesosomal separation. Two
variables were considered: specimen type (molt or carcass)
and tumbler type (smooth interior/bails). P-values less than
0.05 were considered significant. The sequence of disarticu-
lation suggested by the mean time-to-separation of each

'

1

10

^ 1

1

' 1

1

, 1

1 '

1

• UpperCl bound

1

1 1

1 '

1 1

1

1

1 1

i

• Lower Cl bound

1

1

. HMean

IV1345 PL PC IV12 DL Ca Ml PA Ch TMS

1

1 1

1 , J

1 1

j ■ • UpperCl bound

PC Ca PA M345 M2 DL PL Ch TMS Ml

Figures 1, 2. — Plots of mean time to separation data showing order and timing of disarticulation for scorpion molts under different tumbling
conditions. The y-axis is days. The vertical bars represent 95% confidence intervals. A small sample size results in a large standard deviation,
which explains ‘negative days’ in Figure 2. The bars are ordered according to increasing mean time to separation (the center of each bar); the
arms of the bars show the entire expected range of time to separation: 1. molts tumbled in smooth canister n = 9; 2. molts tumbled with agitation
(bail), n = 2; PC = pedipalp claws, PA = pedipalp appendages, DL = distal leg segments, PL = proximal leg segments, M345 = third, fourth
and fifth metasomal segments as a single unit, M2 = second metasomal segment. Ml = first metasomal segment, Ch = chelicerae, Ca =
carapace, and TMS = total mesosomal separation (= total exoskeletal disarticulation).

MCCOY & BRANDT— SCORPION TAPHONOMY

315

10

1

1 1

1 1

1

1 • uppf*r n limit

• lower Cl limit

^ ^ ^ ^

-s ■■

PC PA Ch M345 M2 DL PL IV

.1 Ca TMS

Figures 3, 4. — Plots of mean time to separation data showing order and timing of disarticulation for scorpion carcasses under different
tumbling conditions. The y-axis is days. The vertical bars represent 95% confidence intervals. The bars are ordered according to increasing mean
time to separation (the center of each bar), the arms of the bars show the entire expected range of time to separation: 3. carcasses tumbled in
smooth canister, n = 8; 4. carcasses tumbled with agitation, n = 5. PC = pedipalp claws, PA = pedipalp appendages, DL = distal leg segments,
PL = proximal leg segments, M345 = third, fourth and fifth metasomal segments as a single unit, M2 = second metasomal segment. Ml = first
metasomal segment, Ch = chelicerae, Ca = carapace, and TMS = total mesosomal separation (= total exoskeletal disarticulation).

exoskeletal element was also tested using 95% confidence
intervals.

The Mests and 95% confidence intervals were generated
using SAS software, version 9.2 of the SAS System for
Windows (SAS Institute Inc., Cary, NC, USA).

RESULTS

Taphonomic patterns, 1. Death/molt posture. — The scorpion
carcasses examined for this study exhibited the following
characters: 1) retracted chelicerae (100% of the specimens); 2)
straight body line (in dorsal view) with the metasoma extended
straight back (100%); 3) pedipalps pulled in toward the
prosoma (54%); and 4) walking legs folded against the body
(77%) (Table 1 and Figures 7-18). In contrast, molts were
characterized by the following features: 1) extended chelicerae
(92%); 2) curved body line and curved metasoma (62%); 3)
pedipalps pulled well back from the prosoma in an extended
position (85%); and 4) splayed walking legs (100%) (Table 1
and Figures 7-18). Four of 13 molts (31%) exhibited
telescoped mesosomal segments and overlap of the ventral
surface on the dorsomedial surface (Figures 11, 15). These
observations held across the six different scorpion genera
examined for this study, suggesting taxonomic independence
of these body-posture criteria. The low P-values obtained in

the chi-squared test (Table 1) indicate that our sample size is
sufficiently large to reflect a significant difference (P < 0.05)
between scorpion death and molt postures.

Taphonomic pattens, II. Disarticulation. — Previous actuo-
taphonomic studies on tumbling segmented, exoskeleton-
bearing invertebrates suggested that freezing the animals does
not affect the exoskeletal disarticulation sequence (Kidwell &
Baumiller 1990 on echinoids). We detected no differences in
disarticulation between scorpions that had been frozen versus
scorpions that had died naturally. Time-to-total disarticula-
tion of the specimens is given in Table 2.

Wet V5. Dry: Both exuvia and carcasses released complete,
intact tergites under wet tumbling conditions. These elements
were not broken by subsequent tumbling (tumbling was
terminated with the dissociation of the last exoskeletal
elements). Carcasses tumbled dry behaved as carcasses
tumbled wet. However, molts tumbled dry were quickly (2-3
days) reduced to ragged-edged, broken exoskeletal pieces
rather than separated, unbroken tergites.

Disarticulation Sequence: A disarticulation sequence com-
prises two components: order and timing. Order refers to the
sequence in which a given exoskeletal element separates in
relation to the other exoskeletal elements. For example, “legs,
pedipalps, chelicerae” is a different disarticulation order than

316

THE JOURNAL OF ARACHNOLOGY

1 a

\

10

*

1

1

'

1

1

1 ■

1

• upperCI limit

, 1

1 I

1 1

1 I

1

• lower Cl lirnll

1

1 '

1

. ■mean

M345 PC DL PL M2 Ca PA Ch Ml TMS

8

■

1

1

1 1

1

1

• UpperCI bound

1 1

1 1

1 1

1 I

1 '

1 '

1

• Lower Cl bound

4

■ Mean

0 -

M345 M2 PL DL PC PA Ch Ml Ca TMS

Figures 5, 6. — Plots of mean time to separation data showing order and timing of disarticulation for all scorpion molts and carcasses. The y-
axis is days. The vertical bars represent 95% confidence intervals. The bars are ordered according to the mean time-to-separation (the center of
each bar); the arms of the bars show the entire expected range of time to separation: 5. all molts, « = 1 1; 6. all carcasses, n = 13. PC = pedipalp
claws, PA = pedipalp appendages, DL = distal leg segments, PL = proximal leg segments, M345 = third, fourth and fifth metasomal segments
as a single unit, M2 = second metasomal segment, M I = first metasomal segment, Ch = chelicerae, Ca = carapace, and TMS = total mesosomal
separation (= total exoskeletal disarticulation).

“pedipalps, legs, chelicerae.” Timing refers to the elapsed time
before an exoskeletal element separates. Only differences in
order are noticeable in fossils, but a difference in order
necessarily requires a difference in timing.

Molts and carcasses showed no significant differences in
mean time to separation in any of the exoskeletal elements
(Table 3, Row 1), indicating that there is no significant
difference in the sequence or order of disarticulation for
scorpion molts as compared to scorpion carcasses. Plots of
95% confidence intervals of the mean time to separation for
each exoskeletal element (Figures 1-6) show a general pattern
shared by molts and carcasses. The appendages (PC, PA, DL,
PL, Ch in Figures 1-6) and metasoma (M345, M2, Ml in
Figures 1-6) tend to separate from the rest of the body before
the mesosoma disarticulates (TMS in Figures 1-6). For
example, in Figure 6, the pedipalps separate on day 6, and
the mesosoma disarticulates on day 12.

Using the rock-tumbler with the internal bails shortened
disarticulation time for both molts (compare Figures 1 and 2)
and carcasses (compare Figures 3 and 4). This tumbler effect is
also reflected in the statistically significant values in Table 3,
Row 2. However, the tumbler effect does not result in a
statistically significant difference in disarticulation timing

between molts and carcasses, as shown by the overlapping
confidence intervals (compare Figure 1 with Figure 3 and
Figure 2 with Figure 4).

A somewhat unexpected result was the comparison of times
to total disarticulation. Molts proved to be as durable as
carcasses when tumbled in water, as evidenced by the fact that
the mean time to total disarticulation (see Table 3, Row 1,
Column TMS, and Figures 5, 6) is not significantly different
between molts and carcasses.

To summarize the statistical analysis of the tumbling data,
molts and carcasses generally release appendages and meta-
somal segments early, the mesosoma is the most resilient
exoskeletal unit for both exuvia and carcasses, and molts take
as long as carcasses to reach total disarticulation.

DISCUSSION AND APPLICATION TO
FOSSIL SCORPIONS

Scorpion molting and exuvia taphonomy. — Observations on
ecdysis in modern scorpions support the criteria defined here of
appendage position and body posture as diagnostic for
distinguishing carcasses and exuvia. De Armas (1986) conclud-
ed that the position of the legs on a suitable substrate is critical
for scorpions to successfully complete the process of ecdysis,

Table 4. — YPM specimens and identification as molt or carcass based on the criteria defined herein; “+”indicates presence, denotes missing features or that determination could
not be made, “?” denotes identification could not be determined due to poor preservation. E = extended, R = retracted, C = curved, St = straight, Sp = splayed, F = folded, P =
present, A = absent. In the age column, MP = Middle Pennsylvanian, LS = Late Silurian.

MCCOY & BRANDT— SCORPION TAPHONOMY

317

<

U

w

cd

c

o

X)

cd

U

o

U

>>

c

Ui *

o

o

U

o

U

o

Xi

u

cd

U

o

U

'-o

o

X

3-1

cd

U

S

o

U

u ^ a
(U ^ o

^ rh •£

:s ^ S

U

'O

d

o

X)

u

cd

U

O.

d

o

X

u

cd

U

cd

d

o

X

u<

cd

U

o. Pm

d

o

X

u

cd

U

0^

*> ’> ■>

00 C/5

a a 6 53

cd cd cd ^

(U (U (U <u

00

= = a ? o =
^ ^ ^ ^

> >
Cfl CO

a a

CS CS

: E

g ^

Id o

5 6

" +

u o
cd fi X
a d rv|

11s

O14 Qm Pm

^ ^>M

^ 'O

§ ri § ri § J =

l-i U. 3-1

0 0 0

^ 53^ ^ o

kq Eij

os

SO

so

O

00

^ On

o cn

o

s s

CL, Cu

cu

>

o

. u
'2

S ^

Z e 1

13 O B «

o

O

o
U

>0 >,

-o "O

3 d 3 d

o

u

o

u

"O *0

d I d

3 d 3 !

i—, ►=( s_ ^

^ m on
mJ mJ J

>-

<D O

s s

a> <u

s s

^ ^ ^ ^

+ -I-

o

o

s

n

00000

+ +

s 3

~S a -S
S ^ K

Ҥ -s -s

'ft -S
Q„ a,

f"

— O'!

s s s s

&M cu &

>H >-l >-< >1

a

-a

'3

-ft ft

Cl,

? S
-S ^

3^^

o o m

^ m

s s s

Pm Pm Pm

>1 >H

QJ 0^

X X

S X

000

a a a

-h , I +

-I- -I- -I-

+ -1- ■ -I-

+ -I- -I- + + -I-

00 00 00 00
0000

(N oi CN

s s s s

Pm Pm P^ Pm
>L >1 >1

^ ^ ^ ^ r

d d d g

<u h <u h c

u JJ u ^ aj ca

■ d d

QO W M M CO
J J _1 J J

(O 10

s s

l-< 3-1

(U 0?

d d

(U 13^ (D

ess

X X S

U I-I ^

<u 0^

d d d

Cd

o X
d c

a a

+ ■ + -I-

+ ■ +

+ -I- -I- +

S R K c a

.

CJ C> C) O O

C) O O

ON O <— < rn X

rs m m CN

^ ^ 00 ON

00 00 00 On (N

0000^
r'4

s s s s s

Pm Pm Pm Pm Pm

>H >H >M >M

CO

X

S

X

S

+ » + + • 4- « + +

r-

ON

P.

00

(N

On

Pm

>H

s

2

CO

ON

X

X

o

s s

Pm

>H

318

THE JOURNAL OF ARACHNOLOGY

Figures 7-18. — Comparison of fossil scorpion molts and carcasses: 7. Modern scorpion molt with extended chelicerae, scale bar is 5 mm; 8.
Fossil scorpion molt with extended chelicerae, scale bar is 5 mm; 9. Modern scorpion carcass with retracted chelicerae, scale bar is 5 mm; 10.
fossil scorpion carcass with retracted chelicerae, scale bar is 10 mm; 1 1. Modern scorpion molt with highly curved body line, scale bar is 10 mm;
12. Fossil scorpion molt with highly curved body line, each scale bar is 5 mm, from Treatise on Invertebrate Paleontology, ®1956, courtesy of the
Geological Society of America and the University of Kansas; 13. Modern scorpion carcass with straight body line, scale bar is 10 mm; 14. Fossil
scorpion carcass with straight body line, scale bar is 10 mm; 15. Modern scorpion molt with extended pedipalps, scale bar is 10 mm; 16. Fossil
scorpion molt with extended pedipalps, scale bar 10 mm; 17. Modern scorpion carcass with retracted pedipalps, scale bar 10 mm; 18. Fossil
scorpion carcass with retracted pedipalps, scale bar is 5 mm.

MCCOY & BRANDT— SCORPION TAPHONOMY

319

Table 5. — Literature survey with identification of illustrated specimens as molt or carcass based on the criteria defined herein; indicates
presence, denotes missing features or poor preservation. E = extended, R = retracted, C = curved, St = straight, Sp = splayed, F = folded, P
= present, A = absent.

Author

Figure

Chelicerae

Pedipalps

Body Line

Walking legs

Telescoping

Identification

E

R

E

R

C

St

Sp

F

P

A

typical carcass

-

+

-

-1-

-

+

-

-1-

-

+

carcass

typical molt

-1-

-

+

-

+

-

-1-

-

+

-

molt

Carvalho & Lourenco

2001

Figure 2

-

+

+

-

-

+

-

+

-

-

carcass

Lourenco & Gall 2004

-

-

-h

-

-

-1-

-

-h

-

-

molt

Santiago-BIay 1988

Figure 1-2

-

-

+

-

+

-

-t-

-

-

-

molt

Santiago-Blay et al.

2004

Figure 1

-

+

-

-

-

-1-

-

-

-

-

carcass

and Gaban & Farley (2002) noted that the walking legs become
rigid and fixed in place before ecdysis begins. These observa-
tions suggest that ecdysial leg position may be retained after
ecdysis, although Gaban & Farley (2002) also observed that
Pamroctonus mesaensis (Stahnke 1957) pulls its appendages
medially during ecdysis. The exuvia of all the scorpion genera
examined in our study show the splayed walking leg posture.

The position of the pedipalps differs between the exuvia and
carcasses examined for this study. Our observations compliment
those of Gaban & Farley (2002), who noted that Androctonus
australis (Linnaeus 1758) holds its pedipalps in a distinctive
position before ecdysis, with the chelae sharply angled toward each
other. These authors noted that this pre-molt posture contrasts
with the posture of scorpions at rest, in which the pedipalps are
retracted. More than half of the carcasses examined in this study
retained their pedipalps in a retracted position, suggesting that the
relaxed position also served as a post-mortem posture.

Observations on modern scorpions also support body line
as a criterion for distinguishing exuvia (curved) from carcasses
(straight). Gaban & Farley (2002) cite examples of exuvia of
A. australis, P. mesaenis, and Parabuthus transvaaticus Purcell
1899 that are curved dorsally or to one side.

Fossil scorpion taphonomy. — We examined 24 well-preserved,
nearly complete fossil scorpions from the Ciurca collection at
the Yale Peabody Museum (YPM; Table 4) and applied our
taphonomic criteria in an attempt to characterize each specimen
as molt or carcass. One or more of the four distinguishing
criteria (body line, position of chelicerae, position of walking
legs, and position of pedipalps) could be identified in 85-90% of
the YPM specimens (Table 4). Approximately 89% of the
classifiable YPM specimens fit the criteria for molts (Table 4).
The results of the YMP census support the conclusion of other
authors that the bulk of the scorpion fossil record comprises
exuvia (Kjellesvig-Waering 1986).

Most authors of papers on scorpion systematics do not
include taphonomic data in their description of new speci-
mens, nor do they make a determination of whether their
illustrated specimens represent molts or carcasses. The
taphonomic criteria developed herein can also be used to
make the molt/carcass determination for fossil scorpions
illustrated in the literature (Table 5).

The results of the tumbling experiments suggested that
scorpion molts in water are nearly as durable as scorpion
carcasses, as measured by time to total disarticulation
(Tables 2, 3). This observation is interesting because it runs

counter to the intuitive conclusion that molts must be more
fragile than carcasses, and it calls into question the assumption
that intact fossil scorpions are more likely the remains of the
more “robust” carcass than the presumably “fragile” exuvium
(Wills 1959; Jeram 2001; Gaban & Farley 2002).

The scorpion body plan is strikingly similar to that of the
extinct eurypterids (Chelicerata: Eurypterida), and on this basis,
scorpions can be considered reasonable taphonomic analogues
for their extinct chelicerate cousins. Therefore, the criteria for
distinguishing fossil scorpion molts and carcasses may also be
useful for distinguishing fossil eurypterid molts and carcasses. A
recent study of eurypterid molting (Tetlie et al. 2008) concluded
that eurypterids indeed molted in much the same manner as
modem scorpions, in spite of the obvious ecological differences
between the aquatic eurypterids and the terrestrial scorpions.

ACKNOWLEDGMENTS

We would like to thank Derek Briggs and Erik Tetlie for
their discussions and collaboration and Mary Marazita for
help with the statistical analyses. Many thanks also to Kevin
Cash, Matt Wilhm, Matt Grahm, and Mike Weissman for
their donation of scorpion molts and carcasses, and Tomasz
Baumiller for the use of his tumbling apparatus. The paper
was improved by the constructive comments of Derek Briggs,
Robert Anstey, and two anonymous reviewers. Funding in
support of this project was provided by the Schuchert Fund of
the Yale Peabody Museum, the North-Central Section of the
Geological Society of America, the College of Natural Science,
the Department of Geological Sciences, and the Honors
College at Michigan State University.

LITERATURE CITED

Allison, P.A. 1986. Soft-bodied animals in the fossil record: The role
of decay in fragmentation during transport. Geology 14:979-981.
Babcock, L.E., D.F. Merriam & R.R. West. 2000. Paleolimulus, an
early limuline (Xiphosurida), from Pennsylvanian-Permian Lager-
statten of Kansas and taphonomic comparison with modern
Limulus. Lethaia 33:129-141.

Brownell, P., & G.A. Polis, eds. 2001. Scorpion Biology and
Research. Oxford University Press, New York.

Carvalho, M.G.P. & W.R. Lourenco. 2001. A new family of fossil
scorpions from the Early Cretaceous of Brazil. Comptes Rendus de
I’Academie de Sciences de Paris, Earth and Planetary Sciences
332:711-716.

De Armas, L.F. 1986. Biologia y morfometria de Rhopaliirus garridoi
Armas (Scorpiones, Buthidae). Poeyana 333:1-27.

320

THE JOURNAL OF ARACHNOLOGY

Gaban, R.D. & R.E. Farley. 2002. Ecdysis in scorpions; supine
behavior and exuvial ultrastructure. Invertebrate Biology
12:136-147.

Jeram, A.J. 2001. Paleontology. Pp. 370-392. In Scorpion Biology
and Research. (P. Brownell & G.A. Polls, eds.). Oxford University
Press, New York.

Kidwell, S.M. & T. Baumiller. 1990. Experimental disintegration of
regular echinoids: roles of temperature, oxygen, and decay
thresholds. Paleobiology 16:247-271.

Kjellesvig-Waering, E.N. 1986. A restudy of the fossil Scorpionida of
the world. Palaeontographica Americana 55:1-287.

Lourenco, W.R. & J.C. Gall. 2004. Fossil scorpions from the
Buntsandstein (Early Triassic) of France. Comptes Rendus Palevol
3:369-378.

Menon, F. 2006. Higher systematics of the scorpions from the Crato
Formation, Lower Cretaceous of Brazil. Palaeontology 50;
185-195.

Polls, G.A. (ed.). 1990. The Biology of Scorpions. Stanford University
Press, Stanford, California.

Preacher, K.J. 2001. Calculation for the chi-square test: An interactive
calculation tool for chi-square tests of goodness of fit and indepen-
dence [Computer software]. Online at http://www.quantpsy.org.

Rosin, R. & A. Shulov. 1962. Studies on the scorpion Neho
hierochontkus. Proceedings of the Zoological Society of London
140:547-575.

SAS Institute Inc. 2004. SAS 9.2. SAS Institute Inc, Cary, North
Carolina.

Santiago-Blay, J.A. 1988. A fossil scorpion Tityus geratus new species
(Scorpiones: Buthidae) from Dominican amber. Historical Biology
1:345-354.

Santiago-Blay, J.A., M.E. Soleglad & V. Fet. 2004. A redescription
and family placement of Uintascorpio Perry, 1995 from the
Parachute Creek Member of the Green River Formation (Middle
Eocene) of Colorado, USA (Scorpiones: Buthidae). Revista Iberica
de Aracnologia 10:7-16.

Shuster, C.N. Jr. 1982. A pictorial review of the natural history and
ecology of the horseshoe crab Limiilus polyphemus, with reference
to other Limulidae. Pp. 1-52. In Physiology and Biology of
Horseshoe Crabs. (J. Bonaventura, C. Bonaventura & S. Tesh,
eds.). Alan R. Liss Inc, New York.

Tetlie, O.E., D.S. Brandt & D.E.G. Briggs. 2008. Ecydsis in sea
scorpions (Chelicerata: Eurypterida). Palaeogeography, Palaeocli-
matology, Palaeoecology 265:182-194.

Wills, L.J. 1925. The morphology of the Carboniferous scorpion
Eohiithus Fritsch. Journal of the Linnean Society of London
36:87-97.

Wills, L.J. 1946. A monograph of British Triassic scorpions, Part 1.
Palaeontographical Society Monograph 100:1-74.

Wills, L.J. 1959. The external anatomy of some Carboniferous
‘scorpions’. Part 1. Palaeontology 1:261-282.

Wills, L.J. 1960. The external anatomy of some Carboniferous
‘scorpions’. Part 2. Palaeontology 3:276-332.

Manuscript received 30 January 2009, revised 23 July 2009.

2009. The Journal of Arachnology 37:321-330

Temperature and desiccation effects on the antipredator behavior of Centruroides vittatiis

(Scorpiones: Buthidae)

B, Evan Carlson; Department of Biological Sciences, Bethel University, St. Paul, Minnesota 55112, USA

Matthew P. Rowe‘: Department of Biological Sciences, Sam Houston State University, Huntsville, Texas 77341, USA

Abstract. Temperature can profoundly affect many physiological processes, including muscle performance. Many
ectotherms appear sensitive to this relationship, choosing times and locations of activity permitting high body temperatures
and, thus, quick escape from predators. High body temperatures, however, can lead to dehydration, which in turn affects
muscle performance. Striped bark scorpions Centruroides vittatiis Say 1821 provide an ideal model for assessing the effects
of temperature and water loss on two potentially important antipredator behaviors, sprinting and stinging. Scorpions had
significantly higher sprint speeds at warmer temperatures, with males significantly faster than females. Additionally, sting
latency was longer and sting rate lower when scorpions were cooler. Intriguingly, females appear capable of stinging at a
higher rate than males. Desiccation allowed the scorpions to sprint significantly faster than control (hydrated) scorpions,
probably due to weight loss. The influence of temperature on sprinting and stinging might thus explain bark scorpions’
preference for maintaining high body temperatures during periods when they are exposed to predation. When inactive,
however, scorpions may benefit from maintaining lower body temperatures to decrease resting metabolic rate and
desiccation.

Keywords: Defensive behavior, scorpions, thermal ecology, sting speed, sexual dimorphism

Ectotherms generally exhibit preferred body temperatures
(Tp) that they maintain through behavioral thermoregulation.
Temperature profoundly affects many physiological processes
including, but not limited to, oxygen consumption, digestion
(Bobka et al. 1981; Zhang & Ji 2004), growth (Angilletta et al.
2004), and locomotor performance (Forsman 1999). The latter
process is related to temperature primarily due to well-
recognized thermal dependencies of muscular contraction and
relaxation (Bennett 1984). The influence of temperature on
locomotor performance is ecologically important for many
organisms, as it impacts both hunting ability and predator
avoidance via such activities as sprinting (Waldschmidt & Tracy
1983; Bauwens et al. 1995), flying (Machin et al. 1962),
swimming (Turner et al. 1985), and striking (Greenwald 1974;
Rowe & Owings 1990; Webb & Shine 1998). Indeed, endother-
mic enemies are known to take advantage of ecotherms that find
themselves with body temperatures (Tt,) cooler than they might
prefer (Rowe & Owings 1996; Swaisgood et al. 1999, 2003).
Sprint speed, in particular, can be critical for an ectotherm
attempting escape from a potential predator (van Berkum et al.
1986; Hertz et al. 1988). For this reason, Tp is often strongly
correlated with the optimal temperature (Tq) for sprinting
capacity (Miller 1982; Bauwens et al. 1995; Forsman 1999),
providing evidence that ectothermic organisms typically select
body temperatures that maximize their locomotor capabilities.
If a prey organism’s ability to survive encounters with and
escape from predators is dependent on locomotor performance,
then maintaining Tp will favor survival in the face of predation;
indeed, this has been demonstrated in wild populations
(Christian & Tracy 1981).

The match, however, between Tp and To is not always
perfect (see reviews in Huey & Slatkin 1976; Huey 1982). Some
ectothermic organisms may have a broad range of tempera-
tures over which performance varies little (Schmalhofer &

' Corresponding author. E-mail: rowe@shsu.edu

Casey 1999), negating the advantage of finely tuned behav-
ioral thermoregulation. Other species may face competing
physiological demands with different TqS. Side-blotched
lizards [Uta stanshuriana), for example, select microhabitats
that maximize their sprint speeds during the morning and late
afternoon, but chose sub-optimal, shaded habitats during
midday to avoid desiccation (Waldschmidt & Tracy 1983).
Water conservation can itself feed back on locomotor
performance. Dehydration has been found to decrease
endurance but not burst activity in frogs and lizards (Crowley
1985; Moore & Gatten 1989; Wilson & Havel 1989) and to
decrease walking velocity in crayfish (Claussen et al. 2000).
Because higher temperatures typically increase the rate of
water loss (Slobodchikoff 1983), many ectotherms might face
a difficult physiological trade-off: a high Tb may support
increased performance through its effect on muscle contrac-
tion while simultaneously decreasing performance from
desiccation.

Desert scorpions provide an excellent model organism for
examining the effects of temperature and dehydration on
antipredator motor patterns. They flourish within desert
ecosystems (Hadley 1974), forming an ecologically important
link as the dominant predators of small herbivorous and
detritivorous arthropods (McCormick & Polis 1990) and as
the prey of many vertebrates (Polis et al. 1981; Ayal 2007).
Indeed, the total biomass of these highly successful arachnids
can exceed that of all vertebrates combined in some desert
systems (Polis 1990). Predation is the dominant mortality
factor (Polis 1990), making scorpions especially valuable for
studies of antipredator behavior. They tend to exhibit
nocturnal, “time-minimizing” activity patterns (Schoener
1971) in which they are rarely active and foraging, preferring
to remain in refugia and thus limiting their risk of falling prey
(Polis 1980). Surface activity of many scorpions is highest
during the earlier hours of the night and gradually decreases as
dawn nears (Hadley & Williams 1968; Polis 1980; Warburg &

321

322

THE JOURNAL OF ARACHNOLOGY

Polis 1990; Carlson et al., unpublished data). This may afford
them access to sun-warmed substrates, returning to refugia as
the surface substrates cool. Additionally, scorpions appear
most active during both warm (Polis 1980) and humid
(Skutelsky 1996) weather conditions, decisions that may help
them maximize body temperature while minimizing water loss.

Anecdotal reports suggest that temperature may indeed affect
locomotor performance in scorpions (Hadley 1974; Warburg &
Polis 1990), though there is, as yet, little experimental evidence
of this. The water loss rates of scorpions, though low (Hadley
1974), are known to rise with temperature (Hadley 1970; Gefen
& Ar 2006), and desiccated scorpions exhibit limited function-
ality of the limbs (Sensenig & Shultz 2004). This is likely due to
the use of hydraulics in some locomotor activities (Sensenig &
Shultz 2004) and the decreased effectiveness of hydrostatic
systems with fluid loss (Anderson & Prestwich 1975). Addition-
ally, dehydration may reduce the oxygen-transporting abilities
of the hemolymph, further inhibiting locomotor performance
(Gefen & Ar 2005). All of this serves to highlight the potential
importance of thermoregulation and water balance for locomo-
tor-based antipredator behaviors. Indeed, the primary defensive
mechanisms of scorpions, sprinting (Shaffer & Formanowicz
1996, 2000) and stinging, fall into this category. The purpose of
this research is to determine the effects of temperature and
desiccation on these antipredator behaviors in a model scorpion.

Centniroides vit talus Say 1821, the striped bark scorpion, is an
excellent species for such research. In addition to being the most
common and active scorpion in many lithic deserts of the
southwestern United States and northern Mexico (Brown et al.
2002), scorpions of the family Buthidae are noted for their high
temperature preferences (Warburg & Ben-Horin 1981), low
water loss rates (Gefen & Ar 2004), and well-developed
osmoregulatory ability (Gefen & Ar 2005). C. vittatus is
subjected to a wide range of temperatures, from searing summer
heat to subfreezing winter nights (Whitmore et al. 1985). Striped
bark scoipions appear sensitive to the influence of temperature
on their defensive capabilities, as they are less likely to be active
on the surface during cool weather (Brown & O’Connell 2000;
Brown et al. 2002; Yamashita 2004). And when they are active,
they are less likely to seek refuge in bushes (an antipredator
tactic; Brown & O’Connell 2000; McReynolds 2004) when
temperatures are warm (McReynolds 2008).

Striped bark scorpions are also appropriate because stinging
and sprinting are important components of their antipredator
behavior. Although bark scorpions {Centniroides spp.) are well
known for possessing potent neurotoxins that might deter a
vertebrate enemy, grasshopper mice {Onychomys spp.) are
resistant to the lethal effects of the venom and feed voraciously
on Centniroides (Rowe & Rowe 2006, 2008); an effectively
delivered sting, however, often causes sufficient pain for the
mouse to drop the scorpion, providing an opportunity for its
escape (Rowe & Rowe 2006). The broad range of environmental
temperatures to which striped bark scorpions are exposed,
coupled with the importance of stinging and sprinting for
evading predators, make C. vittatus an ideal model for testing
hypotheses about scorpion thermal ecology and physiology in
relation to defensive behavior. In another report (Carlson et al.,
unpublished data), we show that C. vittatus from two different
populations both select substrates generating high body
temperatures (36-38°C) when tested in a thermal gradient.

Figure 1. — Sexual dimorphism in the metasomas of male and
female C. vittatus.

Here, we examine the effects of temperature, and of tempera-
ture-induced dehydration, on their defensive ability.

METHODS

Animals. — From a large group (600+) of Centniroides
vittatus scorpions collected during May 2008 in the Organ
Mountains in New Mexico, USA, 180 experimental subjects
(60 adult males, 60 adult females, and 60 unsexed juveniles)
were selected randomly. All scorpions were kept at room
temperature (—25° C) in plastic sweater boxes with gravel
substrates, egg crate refugia, and petri dishes for water. The
communal nature of C. vittatus (McAlister 1966; Polis &
Lourengo 1986) allowed us to house 30-60 individuals per box
with no signs of aggression. Due to concerns that the added
weight and the energetics of digestion may influence thermal
ecology, the scorpions were only fed between experiments (i.e.,
after all trials for a single experiment were completed), and all
were fed at the same time. Food items offered included
crickets {Acheta doinesticiis), “superworms” {Zoophobas morio
larvae), and mealworms {Tenebrio molitor larvae). To best
approximate the light conditions faced by scorpions in the
wild (where they are active at night and hiding in crevices or
burrows during the day), lights in the room housing the
scorpions were kept on 24-h D; a small window nonetheless
provided diffuse daylight cueing.

The scorpions were each marked with a unique four-dot
pattern of fluorescent paint (Hadley & Williams 1968; Bradley
1988). Individuals were sexed by the length and thickness of the
segments of the metasoma (tail). Males have thin and elongate
segments whereas females possess shorter and thicker segments
(Polis & Sissom 1990; Fig. 1 ). Pregnancies in females to be tested
were noted; because most of the females were gravid (> 90%),
this variable could not be avoided while maintaining adequate
sample sizes. Individuals were characterized as juveniles based
on qualitative attributes, such as a lack of clear sexual
dimorphism and smaller total body sizes and masses.

Temperature effects on sprint speed. — All 180 scorpions were
used in the sprint speed trials. Because there is significant and
heritable variation in sprint speed ability between individuals
(Shaffer & Formanowicz 2000), each scorpion was tested at
every temperature and compared with its own performance
(repeated measures). The scorpions were divided equally into
three subgroups (each containing 20 adult males, 20 adult
females, and 20 unsexed juveniles) to systematically counter-
balance treatment effects (i.e., temperatures) across trials.
Trials on the same individual scorpion were separated by a
minimum of 48 h.

CARLSON & ROWE— PHYSIOLOGY AND DEFENSE IN BARK SCORPIONS

323

The sprint speed of each scorpion was assessed at three
temperatures; Tp = 38° C (Carlson et al, unpublished data),
25° C, and 10° C. Test temperatures were chosen to span the
range of temperatures experienced by free-ranging striped
bark scorpions (Brown et al. 2002; Yamashita 2004; Carlson et
al., unpublished data).

Sprint speed trials were performed using the same gradient
track unit employed in earlier temperature preference trials
(Carlson et al, unpublished data), consisting of a 3.2-mm
thick copper base, 1.22 m long by 30.5 cm wide. 10.2-cm-tall
aluminum sides enclosed the base and also divided the unit
lengthwise into three separate tracks, each approximately
10.2 cm wide by 1.22 m long. The base was covered with
approximately 1 cm of fine sand. The experimental tempera-
tures were achieved by placing the track, with its copper
bottom, on six hot plates running the length of the unit for the
38° C trials, and on a continuous line of ice packs for the 10° C
condition; since the experimental room was kept as close as
possible to 25° C, no heating or cooling was needed for this
latter treatment. All of the scorpions to be tested at a given
temperature were individually contained in small acetate
cylinders (~ 6 cm X 6 cm, constructed from overhead
transparency sheets) placed in one of the tracks to prevent
them from becoming fatigued from excessive movement. They
were acclimated to the temperature of the unit for at least
6 min; pilot tests, performed with a Sensortek BAT- 12
thermocouple connected to a Harvard Apparatus microprobe
inserted into the mesosoma (abdomen), show this is more than
sufficient time for a scorpion’s Tb to reach the temperature of
the substrate upon which it rests. The temperature of the
substrate (T^) inside each scorpion’s acetate containment
cylinder, and thus the approximate Tb of that scorpion, was
taken remotely (preventing disturbance to the scorpion) using
an Extech model RHIOI infrared thermometer immediately
prior to sprint testing.

Sprint speed trials were conducted in the two tracks not
used for containment/acclimation. Because of negative pho-
tokinetic behaviors in scorpions (Abushama 1964; Warburg &
Polls 1990), the end of the track was shaded while the
beginning was well-lit to encourage unidirectional movement.
Scorpions were placed on the track and, if necessary, startled
with a tap on the metasoma to induce running. A stopwatch
accurate to 0.01 s was used to record the time elapsed from
crossing the starting line to either stopping or reaching a
50 cm “finish line” marked on the track. The distance at the
final point was measured for individuals who stopped short.
Each individual was induced to sprint three times at each of its
test temperatures; within a trial (i.e., test temperature),
scorpions were allowed to rest for several minutes between
successive sprints. The highest speed achieved by an individual
in its three sprints at a given test temperature was used for
later analysis. Three individuals (two juveniles and one adult
male) that refused to sprint during one or more of its test
temperatures were removed from the analysis, reducing the
sample size to 177. The sand in the apparatus runway was
shifted and smoothed to even out the surface between tests on
different individuals.

Temperature effects on sting speed. — A subset of 54
individuals (18 adult females, 18 adult males, and 18 unsexed
juveniles) of the original 180 marked scorpions from the sprint

speed trials were selected for sting trials. In the same manner
as the sprint speed measurements, these scorpions were split
into three equal-sized subgroups that were tested at all three
temperatures in counterbalanced order. Once again, successive
trials on the same scorpion were separated by a minimum of
48 h. One adult male died before completing all three of his
trials and was removed from the analyses.

Trials were conducted using a 9.5-mm thick copper plate,
approximately 19 cm X 16 cm, on top of which was placed a
thin layer of fine sand. An acetate cylinder (approximately
7 cm tall, 8 cm long and 2.5 cm wide at the center) similar to
those used in the sprint trials was used to contain the scorpion,
with the simple modification of partially flattening the cylinder
into an ellipse to keep the scorpion oriented perpendicularly to
the video camera (see below). Test temperatures were induced
by placing the copper plate on ice (10° C), on the lab bench
(25° C), or on a hot plate (38° C), using insulation and spacers
to fine-tune the temperature. A sting-eliciting probe was
constructed from a 15-cm stick, onto which a 2-cm X 2.5-cm
piece of index card was affixed to one end as a target.

Scorpions were placed individually into the acetate ellipse and
allowed 6 min to equilibrate to the temperature of the testing
apparatus. As with sprint trials, the Tg inside each scorpion’s
acetate ellipse was taken immediately preceding its sting test.
Each test was filmed with a Canon XL2 digital video camera at
30 fps. To elicit a sting, the probe, target end down, was lowered
slowly from above and pressed firmly on the 4**^ or 5*’’ tergite
(dorsal cuticular segment) of the scorpion; both the scorpion
and the target were oriented perpendicular to the camera. The
probe was held in place for at least two seconds after the
scorpion stung. Scorpions that did not sting within 10 s were
characterized as failing to sting and the trial was terminated.

Video was analyzed frame-by-frame. Elapsed time from first
contact of the probe with the scorpion’s body to the moment
when the scorpion’s aculeus (stinger) touched the target was
recorded as sting latency. The number of separate contacts
between the aculeus and the target (i.e., number of stings) in
the following 2 s was used to calculate a sting rate (stings/s).
Individuals whose aculeus failed to contact the target were
excluded from sting latency analyses. Occasionally, a scorpi-
on’s aculeus stuck in the target, or lodged between the target
and the scorpion’s body; individuals whose aculeus remained
stuck for more than 1 s of the 2-s “post-first-sting” filming
window were excluded from sting rate analyses.

The speed with which a scorpion can deliver a sting might be
influenced not only by temperature, age, or gender, but also by
the starting position of its telson (the last tail segment, to
which the aculeus is attached) and its metasoma (tail). We
therefore used the videotapes to quantify the starting position
of each scorpion’s telson along its longitudinal axis, employing
an ordinal scale from 1-7 (Fig. 2a), generating an estimate of
metasomal “curling”; similarly, we used a three-level ordinal
scale to quantify the lateral (sagital) angle of each scorpion’s
metasoma (Fig. 2b) just prior to being touched with the sting
stimulus.

Dehydration effects on sprint speed. — From the marked
scorpions that were not employed in the sting experiments, 60
individuals were randomly selected and divided into a control
and a treatment group, each with 10 adult males, 10 adult
females, and 10 juveniles. All scorpions were measured again

324

THE JOURNAL OF ARACHNOLOGY

Upright Angled Resting

b

Figure 2. — Ordinal system for quantifying a) degrees of metasomal
curling and b) metasomal angle in C. vittatus just prior to delivering
a sting.

for sprint speed at room temperature (25° C) using the same
procedure described earlier. The treatment group was then
placed in a glass desiccator with 3.2 cm of Drierite layering the
bottom; a porcelain plate covered with a paper towel was used
to prevent scorpions from making direct contact with the
desiccant. The entire unit, desiccator plus scorpions, was then
placed inside a Percival model I36LLC8 environmental
chamber for 70 h at 36° C. The control group was placed in
a twin desiccator, again with a plate and paper towel barrier,
but with petri dishes of water supplied for drinking and no
Drierite. The control group was kept in the environmental
chamber during the same period as the experimental group,
for which mass losses of 10-20% were targeted.

After desiccation, all the scorpions were again tested at 25°
C for sprint speed. The change in a scorpion’s sprint speed was
calculated as a percent of its original speed. The paper towel
and porcelain plate barrier was not entirely effective; scorpions
that slipped past the barrier and came in contact with either
the desiccant or water condensed on the bottom were excluded
from subsequent analyses, reducing the control and treatment
group sample sizes from 30 each to 19 and 22, respectively.

Statistical analysis. — A mixed-design, repeated measures
ANOVA, with one between-group effect (age/sex) and one
within-subject effect (temperature), was performed on the sprint
speed, sting speed, and sting rate data. The possible influences of
sensitization, habituation, or fatigue across trials was assessed
by substituting “trial number” for “temperature” in a second set
of ANOVAs. Dependent variables (DVs) in each of these six
ANOVAs were tested for the standard parametric assumptions
of normality, homogeneity of variances, and repeated-measures
sphericity (Keppel 1991). Two of the six DVs (sprint speed and
sting latency with temperature) violated the sphericity assump-
tion and were thus subjected to the conservative Greenhouse-
Geisser correction (Keppel 1991, Field 2005). One DV (sting
latency with temperature) violated the normality and homoge-
neity assumptions and could not be corrected using a data
transformation; we therefore applied a conservative level of
significance {P <0.01) to this DV to minimize any risk of Type I
error (Keppel 1991). Effect sizes (strengths of association) are
reported as partial (partial eta squared). Multiple compar-
isons for any of the mixed-design ANOVAs with significant
main effects were conducted using Bonferroni’s adjustments, the
preferred post-hoc test for repeated measures (Field 2005).

Two-way independent ANOVAs were used to assess the
influences of dehydration and age/sex on two dependent
variables, the change in a scorpion’s body mass and the change
in its sprint speed following ~ 3 days at Tb = 36° C. Both DVs
were tested for the standard parametric assumptions of
normality and variance homogeneity. Sprint speeds were
significantly non-normal and could not be transformed to
meet this assumption; we therefore applied a conservative level
of significance (P < 0.01) to this DV to minimize the risk of
Type I error (Keppel 1991). Effect sizes are reported as partial
p-; Bonferroni adjustments were again used to assess the
significance of multiple comparisons.

Two dichotomous variables were analyzed using the
Cochran’s Q test for repeated measures (Siegel 1956): whether
or not the scorpion sprinted the full 50 cm or stopped short;
and whether or not a scorpion’s sting “hit” the target. The
influence of the starting positions of a scorpion’s telson
(metasomal curling) and metasoma (metasomal angle) on
sting latency were analyzed with a Friedman’s non-parametric
repeated measures ANOVA, followed by post-hoc tests using
Wilcoxon’s signed-rank tests (Siegel 1956; Field 2005). The
effect of metasomal curling and metasomal angle on sting
latency was further explored by conducting, for each test
temperature separately, a non-parametric Spearman’s rank
order correlation (Siegel 1956; Field 2005).

The computer program SPSS (Version 15.0, Chicago,
Illinois) was used to perform statistical analyses. Descriptive
results are presented in the text as means +/- 1 SE.

RESULTS

Sprint distances. — The experimental protocol was very
effective in eliciting “straight-line” sprints from all 177
scorpions tested at each of their three body temperatures.
Indeed, 88% of the trials (467/531) resulted in the scorpion
sprinting the full 50 cm marked on the track. The average
distance of the 64 sprints (12%) that stopped short of the 50-
cm mark did not differ considerably across the three test
temperatures (30.4 ± 9.14 cm at 10° C, 35.1 ± 5.37 cm at 25°

CARLSON & ROWE— PHYSIOLOGY AND DEFENSE IN BARK SCORPIONS

325

25

24

23

22

21

20

19

18

17

16

=fi

male female ju¥enile

Age/SeM Category

Figure 3. — Mean + SE sprint speeds of C. vittatus as a function of
age and gender. Different letters signify different mean values at P

< 0.05.

C, and 35.0 ± 6.20 cm at 38° C, respectively). The proportion
of “short sprints” was, however, significantly influenced by a
scorpion’s body temperature: only 10/177 tests (5.65%) at 25°
C and 5/177 tests (2.82%) at 38° C resulted in the scorpion
stopping short, compared to 49/177 (27.68%) truncated runs
for scorpions at 10° C (Q2 = 64.48, P < 0.001). Moreover, the
10 shortest runs (ranging from 10-23 cm) recorded in all 531
trials were for scorpions tested at their coolest temperature.

Sprint speeds. — There was a highly significant difference in the
sprint speeds between each age/sex category {F = 34.2; df = 2,
174; F < 0.001; Fig. 3), with males sprinting fastest (23.1 ±
0.51 cm/s), followed by females (19.8 ± 0.50 cm/s) and then
juveniles (17.1 ± 0.51 cm/s); the partial rf for this main effect
was 0.282. Post-hoc Bonferroni comparisons show that male
sprint speeds were significantly faster than either females or
juveniles {P < 0.001), while females were faster than juveniles (P
= 0.001). It is worth noting that 57 of the 60 female scorpions
involved in these sprint tests were gravid; when we removed the
three non-gravid females (one each from the three subgroups of
adult females) from the analysis, results were unchanged.

Temperature significantly affected sprint speeds (F =
1712.42; df = 1.45, 252.67; P < 0.001; Fig. 4): at the warm
temperature (38° C), scorpions averaged speeds of 32.8 ±
0.65 cm/s; at medium temperatures (25° C), their speed
averaged 20.4 ± 0.37 cm/s; at the cooler temperature (10°
C), speeds averaged only 6.8 ± 0.15 cm/s. The partial rj^ for
this main effect was 0.908. Post-hoc Bonferroni comparisons
revealed significant differences between each of the tempera-
ture treatments (P < 0.001). The positive effects of increasing
body temperature on sprint speeds appears more pronounced
in males than in females and juveniles, as evidenced by a small
(partial = 0.142) but nonetheless significant interaction
between age/sex and temperature (F = 14.36; df = 2.94,
252.67; P < 0.001). There was no significant effect of trial
number on sprint speed (F = 0.42; df = 2, 348; P = 0.657), nor
was there a significant interaction between trial number and
age/sex (F = 0.93; df = 4, 348; P = 0.983).

Although our test apparatus was unsophisticated (i.e., a
copper-bottomed runway placed on ice bags, a counter top, or

40 I

10 °C 25 °C 38 °C

Temperature

Figure 4. — Mean + SE sprint speeds of C. vittatus as a function of
temperature. Different letters signify different mean values at F < 0.05.

hot plates), the variability in our targeted treatment temper-
atures (10, 25, and 38° C) was small; the Tg (and, thus, the Tb)
for scorpions during their 10° C trial was 10.4 ± 0.20° C; for
scorpions during their 25° C trial it was 25.4 ± 0.05° C; and
for scorpions in their 38° C trial, it was 40.9 ± 0.08° C. The
low variability in substrate temperatures, recorded as they
were along the full length of the runway, attests to the
effectiveness of the copper plate in diffusing temperatures
evenly across the apparatus.

Sting speeds and effectiveness. — Eliciting stings proved more
difficult than inducing sprints. Indeed, 24.5% of the 159 sting
trials ended with the scorpion either not stinging or attempting
to sting but missing the target. Temperature appeared to
influence stinging success, as 84.9% and 75.5% of the stings
delivered by scorpions when at 25° C and 38° C Tb hit the
target, respectively, while only 66.0% of the stings delivered by
scorpions at 10° C were so effective {Q2 = 6.25, P = 0.044).

For those scorpions who managed to sting the target in each
test, temperature had a significant effect on their latency to
sting (F = 13.27; df = 1.06, 23.26; P = 0.001; Fig. 5); cooler
scorpions (10° C) took significantly longer (Bonferroni
adjustment; P < 0.001) to deliver a sting (1.67 ± 0.40 s) than
either the medium (25° C; 0.29 ± 0.05 s) or warm (38° C; 0.32
± 0.08 s) scorpions, which did not themselves differ. The
partial rj^ for this main effect was 0.376. The age and sex of the
scorpion had no effect on sting latency (F = 1.44; df 2, 22; P
= 0.258), nor was there a significant interaction between age/
sex and temperature (F = 1.77; df = 2.11, 23.26; P = 0.191).
Note that 17 of the 18 female scorpions included in this
analysis were gravid; when the non-gravid female was
removed from the analysis, results were unchanged.

The significant influence of Tb on sting latency is unlikely to
be the result of systematic differences in the initial positions of
a scorpion’s telson or metasoma. There were no significant
differences in metasomal curling {■^ = 2.14; <//’= 2; F = 0.343)
or angle {-^ = 3.60; df = 2; P = 0.166) across the three test
temperatures. Moreover, neither metasomal curling (fs =
0.218; F = 0.208) nor angle (r, = 0.053; P = 0.763) was
significantly correlated with sting latency for scorpions during
their 10° C test; results at 38° C were similar, with insignificant

326

THE JOURNAL OF ARACHNOLOGY

Temperature

8 1
7 -

b

10 °C 25 °C 38 °C

Temperature

Figure 5. — Mean + SE sting latencies of C. vittatus as a function of
temperature. Different letters signify different mean values at R < 0.05.

Figure 6. — Mean + SE sting rates of C. vittatus as a function of
temperature. Different letters signify different mean values at P < 0.05.

correlations of sting latency with metasomal curling {t's =
0.049; P = 0.762) and angle (r, = 0.077; P = 0.637).
Intriguingly, both curling and angle were significantly
correlated with sting latency for scorpions at intermediate
body temperatures; e.g., at 25° C, scorpions with higher values
of metasomal curling (positions 4-6 in Fig. 2a; note that no
scorpion had a curling value of 7 at this temperature) delivered
slower stings (r^. = 0.449; P = 0.002) than individuals with
lower values (positions 1-3 in Fig. 2a). Similarly, individuals
whose metasomas where held at ~ 45° delivered slower stings
than those holding them vertically (r, = 0.491; P = 0.001);
only one individual at 25° C held its metasoma in a resting
position (Fig. 2b), preventing any meaningful comparison for
this angle. There was no significant effect of trial number on
sting latency {F = 0.34; df = 2, 44; P = 0.717), nor was there a
significant interaction between trial number and age/sex (F =
1.26; df= 4, 44; P = 0.301).

A similar pattern was found in the effects of temperature on
sting rate (F = 41.43; df = 2, 86; P < 0.001; Fig. 6); cooler
scorpions (10° C) delivered significantly fewer (Bonferroni
adjustment; P < 0.001) stings per second (2.97 ± 0.27) than
did scorpions at intermediate (25° C; 6.16 ± 0.39 stings/s) and
warm (38° C; 6.33 ± 0.32 stings/s) body temperatures, which
themselves did not differ. The partial r|“ for this main effect was
0.491. And while there was no significant interaction between
age/sex and temperature on sting rate (F = 0.32; df= 4, 86; P =
0.864), the main effect of age/sex approached significance {F =
2.81; df = 2, 43; P = 0.072). Though caution is merited, it is
noteworthy that female bark scorpions appear to deliver rapid,
probing stings (5.90 ± 0.39 stings/s) more quickly than either
juveniles (4.86 ± 0.38 stings/s) or males (4.71 ± 0.39 stings/s).
Removing the sole non-gravid female from sting rate analyses
did not influence these results. There was no significant effect of
trial number on sting rate (F = 0.58; df= 2, 86; P = 0.561), nor
was there a significant interaction between trial number and age/
sex (F = 0.44; df = 4, 86; P = 0.779).

Our test apparatus for assessing sting speed and sting rates
was, like our apparatus for assessing sprint speeds, rather
crude. It was, nonetheless, effective and reliable at generating
our targeted test temperatures. The Tj, and thus Tb, for

scorpions tested at their targeted Tb of 10° C was 9.4 ± 0.18°
C; for scorpions during their 25° C trials, Tj was 25.0 ± 0.05°
C; and for 38° C trials, T^ was 37.8 ± 0.09° C.

Dehydration effects. — The scorpions receiving the desicca-
tion treatment lost a significantly greater percentage of their
body mass (— 14.95 ± 1 .3%) than did control scorpions (—0.44
± 0.9%; F = 78.32; df = 1, 35; P < 0.001; Fig. 7). The partial
r|" for this main effect was 0.691 . The effect of age/sex on mass
loss was insignificant (F = 0.97; df= 2, 35; P = 0.391), as was
the interaction of dehydration/control with age/sex (F = 0.41;
df = 2, 35; P = 0.670).

The effect of hydration on sprint speeds was highly
significant (F = 22.84; df = 1, 35; P < 0.001; Fig. 8);
desiccated scorpions increased their sprint speed by +7.57 ±
7.1% between their pre- and post-tests, while control scorpions
decreased in speed by —27.69 ± 4.8%. The partial r|“ for this
main effect was 0.395. At first glance, the age and sex of a
scorpion also appear to influence how its sprint speed changed
(F = 3.68; df= 2, 35; P = 0.035); females suffered the greatest

20 -1

18 -

«/)

16 -

>

14 -

■c

o

12 -

QQ

10 -

H-

o

8 -

6 -

O

4 -

SO

2 -

0 -1

Control Dehydrated

Test Group

Figure 7. — Mean + SE percent of body mass lost in dehydrated
and control C. vittatus. Different letters signify different mean values
at F < 0.05.

CARLSON & ROWE— PHYSIOLOGY AND DEFENSE IN BARK SCORPIONS

327

Figure 8. — Mean + SE percent changes between pre- and post-
treatment sprint speeds for dehydrated and control C vittatus.
Different letters signify different mean values at F < 0.05.

reduction in sprint speeds (—19. 52 ± 6.5%) during their time
in the environmental chamber, followed by males (—9.80 ±
6.8%), while juveniles actually ran faster (3-10.17 ± 8.8%) after
spending ~ 3 days at 36° C. Note, however, that this DV did
not meet the parametric assumption of normality, forcing us
to reduce the level of significance (to P < 0.01) to minimize
Type I error; by this more stringent criterion, age/sex had no
significant influence on sprint speeds following desiccation.
The highly significant effect of hydration, but not age/sex, on
changes in scorpion sprint speeds was confirmed by running a
Kruskal-Wallis non-parametric ANOVA (Siegel 1956; Field
2005) on both variables; the former DV remained highly
significant = 14.37; df = 1; P < 0.001), while the latter was
not (x^ = 2.33; df — 2; P = 0.312). And finally, the interaction
between age/sex and dehydration/control was insignificant (F
= 1.21; df= 2, 35; P = 0.31).

DISCUSSION

Carlson et al. (unpublished data) show that striped bark
scorpions from two widely separated populations inhabiting
ecologically distinct habitats (one a rocky scrub desert in the
foothills of the Organ Mountains in southern New Mexico,
the other from the piney woods region of southeastern Texas)
exhibit activity patterns and microhabitat preferences favoring
high Tb’s; moreover, when tested in a thermal gradient in the
laboratory, scorpions from both populations had very high
Tp’s (36-38° C). The results presented here may help explain
why; i.e., a C. vittatus that is warm sprints significantly faster
and frequently sprints farther, delivers a quicker and more
accurate sting, and delivers more stings per second than when
it is cold. The influence of body temperature on this suite of
defensive behaviors is the most robust finding from this study,
as demonstrated by the uniformly higher effect sizes (partial
q^) for temperature than for age and gender. Temperature
appears to have a more profound effect on sprint speed than
sting efficacy, as evidenced by 1) the larger effect size (partial
= 0.908) for maximal sprint velocity than for sting latency
(0.376) or sting rate (0.491); and 2) by the significant
differences (Bonferroni multiple comparisons) in sprint speeds

across all three body temperatures (10, 25, and 38° C), but
only between the coldest (10° C) vs. two warmer (25 and 38°
C) temperatures for sting latency and rate. The preference of
striped bark scorpions for the hottest of these temperatures
may nonetheless make sense; when in the clutches of a
grasshopper mouse, even a moderately warm scorpion can
deliver a painful sting that will get it dropped, but only the
warmest and fastest scorpion might then sprint to safety
before the mouse re-attacks (Rowe & Rowe 2006).

Desiccation also had a significant effect on sprint velocity,
but not in the manner we originally predicted. Surprisingly,
scorpions that had lost nearly 15% of their body mass due to
dehydration ran faster than they did prior to desiccation;
control scorpions, whose body masses barely changed over the
same three-day period while housed at 36° C, had sprint
speeds ~ l/3'^‘* lower than their initial runs. The lack of any
substantive effects of desiccation on sprinting may have
several explanations. First, scorpions as a group use hydraulic
pressure less in limb extension than most other arachnids
(Shultz 1992), suggesting that tluid loss might have little
impact on running speed. At high (though unspecified) levels
of desiccation, however, scorpions do have difficulty moving
(Sensenig & Shultz 2004); it is therefore possible that the
amount of water loss in our trials was too low to have
produced noticeable locomotor effects. Second, scorpions
may, when suffering from dehydration stress, shift to
anaerobic metabolism (Gefen 2008), which is less sensitive to
the debilitating effects of desiccation than is aerobic metab-
olism (Weinstein 1998), allowing scorpions to maintain
comparable levels of performance across varying levels of
hydration. Finally, desiccation appears to affect activity
bursts, such as sprinting, much less than it does endurance
(Crowley 1985; Wilson & Havel 1989) mediated again through
the different sensitivities of anaerobiosis vs. aerobiosis to
desiccation. This study was not designed to measure endur-
ance, so conclusions cannot presently be drawn about any
possible inhibitory effects of desiccation on scorpion locomo-
tor performance over longer time frames. The small positive
effect of dehydration on scorpion sprint speeds (-1- 7.6% faster
following desiccation) is most likely the result of their
significant loss of body mass (Crowley 1985).

In contrast, the dramatic reduction in sprint speeds for the
control scorpions may reflect the negative physiological effects
of supporting a high T(, (36° C) for a moderately long duration
(70 h) without feeding. Studies have shown that ectotherms
subject to high temperatures, but within the normal range they
experience in the field, suffer denaturation of cellular proteins
(Hofmann & Somero 1995); replacing these proteins requires
energy that must be diverted from other tasks including,
perhaps, maintaining the structures required for limb exten-
sion and retraction. Moreover scorpions, like many ecto-
therms, exhibit temperature-sensitive metabolic rates (Lighton
et al. 2001); the energetic demands of higher TbS would reduce
glycogen stores (Sinha & Kanungo 1967), further squeezing
the pipeline that fuels locomotion. Whether desiccated
scorpions are buffered from protein damage at high TbS is
unknown. Dehydration, however, might actually protect
scorpions from depleting their carbohydrate reserves, as
desiccation significantly reduces their metabolic rates (Gefen
2008). Thus, for striped bark scorpions, the cost of maintain-

328

THE JOURNAL OF ARACHNOLOGY

ing a high Tb is imposed not by desiccation, to which they
appear well adapted, but through exhaustion of their energy
resources. This cost may help explain the lengthy hiatus that
scorpions spend between meals (Bradley 1982; Quinlan et al.
1993) in presumably cool refugia. Although speculative,
proximate cues inducing a bark scorpion to leave the safety
of its shelter could be a loss of body mass and/or a level of
desiccation that maximize its sprint speed.

The influence of age and gender on the defensive behavior of
striped bark scorpions also produced both anticipated and novel
results. In a simple but elegant study, Shaffer and Formanowicz
(1996) demonstrated that non-gravid female C. vittatus had
significantly faster sprint speeds than gravid females, who
themselves were faster than females carrying first-instar larvae
on their backs. Moreover, female sprint speed was inversely
related to the weight of the clutch they were carrying, either
internally or dorsally. Given that all but three of the 60 females
used in our sprint speed trials were gravid (and thus heavy, at
0.64 ± 0.016 gm), it is not surprising that their sprint speeds
were significantly slower than the much leaner males (0.40 ±
0.12 gm). Juveniles, although light (0.17 ± 0.007 gm), were
significantly slower than either males or females, likely resulting
from their shorter limbs or incomplete motor development. Our
results demonstrating both age and gender effects on sprint
speeds thus support and extend the original findings of Shaffer
and Formanowicz (1996). Uniquely, however, our results may
help explain the dramatic sexual differences in Centniroides
metasomas (Fig. 1) by pointing, tentatively, to a sexual
dimorphism in antipredator behavior. Gestation in C. vittatus
lasts eight months (Polis & Sissom 1990), a lengthy and risky
period for female bark scorpions. Limited by the weight of their
larger bodies and developing embryos, and thus incapable of
achieving the sprint speeds of males, female C. vittatus may have
selectively compensated for their increased vulnerability with
better stinging ability. The longer, thinner metasomas of males
appear to limit their ability to deliver fast, repetitive stings.
While possession of a tail morphology capable of probing an
enemy’s epidermis for a weak spot certainly makes sense for a
sluggish female bark scorpion, its absence in males is perplexing.
Perhaps the longer and thinner metasoma of males enhances
maneuverability, complementing their reliance on quick sprints
to safety. Thin tails with an extended reach might also prove
useful in courtship, or in male-male combat. We hope to explore
these and other functional explanations concerning sexual
differences in bark scorpion moiphology and defensive behavior
in future projects.

Our results documenting the influence of temperature on
the defensive behavior of bark scorpions appear relevant to
two additional, related aspects of their ecology; namely, their
“errancy” and their proclivity for climbing bushes. Bark
scorpions (Centruroides spp.) have earned the ecomorpholo-
gical label of “errant” (McCormick & Polis 1990; Polis 1990)
because, when hunting, they actively search for prey, quite
unlike the ambushing strategy adopted by the majority of
scorpion species. A cursorial foraging mode would, we argue,
be more likely to attract the attention of visually hunting
predators (e.g., grasshopper mice, Onycyhomys spp.; see Rowe
& Rowe 2006) than would lying in wait. The foraging behavior
of bark scorpions appears sensitive to their greater risk of
predation, as evinced by their tendency of climbing into

bushes to evade their enemies (Brown & O’Connell 2000), a
behavior that becomes even more pronounced when the moon
is bright (McReynolds 2004). Intriguingly, one report suggests
that female bark scorpions are more likely than males to use
bushes (Yamashita 2004), perhaps reflecting the females’
greater vulnerability imposed by their slower sprint speeds. In
a recent report, McReynolds (2008) shows that striped bark
scorpions shift from foraging in vegetation to foraging on the
ground when nighttime air temperatures are extremely warm
(> 30° C); the author suggests this shift may be due to
increased availability of terrestrial prey during the hottest
months of the year. A complementary interpretation is that
bark scorpions are more likely to be found foraging on the
ground when temperatures are warm enough to permit quick
sprints to safety.

ACKNOWLEDGMENTS

We would like to thank Dr. Jeffrey Port, Dr. Linden
Higgins, and two anonymous reviewers for helpful comments
on earlier drafts of this manuscript. We are indebted to Drs.
Jerry Cook and Sibyl Bucheli for the use of their environ-
mental chamber. Special thanks go to the biologists and staff
of the Las Cruces, New Mexico, office of the Bureau of Land
Management for their invaluable logistical support. This
research was funded by the Department of Biological Sciences
and the College of Arts & Sciences at Sam Houston State
University, and by the National Science Foundation Research
Experience for Undergraduates (Award #0649187). Collec-
tion of the scorpions used in this study and the research
protocols employed were compliant with all current state and
federal laws in the USA.

LITERATURE CITED

Abushama, F.T. 1964. On the behaviour and sensory physiology of
the scorpion Leiurus qiimqiiestriatus (H. & E.). Animal Behaviour
12:140-153.

Anderson, J.F. & K.N. Prestwich. 1975. The fluid pressure pumps of
spiders (Chelicerata, Araneae). Zeitschrift fiir Morphologie der
Tiere 81:257-277.

Angilletta, M.J., Jr., T.D. Steury, T.D. Sears & M.W. Sears. 2004.
Temperature, growth rate, and body size in ectotherms: fitting
pieces of a life-history puzzle. Integrative and Comparative
Biology 44:498-509.

Ayal, Y. 2007. Trophic structure and the role of predation in shaping
hot desert communities. Journal of Arid Environments
68:171-187.

Bauwens, D., T. Garland, Jr., A.M. Castilla & R. Van Damme. 1995.
Evolution of sprint speed in lacertid lizards: morphological,
physiological, and behavioral covariation. Evolution 49:848-863.
Bennett, A.F. 1984. Thermal dependence of muscle function.

American Journal of Physiology 247:R217-R229.

Bobka, M.S., R.G. Jaeger & D.C. McNaught. 1981. Temperature
dependent assimilation efficiencies of two species of terrestrial
salamanders. Copeia 1981:417-421.

Bradley, R. 1982. Digestion time and reemergence in the desert
grassland scorpion Paruroc tonus iitahensis (Williams) (Scorpio-
nida, Vaejovidae). Oecologia 55:316-318.

Bradley, R.A. 1988. The influence of weather and biotic factors on the
behaviour of the scorpion (Parwoctonus iitahensis). Journal of
Animal Ecology 57:533-551.

Brown, C.A. & D.J. O’Connell. 2000. Plant climbing behavior in the
scorpion Centruroides vittatus. American Midland Naturalist
144:406-418.

CARLSON & ROWE— PHYSIOLOGY AND DEFENSE IN BARK SCORPIONS

329

Brown, C.A., J.M. Davis, DJ. O’Connell & D.R. Formanowicz, Jr.
2002. Surface density and nocturnal activity in a west Texas
assemblage of scorpions. Southwestern Naturalist 47:409-419.

Christian, K.A. & C.R. Tracy. 1981. The effect of the thermal
environment on the ability of hatchling Galapagos land iguanas to
avoid predation during dispersal. Oecologia 49:218-223.

Claussen, D.L., R.A. Hopper & A.M. Sanker. 2000. The effects of
temperature, body size, and hydration state on the terrestrial
locomotion of the crayfish Orconectes rusticus. Journal of
Crustacean Biology 20:218-223.

Crowley, S.R. 1985. Insensitivity to desiccation of sprint running
performance in the lizard, Sceloporus imdulatus. Journal of
Herpetology 19:171-174.

Field, A. 2005. Discovering Statistics Using SPSS, 2"*^ edition. Sage
Publications, London.

Forsman, A. 1999. Variation in thermal sensitivity of performance
among colour morphs of a pygmy grasshopper. Journal of
Evolutionary Biology 16:869-878.

Gefen, E. 2008. Sexual dimorphism in desiccation responses of the
sand scorpion Smeringurus mesaensis (Vaejovidae). Journal of
Insect Physiology 54:798-805.

Gefen, E. & A. Ar. 2004. Comparative water relations of four species
of scorpions in Israel: evidence for phylogenetic differences.
Journal of Experimental Biology 207:1017-1025.

Gefen, E. & A. Ar. 2005. The effect of desiccation on water
management and compartmentalisation in scorpions: the hepato-
pancreas as a water reservoir. Journal of Experimental Biology
208:1887-1894.

Gefen, E. & A. Ar. 2006. Temperature dependence of water loss rates
in scorpions and its effect on the distribution of Buthotus judaicus
(Buthidae) in Israel. Comparative Biochemistry and Physiology A
144:58-62.

Greenwald, O.E. 1974. Thermal dependence of striking and prey
capture by gopher snakes. Copeia 1974:141-148.

Hadley, N.F. 1970. Water relations of the desert scorpion, Hadrurus
arizonensis. Journal of Experimental Biology 53:547-558.

Hadley, N.F. 1974. Adaptational biology of desert scorpions. Journal
of Arachnology 2:1 1-23.

Hadley, N.F. & S.C. Williams. 1968. Surface activities of some North
American scorpions in relation to feeding. Ecology 49:726-734.

Hertz, P.E., R.B. Huey & T. Garland, Jr. 1988. Time budgets,
thermoregulation, and maximal locomotor performance: are
reptiles Olympians or Boy Scouts? American Zoologist
28:927-938.

Hofmann, G.E. & G.N. Somero. 1995. Evidence for protein damage
at environmental temperatures; seasonal changes in levels of
ubiquitin conjugates and hsp70 in the intertidal mussel Mytilus
trossulus. Journal of Experimental Biology 198:1509-1518.

Huey, R.B. 1982. Temperature, physiology, and the ecology of
reptiles. Pp. 25-91. In Biology of the Reptilia, Volume 12.
Physiology C, Physiological Ecology (C. Cans & F.H. Pough,
eds.). Academic Press, London.

Huey, R.B. & M. Slatkin. 1976. Cost and benefits of lizard
thermoregulation. Quarterly Review of Biology 51:363-384.

Keppel, G. 1991. Design and Analysis: a Researcher’s Handbook, 3'''^
edition. Prentice Hall, Upper Saddle River, New Jersey.

Lighten, J.R.B., P.H. Brownell, B. Joos & R.J. Turner. 2001. Low
metabolic rate in scorpions: implications for population biomass
and cannibalism. Journal of Experimental Biology 204:607-613.

Machin, K.E., J.W.S. Pringle & M. Tamasige. 1962. The physiology
of insect fibrillar muscle. IV. The effect of temperature on a beetle
flight muscle. Proceedings of the Royal Society of London, Series B
155:493^99.

McAlister, W.H. 1966. The aggregating tendency of Centruroides
vittatus Say (Arachnida: Scorpionida). Texas Journal of Science
18:80-84.

McCormick, S.J. & G.A. Polis. 1990. Prey, predators, and parasites.
Pp. 294-320. In The Biology of the Scorpions. (G.A. Polis, ed.).
Stanford University Press, Stanford, California.

McReynolds, C.N. 2004. Temporal patterns in microhabitat use for
the scorpion Centruroides vittatus (Scorpiones:, Buthidae). Eu-
scorpius 17:35^5.

McReynolds, C.N. 2008. Microhabitat preferences for the errant
scorpion, Centruroides vittatus (Scorpiones, Buthidae). Journal of
Arachnology 36:557-564.

Miller, K. 1982. Effect of temperature on sprint performance in the
frog Xenopus laevis and the salamander Necturus maculosus.
Copeia 1982:695-698.

Moore, F.R. & R.E. Gatten, Jr. 1989. Locomotor performance of
hydrated, dehydrated, and osmotically stressed anuran amphibi-
ans. Herpetologica 45:101-1 10.

Polis, G.A. 1980. Seasonal patterns and age-specific variation in the
surface activity of a population of desert scorpions in relation to
environmental factors. Journal of Animal Ecology 49:1-18.

Polis, G.A. 1990. Ecology. Pp. 247-293. In The Biology of the
Scorpions (G.A. Polis, ed.). Stanford University Press, Stanford,
California.

Polis, G.A. & W.R. Lourenqo. 1986. Sociality among scorpions.
Proceedings of the 10th International Congress of Arachnology,
Jaca 1:111-115.

Polis, G.A. & W.D. Sissom. 1990. Life history. Pp. 161-223. In The
Biology of the Scorpions (G.A. Polis, ed.). Stanford University
Press, Stanford, California.

Polis, G.A., W.D. Sissom & S.J. McCormick. 1981. Predators of
scorpions: field data and a review. Journal of Arid Environments
4:309-326.

Quinlain, T.G., M.C. Calver & G.T. Smith. 1993. Immunological
determination of digestive rates in the syntopic scorpions
Urodactus armatus Pocock and Urodactus novaehoUandiae Peters.
Oecologia 95:459-462.

Rowe, A.H. & M.P. Rowe. 2006. Risk assessment of grasshopper
mice (Onychomys spp.) feeding on neurotoxic prey {Centruroides
spp.). Animal Behaviour 71:725-734.

Rowe, A.H. & M.P. Rowe. 2008. Physiological resistance of
grasshopper mice (Onychomys spp.) to Arizona bark scorpion
(Centruroides exUicauda) venom. Toxicon 52:597-605.

Rowe, M.P. & D.H. Owings. 1990. Probing, assessment, and
management during interactions between ground squirrels and
rattlesnakes. Part 1: Risks related to rattlesnake size and body
temperature. Ethology 86:237-249.

Rowe, M.P. & D.H. Owings. 1996. Probing, assessment, and
management during interactions between ground squirrels and
rattlesnakes. Part 2: Cues afforded by rattlesnake rattling.
Ethology 102:85^874.

Schmalhofer, V.R. & T.M. Casey. 1999. Crab spider hunting perfor-
mance is temperature insensitive. Ecological Entomology 24:345-353.

Schoener, T.W. 1971. Theory of feeding strategies. Annual Review of
Ecology and Systematics 2:369-404.

Sensenig, A.T. & J.W. Shultz. 2004. Elastic energy storage in the
pedipalpal joints of scorpions and sun-spiders (Arachnida,
Scorpiones, Solifugae). Journal of Arachnology 32:1-10.

Shaffer, L.R. & D.R. Formanowicz, Jr. 1996. A cost of viviparity and
parental care in scorpions: reduced sprint speed and behavioural
compensation. Animal Behaviour 51:1017-1024.

Shaffer, L.R. & D.R. Formanowicz, Jr. 2000. Sprint speeds of
juvenile scorpions: among family differences and parent offspring
correlations. Journal of Insect Behavior 13:45-53.

Shultz, J.W. 1992. Muscle firing patterns in two arachnids using
different methods of propulsive leg extension. Journal of Exper-
imental Biology 162:313-329.

Sinha, R.C. & M.S. Kanungo. 1967. Effect of starvation on the scorpion
Palamnaeus hengalensis. Physiological Zoology 40:386-390.

330

THE JOURNAL OF ARACHNOLOGY

Siegel, S. 1956. Nonparametric Statistics. McGraw-Hill, New York.

Skutelsky, O. 1996. Predation risk and state-dependent foraging in
scorpions: effects of moonlight on foraging in the scorpion Buthus
occitcmus. Animal Behaviour 52:49-57.

Slobodchikoff, C.N. 1983. Water balance and temperature prefer-
ences, and their role in regulating activity times of tenebrionid
beetles. Oikos 40:113-119.

Swaisgood, R.R., M.P. Rowe & D.H. Owings. 1999. Assessment of
rattlesnake dangerousness by California ground squirrels: infor-
mation extraction from rattling sounds. Animal Behaviour
57:1301-1310.

Swaisgood, R.R., D.H. Owings & M.P. Rowe. 2003. Antipredator
responses of California ground squirrels to rattlesnakes and
rattling sounds: the roles of sex, reproductive parity, and offspring
age in assessment and decision-making rules. Behavioral Ecology
and Sociobiology 55:22-31.

Turner, J.S., C.R. Tracy, B. Weigler & T. Baynes. 1985. Burst
swimming of alligators and the effect of temperature. Journal of
Herpetology 19:450-458.

van Berkum, F.H., R.B. Huey & B.A. Adams. 1986. Physiological
consequences of thermoregulation in a tropical lizard {Aineivci
festiva). Physiological Zoology 59:464-472.

Waldschmidt, S. & C.R. Tracy. 1983. Interactions between a lizard
and its thermal environment: implications for sprint performance
and space utilization in the lizard Uta stcmshuriami. Ecology
64:476-484.

Warburg, M.R. & A. Ben-Horin. 1981. The response to temperature
gradients of scorpions from mesic and xeric habitats. Comparative
Biochemistry and Physiology, A 68:277-279.

Warburg, M.R. & G.A. Polis. 1990. Behavioral responses, rhythms,
and activity patterns. Pp. 224—246. In The Biology of the Scorpions
(G.A. Polis, ed.). Stanford University Press, Stanford, California.

Webb, J.K. & R. Shine. 1998. Thermoregulation by a nocturnal
elapid snake (Hoplocephalus himgaroides) in southeastern Austra-
lia. Physiological Zoology 71:680-692.

Weinstein, R.B. 1998. Effects of temperature and water loss on
terrestrial locomotor performance in land crabs: integrating
laboratory and field studies. American Zoologist 38:518-527.

Whitmore, D.H., R. Gonzalez & J.G. Baust. 1985. Scorpion cold
hardiness. Physiological Zoology 58:526-537.

Wilson, B.S. & P.J. Havel. 1989. Dehydration reduces the endurance
running capacity of the lizard Uta stansbiiriana. Copeia
1989:1052-1056.

Yamashita, T. 2004. Surface activity, biomass, and phenology of the
striped scorpion, Centrnroicles vittatiis (Buthidae) in Arkansas,
USA. Euscorpius 17:25-33.

Zhang, Y.-P. & X. Ji. 2004. The thermal dependence of food
assimilation and locomotor performance in southern grass lizards,
Takvdromus sexlineatus (Lacertidae). Journal of Thermal Biology
29:45-53.

Manuscript received 27 January 2009, revised 29 July 2009.

2009. The Journal of Arachnology 37:331-337

Cytogenetics of three species of scorpions of the genus Brachistosternus from Argentina

(Scorpiones: Bothriuridae)

Sergio G. Rodriguez-Gil‘, Andres A. OJanguren-Affiiastro-"^ Leonel M. Barral', Cristina L. Scioscia-, and Liliana M.
Mola’ -^: 'Laboratorio de Citogenetica y Evolucion, Departamento de Ecologia, Genetica y Evolucion, Facultad de
Ciencias Exactas y Naturales, Universidad de Buenos Aires, Ciudad Universitaria, Pabellon 2, Cuarto Piso,
(CI428EHA) Buenos Aires, Argentina; ^Division Aracnologia, Museo Argentino de Ciencias Naturales “Bernardino
Rivadavia” (CONICET), Avenida Angel Gallardo 470 {C1405DJR) Buenos Aires, Argentina; ■''Carrera del Investigador
(CONICET)

Abstract. Meiotic studies on three phylogenetically distant species of the genus Brachistosternus Pocock from Argentina
were conducted. One species is from the subgenus Ministermis Francke 1985, B. ferrugineus Thorell 1876, and two species
are from the subgenus Brachistosternus Pocock 1893, B. montanus Roig-Alsina 1977 (Andean species group), and
morphologically different populations of B. pentheri Mello-Leitao 1931 (plains species group). All species showed
achiasmatic meiosis, absence of heteromorphic bivalents, and bibrachial and monobrachial chromosomes of different sizes.

Males of Brachistosternus ferrugineus, B. montanus. and one population of B. pentheri have 2n = 46. Males of the typical
populations of B. pentheri have In = 42. These results suggest that B. pentheri may comprise two species.

Keywords: Achiasmatic meiosis. Neotropics

The family Bothriuridae contains about 150 described
species; it shows a Gondwanan distribution and has diversified
mainly in southern South America. The systematics of this
family has been resolved satisfactorily and the taxonomic
status of most genera and species is well defined (Prendini
2003; Ochoa 2004a; Ojanguren-Affilastro & Ramirez 2009),
making it particularly suitable for studying patterns of
cytogenetic evolution. Chromosome number in Bothriuridae
varies between 28 and 50 (Piza de Toledo 1947; Ferreira 1968;
Giacomozzi 1977). Cytogenetic analyses have been performed
on seven species, four species from the genus Bothriurus Peters
1861, Timogenes elegans (Mello-Leitao 1931) and two species
of Brachistosternus Pocock 1893 {B. pentheri Mello-Leitao
1931 and B. alienus Lonnberg 1898) (Piza de Toledo 1947;
Ferreira 1968; Giacomozzi 1977). The present study focuses
on exploring cytological diversity among species and popula-
tions of Brachistosternus from Argentina.

Brachistosternus is the most diverse genus of the family, with
about 40 known species (Ochoa 2002, 2004b; Ochoa & Acosta
2002; Ojanguren-Affilastro 2003a, b, 2005a, b; Ochoa &
Ojanguren-Affilastro 2007; Ojanguren-Affilastro et al. 2007a,
b; Ojanguren-Affilastro & Scioscia 2007). It inhabits arid areas
in the western and southern parts of South America, from
Ecuador (Cekalovic 1969) to southern Argentinean Patagonia
(Ojanguren-Affilastro 2003b). The genus is divided into two
subgenera (Ojanguren-Affilastro & Ramirez 2009), namely
Brachistosternus Pocock 1893 and Ministernus Francke 1985.
Both subgenera are present in Argentina. The subgenus
Brachistosternus includes two large groups, one including
lowland or plains species and the other mountain species from
the Andes from altitudes between 2500 and 4500 m asl.

Here we report a cytogenetic study of three phylogenetically
distant species of Brachistosternus in order to reveal possible
chromosome variations in the genus. Brachistosternus ferrugi-

Corresponding author. E-mail: andres.ojanguren(^gmail.com;
ojanguren(^macn.gov.ar)

neus (Thorell 1876) was selected as a representative of the
subgenus Ministernus. This species is widely distributed in
central and northern Argentina, eastern Bolivia, Paraguay and
possibly in southwestern Brazil (Maury 1974). Two species
were selected as representatives of the subgenus Brachistos-
ternus, one belonging to the Andean group and the other to
the plains group. Brachistosternus montanus Roig-Alsina 1977
(Andean group) is restricted to high-altitude areas of the
Andean region (2700 to 3500 m asl) in central-western
Argentina, in the provinces of Mendoza, San Juan and La
Rioja (Ojanguren-Affilastro 2003a; Roig-Alsina 1977). Bra-
chistosternus pentheri Mello-Leitao 1931 (plains group) is also
found exclusively in Argentina, with a widespread distribution
from Salta province to the southern part of Buenos Aires
province. The morph of the northernmost B. pentheri found in
the provinces of La Rioja, Catamarca and Salta (here
designated as the northern morph) differs slightly from the
type material by larger size and less pronounced pigmentation
(Roig-Alsina & Maury 1984; Ojanguren-Affilastro 2005b).
Both morphs of B. pentheri were analyzed cytogenetically to
determine whether they also differ in karyotype.

METHODS

Specimens. — Brachistosternus ferrugineus: The specimens be-
long to two populations from an area near the center of the
known distribution of the species: three males from the locality of
Chepes, La Rioja province, Argentina (31°21 '00"S; 66°35'60"W),
and three males from the locality of San Marcos Sierra, Cordoba
province, Argentina (30°46'60"S; 64°39'00"W). Brachistosternus
montanus: Four males were obtained from the locality of Laguna
Brava, La Rioja province (28°25'50"S; 69°00'3L3"W). This
population has been previously mentioned under the name B.
affinis montanus (Ojanguren-Affilastro 2003a) because it shows
minor morphological differences from the typical morph of the
species, which is present in San Juan and Mendoza provinces.
However, we consider this population to be the same species after
examining many additional specimens.

331

332

THE JOURNAL OF ARACHNOLOGY

♦

#

b

Figure F a-c. — Bracliistostennis fernigmeus (n = 23). a. Late postpachytene; b. Two prometaphase I; c. Metaphase II. — d-f.
Brachistostermts montcimis (2n = 46, n = 23). d. Pachytene; e. Prometaphase I; f. Prometaphase II. Scale bar: 10 pm.

Brachistostemus pentherv. Four males were obtained from
the locality of Villa Union, northern La Rioja province
(29°18'00"S; 68°12'00"W) and belong to the northern morph.
The typical morph was represented by three males from
Chepes (southern La Rioja province) (31°21'00"S; 66°35'
60"W), and one male from Oriente (coastal area of southeast-
ern Buenos Aires province) (38°36'29"S; 60°37'07.3"W).

Cytogenetic methods. — All specimens were carried alive to
laboratory and killed by cooling in a refrigerator. Their
gonads were dissected in a physiological saline solution,
swollen in hypotonic solution (0.56% KCl) for 10 min, and
then fixed in a mixture of ethanol:chloroform:acetic acid
(6:3:1 ). A piece of testis was placed on a slide, dissociated in a
drop of 60% acetic acid with tungsten needles, and spread on
the slide using a heating histological plate at approximately
45° C. Finally, the preparations were air-dried and stained
with 3% Giemsa solution in water (pH = 7.4) for 10 min. Five
postpachytene-prometaphase I of each cytotype of B. pentheri
were measured to determine the meiotic karyotype. Bivalents
measurements were made using the computer application
Micromeasure version 3.3 (Reeves & Tear 2000). The relative
length of each bivalent was calculated as a percentage of total
haploid complement length (TCL). The idiogram of each

cytotype was drawn on the basis of the relative percentage of
each bivalent length to the TCL.

RESULTS

Subgenus Ministerniis Francke 1985
(Figs, la-c; Table 1)

Brachistostemus fermgineus has a karyotype of 2n = 46.
No positively heteropycnotic bodies were observed at early
prophase 1. After pachytene, bivalents have no chiasma;
homologous chromosomes lie parallel to each other, and
condense gradually during prophase I and metaphase 1. All
bivalents are homomorphic (Figs, la, b). Chromosome plates of
prometaphase and metaphase II consist of bibrachial (meta- or
submetacentric) and monobrachial chromosomes of different
sizes (Fig. Ic). No differences were observed between the
specimens from the two localities.

Subgenus Brachistostemus Pocock 1893
(Figs. Id-f, 2a-f, 3a-d; Table 1)

The chromosome complement of B. montanus consists of 46
chromosomes. At early prophase I no positively heteropyc-

RODRIGUEZ-GIL ET AL.— CYTOGENETICS OF BRACHISTOSTERNUS

333

Table 1. — Karyotype characteristics and collecting localities of the Bothriuridae species cytogenetically analyzed.

Species

2n

n

Locality

References

Bothriurus sp.

36

-

Tres Lagoas, Matto Grosso, Brazil

Piza 1947

Bothriunis araguaycie Vellard 1934

44

22

Sao Paulo, Brazil

Ferreira 1968 (sub. B.
asper araguaie)

Bothriurus Jlavidus Kraepelin 1911

48

24

Buenos Aires Province, Argentina

Giacomozzi 1977

Bothriurus prospicuus Mello-Leitao 1934

50

25

Buenos Aires Province, Argentina

Giacomozzi 1977

Brachistosternus alienus Lonnberg 1898

28

14

Chubut Province, Argentina

Giacomozzi 1977

Brachistosternus ferrugineus (Thorell 1876)

46

23

San Marcos Sierra, Cordoba Province, Argentina

This work

B. ferrugineus

46

23

Chepes, La Rioja Province, Argentina

This work

Brachistosternus montanus Roig-Alsina 1977

46

23

Laguna Brava, La Rioja Province, Argentina

This work

Brachistosternus pentheri Mello-Leitao 1931

46

23

Villa Union, La Rioja Province, Argentina

This work

B. pentheri

42

21

Chepes, La Rioja Province, Argentina

This work

B. pentheri

42

21

Oriente, Buenos Aires Province, Argentina

This work

B. pentheri

42

21

Buenos Aires Province, Argentina

Giacomozzi 1977(sub. B.
psainniophilus Maury)

Timogenes elegans (Mello-Leitao 1931)

48

24

Rio Negro Province, Argentina

Giacomozzi 1977

notic bodies are present (Fig. Id). Meiosis is achiasmatic; all
bivalents are homomorphic, condensing gradually until
metaphase I (Fig. le). Bibrachial and monobrachial chromo-
somes of different sizes were observed at prometaphase and
metaphase II (Fig. 10-

Morphs of B. pentheri showed different chromosome
numbers. The karyotype of the northern morph (Villa Union,
northern La Rioja) exhibits 46 chromosomes (cytotype I)
(Fig. 2), whereas karyotype of the typical morph from Chepes
(southern La Rioja) and Oriente (southeastern Buenos Aires)

d e f

Figure 2. a-f. — Brachistosternus pentheri (In = 46, n = 23). a. Spermatogonial prometaphase; b. Pachytene; c. Postpachytene; d.
Prometaphase I; e. Metaphase I; f. Metaphase II. Scale bar: 10 pm.

334

THE JOURNAL OF ARACHNOLOGY

Figure 3. a-d. — Brachistosternus pentheri (In = 42, /; = 21). a. Pachytene; b. Late postpachytene; c. Prometaphase I; d. Metaphase 1. Scale
bar: 10 pm.

consisted of 42 chromosomes (cytotype II) (Fig. 3). Chromo-
somes of both morphs presented the same meiotic behavior.
At spermatogonia! prometaphases and metaphases chromo-
somes of different sizes were observed (Fig. 2a). At prophase I
no positively heteropycnotic bodies were detected (Figs. 2b,
3a); meiosis is achiasmatic, and bivalents are homomorphic,
condensing gradually during prophase I (Figs. 2c-e, 3b-d).
Bibrachial (metacentric or submetacentric) and monobrachial
chromosomes could be identified at metaphase II of individ-
uals with 2n = 46 (Fig. 20. No metaphases II were observed in
the individuals with 2n = 42. Karyotype of B. pentheri
cytotype I (n = 23) was formed by three larger bivalents of
different size and the rest of the complement decreasing
gradually in size (from medium-sized to small bivalents)
(Fig. 4a, Table 2). Karyotype of B. pentheri cytotype II (n =
21) was formed by four larger bivalents of different size, 10
medium-sized bivalents that gradually decreased in size, and
seven smaller bivalents that gradually decreased in size
(Fig. 4b, Table 2).

DISCUSSION

This study focuses on cytogenetics of bothriurid scorpions of
the genus Brachistosternus, comparing representatives of the
subgenera Minis t emus (B. ferrugineus) and Brachistosternus (B.
montanus and B. pentheri). The karyotype of studied species is
formed by a mixture of bibrachial and monobrachial chromo-
somes of different sizes. Meiotic complements of the three

analyzed species of the family Bothriuridae contain no hetero-
morphic chromosome pair, which indicates the absence of
heteromorphic sex chromosomes in males, as is also the case in
Buthus occitanus (Amoreux 1789) and Pandinus imperator (C.L.
Koch 1841) (Guenin 1957, 1961). Shanahan (1989a, 1989b)
showed achiasmatic meiosis in males of Buthidae and Urodaci-
dae. Our study of Brachistosternus, as well as that of Ferreira
(1968) on Bothriurus araguayae Vellard 1934, reveals the
presence of this derived type of meiosis also in Bothriuridae.

Our study provides the first cytogenetic analysis of B.
ferrugineus and B. montanus. The karyotypes of these species
consist of 46 chromosomes, a number that has not been found
in the family Bothriuridae previously (Table 1). Although B.
ferrugineus is widely distributed, there are almost no
morphological differences between populations and no vari-
ation in chromosome number.

In contrast to B. ferrugineus, B. pentheri shows two different
morphs (Roig Alsina & Maury 1984, Ojanguren-Affilastro
2005a), and our analysis revealed karyotypic differences
between them. The northern morph of B. pentheri from Villa
Union (northern La Rioja) shows the same chromosome
number as B. ferrugineus. In contrast, the karyotype of B.
pentheri from Chepes (southern La Rioja) and Oriente
(southeastern Buenos Aires) consists in 42 chromosomes.
These specimens correspond morphologically to the species’
holotype from Mendoza province, Argentina, and they belong
to the typical morph of the species. Although the range of the

RODRIGUEZ-GIL ET AL.— CYTOGENETICS OF BRACHISTOSTERNUS

335

10%

9%

8%

^ 7%

<u 6%

■s 5%

CO

4%

^ 3%

2%

1%

0%

Bivalent number 3

10%

9%

8%

^ 7%

<u 6%

■e 5%

£D

4%

^ 3%

2%

1%

0%

Figure 4. a-b. — Idiograms of Brachistostermis pentheri cytotypes. a. Cytotype I, n = 23; b. Cytotype II, « = 21.

-

—

-

— ■

-

-

—

-

—

-

-

—

-

—

-

—

—

-

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21

Bivalent number

relative lengths of the bivalents in both cytotypes was similar,
three bivalent groups can be distinguished in cytotype II, but
only two in cytotype I (Fig. 4). Since only 1% of Brcichistos-
ternus species were cytogenetically analyzed it was not possible
to determine the ancestral karyotype and therefore the
rearrangements that lead to the different chromosome
numbers (e.g., centric or tandem fusions versus fissions). The
intraspecific morphological and karyotypic variations of B.
pentheri indicate that its marginal northern populations could
be a different subspecies or even species. This proposal could
be tested by attempting hybridization between the typical and
northern morphs.

The genus Brachistosternus was studied cytogenetically for
the first time by Giacomozzi (1977) (Tab. 1), who mentions
that his specimens were collected and determined by Dr. E.
Maury as B. psammophilus Maury 1977 and B. alienus
Lonnberg 1898. B. psammophilus was considered a possible
endemic species confined to coastal dunes in southern Buenos
Aires province (Maury 1977). However, some years later Roig
Alsina & Maury (1984) synonymized B. psammophilus with B.
pentheri, a widespread species from central and northern
Argentina. Therefore, the specimens studied by Giacomozzi
(1977) as B. psammophilus should be considered B. pentheri.
The karyotype of the specimens studied by Giacomozzi as B.
psammophilus consisted of 42 chromosomes, like our speci-

mens of B. pentheri from Oriente. Both samples of specimens
belong to the same group of populations from coastal dunes of
southern Buenos Aires.

On the other hand, the identity of the second species of
Brachistosternus studied by Giacomozzi (1977) is uncertain.
Maury determined these specimens as B. alienus (Giacomozzi
1977), but it is possible that they belong to B. august imanus
Ojanguren-Affilastro & Roig-Alsina 2001. At the time of
Giacomozzi’s investigation, most authors based the identifi-
cation of B. alienus on the redescription by Mello-Leitao
(1938, 1945), whose definition of B. alienus encompassed
another species now known as B. angustimanus (Ojanguren-
Affilastro 2001; Ojanguren-Affilastro & Roig-Alsina 2001).
Both species are sympatric over most of their ranges, but B.
angustimanus is more commonly found than B. alienus because
of its larger size and higher abundance.

Brachistosternus alienus (sensu Giacomozzi 1977) shows the
lowest chromosome number known for the genus (n = 14)
(Giacomozzi 1977). Recent phylogenetic analyses (Ojanguren-
Affilastro 2008; Ojanguren-Affilastro & Ramirez 2009) that
include both morphological and molecular data placed B.
pentheri as the sister group of the clade {B. angustimanus (B.
alienus (Brachistosternus telteca Ojanguren-Affilastro 2000,
Brachistosternus miiltidentatus Maury 1984))). Therefore, the
assessment of whether the low chromosome number is a

336

THE JOURNAL OF ARACHNOLOGY

Table 2. — Relative lengths (RL) of bivalents of Brachistostenms
pentlieri cytotypes.

Bivalent number

RL (%)

Cytotype I (/7=23)

Cytotype II («=21)

1

8.17

8.78

2

7.22

7.99

3

6.57

7.14

4

5.70

6.42

5

5.34

5.79

6

5.13

5.57

7

5.00

5.39

8

4.85

5.12

9

4.65

4.96

10

4.41

4.83

11

4.29

4.71

12

4.08

4.58

13

3.91

4.33

14

3.68

4.17

15

3.63

3.74

16

3.48

3.36

17

3.28

3.02

18

3.11

2.81

19

3.00

2.66

20

2.89

2.46

21

2.78

2.17

22

2.53

23

2.30

synapomorphy of the clade or an autapomorphy of the
specimens determined by Maury as B. alienus should be made
using specimens accurately identified as B. alienus and B.
august imamis.

ACKNOWLEDGMENTS

The present work was supported by grants from the
National University of Buenos Aires (UBA) (Ex 317, L.
Mola) and from the National Council of Scientific and
Technological Research (CONICET) (PIP 5927, L. Mola and
PIP 5654, C. Scioscia). Fieldwork was supported by CON-
ICET grant PIP 6502 (Mercedes Lizarralde de Grosso), and by
a MACN grant for collections improvement to C. Scioscia and
A. A. Ojanguren-Affilastro. The authors wish to thank L.
Compagnucci, J.J. Martinez, and L. Piacentini for their help in
the fieldwork, and H. Dinapoli for maintaining the scorpions
in captivity.

LITERATURE CITED

Cekalovic, K.T. 1969. Brachistosternus pegnai n. sp. deescorpibn para
el Ecuador (Scorpionida-Bothriuridae). Boletin de la Sociedad
Biolbgica de Concepcion 41:163-171.

Ferreira, A. 1968. Contribution to the knowledge of cytology of two
species of Brazilian scorpions: Opisthacaiulnis manaiiarensis
Ferreira, 1967 (Scorpiones, Scorpionidae) and Bothriwus asper
ciraguaie (Scorpiones, Bothriuridae). Anais da Academia Brasileira
de Ciencias 40( 1 ):97-99.

Giacomozzi, R.O. 1977. Citotaxonomia de Bothriuridae (Scorpio-
nida) de la Argentina. Physis (Buenos Aires), C 37:236.

Guenin, H.A. 1957. Contribution a la connaissance cytologique des
Scorpions: les chromosomes de Pandimts imperator Koch. Revue
Suisse de Zoologie 64:349-353.

Guenin, H.A. 1961. Contribution a la connaissance cytologique des

Scorpions: les chromosomes de Biithus occitanus Amor. (1). Vie et
Mileu 12:89-96.

Maury, E.A. 1974. Escorpiofauna Chaquena. 1. La verdadera
identidad de Brachistosternus ( Microsternus) ferrugineus (Thorell
1876) (Bothriuridae). Physis (Buenos Aires), C 33(86):73-84.

Maury, E.A. 1977. Un nuevo Brachistosternus de los medanos
costeros bonaerenses (Scorpiones: Bothriuridae). Physis (Buenos
Aires), C 37(93): 169-176.

Mello-Leitao, C.F. de. 1938. Notas sobre alacranes argentinos. Notas
del Museo de La Plata 3(9):83-95.

Mello-Leitao, C.F. de. 1945. Escorpioes sul-americanos. Arquivos do
Museu Nacional 40:7^68.

Ochoa, J.A. 2002. Nueva especie de Brachistosternus Pocock

(Scorpiones. Bothriuridae) del sur del Peru. Revista Peruana de
Biologia 9:55-63.

Ochoa, J.A. 2004a. Filogenia del genero Orohothriurus y descripcidn
de un nuevo genero de Bothriuridae (Scorpiones). Revista Iberica
de Aracnologia 9:73.

Ochoa, J.A. 2004b. Brachistosternus ninapo una nueva especie
(Scorpiones: Bothriuridae) de los Andes occidentales en el sur del
Peru. Revista Peruana de Biologia 11:139-148.

Ochoa, J.A. & L.E. Acosta. 2002. Two new Andean species of
Brachistosternus Pocock (Scorpiones. Bothriuridae). Euscorpius
2:1-13.

Ochoa, J.A. & A. A. Ojanguren-Affilastro. 2007. Systematics and
distribution of Brachistosternus (Brachistosternus) ehrenhergii
(Gervais, 1841) with the first record of stridulation in this genus
Brachistosternus (Scorpiones: Bothriuridae). Studies on Neotrop-
ical Fauna and Environment 41:79-85.

Ojanguren-Affilastro, A. A. 2001. Sistematica y distribucion de
Brachistosternus alienus Lonnberg (Scorpiones, Bothriuridae).
Revista del Museo Argentino de Ciencias Naturales 3:169-174.

Ojanguren-Affilastro, A. A. 2003a. Las especies andinas de Brachistos-
ternus ( Leptosternus), con la descripcidn de tres nuevas especies
(Scorpiones, Bothriuridae). Revista Iberica de Aracnologia 8:
23-36.

Ojanguren-Affilastro, A. A. 2003b. The genus Brachistosternus in
Argentina, with the description of a new species from Patagonia
(Scorpiones, Bothriuridae). Journal of Arachnology 31:317-331.

Ojanguren-Affilastro, A. A. 2005a. Estudio monografico de los
escorpiones de la Repiiblica Argentina. Revista Iberica de
Aracnologia 11:75-241.

Ojanguren-Affilastro, A. A. 2005b. Notes on the genus Brachistos-
ternus (Scorpiones, Bothriuridae) in Chile, with the description of
two new species. Journal of Arachnology 33:175-192.

Ojanguren-Affilastro, A. A. 2008. Sistematica y filogenia de los
escorpiones del genero Brachistosternus Pocock (Arachnida,
Scorpiones, Bothriuridae). Thesis. Facultad de Ciencias Exactas
y Naturales, Universidad de Buenos Aires.

Ojanguren-Affilastro, A. A. & M.J. Ramirez. 2009. Phylogenetic
analysis of the scorpion genus Brachistosternus (Arachnida,
Scorpiones, Bothriuridae). Zoologica Scripta 38:183-198.

Ojanguren-Affilastro, A. A. & A. Roig Alsina. 2001. Brachistosternus
august imanus, una nueva especie del norte de la Patagonia,
Argentina (Scorpiones, Bothriuridae). Physis (Buenos Aires), C
58:15-22.

Ojanguren-Affilastro, A. A. & C.L. Scioscia. 2007. New data about
genus Brachistosternus in Chile (Scorpiones: Bothriuridae). Journal
of Arachnology 35:102-113.

Ojanguren-Affilastro, A.A., C.L Mattoni & L. Prendini. 2007a. The
genus Brachistosternus in Chile (Scorpiones. Bothriuridae), with
the description of two new species. American Museum Novitates
3564:1^4.

Ojanguren-Affilastro, A. A., P. Agusto, J. Pizarro Araya & C.L
Mattoni. 2007b. Two new scorpion species of genus Brachistos-

RODRIGUEZ-GIL ET AL.— CYTOGENETICS OF BRACHISTOSTERNVS

337

tenms (Scorpiones: Bothriuridae) from northern Chile. Zootaxa
1623:55-68.

Piza de Toledo, S. 1947. Notas sobre crommossomios de alguns
escorpioes brasileiros. Anais da Escola Superior de Agricultura
“Luiz de Queiroz” 4:169-176.

Prendini, L. 2003. A new genus and species of bothriurid scorpion
from the Brandberg Massif, Namibia, with a reanalysis of
bothriurid phylogeny and a discussion of the phylogenetic position
of Lisposoma Lawrence. Systematic Entomology 28:149-172.

Reeves, A. & J. Tear. 2000. MicroMeasure for Windows, Version 3.3.
Free program available online at http://www.colostate.edu/Depts/
Biology/MicroMeasure.

Roig-Alsina, A.H. 1977. Una nueva especie de escorpion andino en
Mendoza, Repiiblica Argentina. Physis (Buenos Aires), C
37:255-259.

Roig-Alsina, A.H. & E.A. Maury. 1984. Sistematica y distribucion
geografica de Brachistostermis (L.) pentheri Mello-Leitao, (1931).
(Scorpiones, Bothriuridae). Physis (Buenos Aires), C 42:17-21.
Shanahan, C. 1989a. Cytogenetics of Australian scorpions. 1.
Interchange polymorphism in the family Buthidae. Genome
32:882-889.

Shanahan, C. 1989b. Cytogenetics of Australian scorpions. 11.
Chromosome polymorphism in species of Urodactus (Family
Scorpionidae). Genome 32:890-900.

Manuscript received 21 November 2008, revised 21 July 2009.

2009. The Journal of Arachnology 37:338-345

Redescription of Plesiochactas mitchelli (Scorpiones: Euscorpiidae): a rare scorpion from Central America

Kaleb Zarate-Galvez: Coleccion de Aracnidos, Facultad de Ciencias Biologicas, Universidad de Ciencias y Artes de
Chiapas. Libramiento Norte Poniente s/n, Colonia Lajas Maciel, C.P. 29039. Tuxtla Gutierrez, Chiapas, Mexico.
E-mail; kalebzg7@gmail.com

Oscar F. Francke: Coleccion Nacional de Aracnidos, Departamento de Zoologia, Instituto de Biologia, Universidad
Nacional Autonoma de Mexico, Apartado Postal 70-153, Ciudad Universitaria, Mexico, D.F. 04510, Mexico

Abstract. Plesiochactas mitchelli Soleglad 1976 was originally described from a juvenile female collected in “Guatemala”
before 1902. The species is redescribed on the basis of an adult female from a specific locality in the state of Chiapas; it is
the first record of this species from Mexico.

Keywords: Euscorpiidae, Chiapas, Mexico, systematics

There is little information available about scorpions of
the genus Plesiochactas Pocock 1900. This genus consists of
only two species: Plesiochactas dilutus Karsch 1881, the
type species which is known from four specimens (Soleglad

1976; Soleglad & Sissom 2001), and Plesiochactas mitchelli
Soleglad 1976. The male is undescribed for this genus, and
P. mitchelli is known only from a single immature
specimen.

Figures 1, 2. — Plesiochactas mitchelli female adult specimen. 1. Dorsal aspect; 2. Ventral aspect. Scale line; 5 mm.

338

zArATE-GALVEZ & FRANCKE— SCORPION PLESIOCHACTAS MITCHELLI

339

Figures 3, 4. — Plesiochactas mitchelli female adult specimen. 3. Carapace; 4. Ventral aspect of sternum and pectines.

Pocock (1902) mentioned a young specimen from “Guate-
mala” nearly allied to P. dilutus. Soleglad (1976) revised the
scorpion subfamily Megacorminae, with limited material
available. He described P. mitchelli without a precise locality
from the same juvenile female reported by Pocock. To this
date, no additional specimens of this species have been
reported in the literature. In the present paper, we redescribe
P. mitchelli on the basis of an adult female from southeastern
Chiapas, Mexico, from a locality not too distant from the
border with Guatemala (see Remarks).

METHODS

The new specimen examined for this study is lodged in the
Coleccion Nacional de Aracnidos (CNAN), Institute de
Biologia, Universidad Nacional Autonoma de Mexico,
Mexico, D.F. Nomenclature and mensuration primarily
follow Stahnke (1970), with the following exceptions: meta-
somal carinal terminology after Francke (1977), pedipalp
carinae terminology after Acosta et al. (2008), trichobothrial
designations after Vachon (1974), and chelal finger dentition
terminology after Soleglad & Sissom (2001). Measurements
were made with an ocular micrometer at 1 OX on a Nikon SMZ
800 stereomicroscope; drawings on the same scope with a
camera lucida attached; photographs also on the saro.e scope
with a Nikon Coolpix 20 camera.

SYSTEMATICS

Family Euscorpiidae Laurie 1896
Subfamily Megacorminae Kraepelin 1899
Genus Plesiochactas Pocock 1900
Plesiochactas Pocock 1900:470.

Type species. — Plesiochactas dilutus (Karsch 1881), by
monotypy.

Plesiochactas mitchelli Soleglad 1976
Plesiochactas mitchelli Soleglad 1976:251, 263, 286, 288, 294—

298, Table 4, Figs. 17, 21, 86, 87, 117, 118, 121, 123-134;

Soleglad & Sissom 2001:29, 58, 67, 92, Figs. 141, 196.
Plesiochactas dilutus (in part) Pocock 1902:16, 17, Table 4,

Figs. 5a-f.

Type data. — GUATEMALA: Holotype juvenile female
from “Guatemala (Sarg),” date unknown (deposited in
Natural History Museum, London).

Additional specimen. — MEXICO: Chiapas: 1 adult ?, Santa
Rosa, [Municipio] La Trinitaria, August 1974, [col.] A.
Ramirez V. (CNAN).

Diagnosis. — Adult female 53.3 mm long. Pectines quite
reduced; 5 teeth for female; fulcra absent. Median carina of
sternite VII obsolete. Stigmata elongate oval. Patella with 23
trichobothria on external face: 5 each in em and et series; 8-10
ventral trichobothria in a straight row along posterior margin.
P. dilutus, the only other species in the genus has 8-9 pectinal
teeth in females, 10 in the male; the pectines have well
developed fulcra; and median carina of sternite VII weak to
moderately strong, smooth.

Description of adult female (Figs. 1, 2). — Coloration: uni-
formly orange-brown, darkening to reddish brown in cara-
pace, pedipalps, metasomal segments and telson. Prosoma:
carapace 1.2 times wider than long, densely covered with
random mixture of large, medium, and small granules,
anterior margin straight. Lateral eyes, two per side, well
developed, posterior eye slightly smaller. Median ocular
tubercle very well developed and prominent, one-fifth width
of carapace at that point; placed on anterior two-fifths of
carapace (Fig. 3). Sternum wider than long with some
granules distally, deep median groove (Fig. 4). Mesosoma:
tergites rough, minutely and densely granulose, with larger.

340

THE JOURNAL OF ARACHNOLOGY

Figures 5, 6. — Stigmata of Plesiochactas mitchelli female adult specimen. 5. Stigmata closed; 6. Stigmata open.

sparse granules on posterolateral regions. Median carina weak
to vestigial in all tergites. Tergite VII with two pairs of
crenato-granulose carinae. Genital operculum: relatively small,
subtriangular, with complete median longitudinal membra-
nous connection. Pectines (Fig. 4): reduced, three marginal

and one middle lamellae well defined; fulcra absent; pectinal
teeth 4 times longer than wide, sensorial areas present distally,
on '/3 to V2 of tooth; tooth count 5-5. Sternites: III-VI
lustrous; VII with a pair of vestigial lateral carinae, hinted by
some granules; median carina obsolete. Stigmata: reniform

ZARATE-GAlVEZ & FRANCKE— SCORPION PLESIOCHACTAS MITCHELL! 341

Figures 7, 8. — Plesiochactas mitcheUi female adult specimen. 7. Lateral aspect of metasomal segments and telson. Scale line: 5 mm; 8. Ventral
aspect of left chelicerae. Scale line: 1 mm.

elongate, with a distinctly sclerotized lid hinged distally, which
opens inward to the book lung cavity (Figs. 5, 6). Metasoma:
intercarinal spaces with sparse granules. Segments I-II wider
than long. III as long as wide and IV longer than wide.
Dorsolateral and lateral supramedian carinae crenulated to
serrated. Posterior spines on dorsal carinae weakly developed,
gradually stronger on distal segments (Fig. 7). Lateral infra-
median carinae on I, granulose and complete; on II with few
granules posteriorly, and on III-IV obsolete. Ventrolateral
carinae on I-II, granulose; on III-IV crenulated to serrated.
Ventral median carina on I weak, granulose; on II-IV
crenulated to serrated. Segment V: dorsolateral and ventro-
lateral carinae crenulated to serrated; lateral median carina
weak, indicate only on proximal region; ventral median carina,
serrated; anal arc with subtle granulation. Telson: segment
elongated, aculeus-vesicle juncture not sharply defined. Vesicle
dorsally smooth, lateral and ventrally covered with low,
rounded granules. Aculeus slightly curved. Subaculear tuber-
cle/spine absent. Chelicera: movable finger with two subdistal
denticles on dorsal edge; inferior margin with 2-A small
denticles; without serrula (Fig. 8). Pedipalps: dorsal face of
patella and femur with scattered granules; ventral face smooth
to rough. Femur orthobothriotaxic (Fig. 9): 2.6 times longer
than wide; retrodorsal, prodorsal and proventral carinae,
strong and crenulated; retroventral carina present only on

proximal region, strong and granulose; retrolateral face with
granulose to serrated median carina; prolateral face with a
median carina indicate by some longer spiniform granules.
Patella neobothriotaxic (Figs. 10, 1 1): retrolateral face with 23
trichobothria (5 et, 4 est, 5 em, 2 esb and 7 eb); ventrally on
right side with 9 v trichobothria, left with 10 v; 2.3 times
longer than wide; all carinae crenulated to serrated; retro-
lateral aspect with well developed median carina, crenulated;
proximal projection of prolateral face with one medium and
one small spine. Chela: Orthobothriotaxic (Figs. 12, 13):
movable finger shorter than carapace and slightly shorter
than fifth metasomal segment. Digital and retrolateral
secondary carinae strong, crenulated; dorsal secondary and
prodorsal carinae weak, granulose; retroventral carina very
strong and crenulated; proventral carina weak, crenulated;
ventral median carina vestigial, only present and granulose
proximally; prolateral median carina crenulated. Intercarinal
spaces with sparse granules. Chelal finger dentition, based on
right movable finger (Fig. 14): median denticle row flanked by
groups of one outer denticle plus two inner denticles; in
addition, the median denticle corresponding to each group is
slightly enlarged, resulting in transverse rows of four denticles
each, which divide the median denticle row into distinct sub-
rows. Distally there are no accessory denticles; basal to distal
group 3 inner accessory denticles appear, and basal to distal

342

THE JOURNAL OF ARACHNOLOGY

Figures 9-11. — Plesiochactas mitchelli female adult specimen. 9. Dorsal aspect of pedipalp femur; 10. Dorsal aspect of pedipalp patella; 11.
Retrolateral aspect of pedipalp patella. Scale line: 5 mm.

group 5 outer accessory denticles are present. Basal to distal
group 6 a poorly defined “double row” formed by the median
denticle row and the inner accessory denticles can be
discerned. Measurements: (in mm; L = length, W = width,
D = depth). Total L 53.3; carapace L/W 8. 0/9. 3; mesosoma L
17.2; metasoma L 19.4: segment I L/W/D 2.3/3.8/3.1, segment
II L/W/D 2. 6/3. 3/3.0, segment III L/W/D 3.0/3.2/3.1, segment
IV L/W/D 4.0/3.0/3.1, segment V L/W/D 7.5/2.7/2.8. Telson L
8.7: vesicle L/W/D 5. 1/3. 0/2. 6; pedipalp L 26.7: femur L/W 6.6/
2.5, patella L/W 6.8/3.0, chela L/W/D 13.3/4.0/4.3, movable
finger L 7.0, fixed finger L 5.9.

Remarks. — Francis Charles Sarg (whom we presume
collected the holotype) was born in Guatemala; in the 1880s
he lived in Coban and around 1902 he owned a farm called “El
Chicabal” in Quetzaltenango. It is possible the holotype was
collected in or near that farm.

There are four settlements called Santa Rosa in the
Municipio of La Trinitaria, Chiapas, and we do not know
from which of them the specimen was collected; thus we can
not provide geographical coordinates or elevation for the
specific location. Quetzaltenango, Guatemala, is approximate-
ly 100 km from the Municipio of La Trinitaria.

DISCUSSION

The principal difference, other than size, between the juvenile
holotype and the adult female of P. mitchelli is the patella

ventral trichobothrial number: eight in the holotype, nine and
ten in the adult female specimen. This character shows similar
variability in the other described species of Plesiochactas (P.
di I lit us: 9-12), and is also variable in species of the genus
Megacormus Karsch 1881 (Soleglad 1976; Sissom 1994;
Francke pers. obs.). The submedian and lateral carinae of
sternite VII are obsolete in the holotype, and on the adult
female a pair of vestigial lateral carinae are present, and the
submedian carinae are also obsolete. One somewhat large
denticle on the distal aspect of the ventral edge of the movable
cheliceral finger of the holotype, the adult female has two to
four spaced denticles. The movable finger of the pedipalp chela
on the adult female is shorter than the carapace and slightly
shorter than the fifth metasomal segment; in the holotype it is
slightly shorter than the carapace, and longer than segment V of
the metasoma. Several differences of mesosomal, metasomal,
and pedipalpal carinae between the holotype and the adult
female are ascribable to ontogenetic changes. Both the presence
or absence of fulcra on the pectines, and the pectinal tooth
counts on females, are diagnostic characters that separate the
two species of Plesiochactas. Also, the median carina on sternite
VII is weak to moderate and smooth on P. diliitus, and it is
smooth and vestigial to obsolete on P. mitchelli.

The taxonomic status of this genus and its species remains
uncertain. Originally, Plesiochactas was separated from
Megacormus by the presence of distinct fulcra in the pectines.

zARATE-GAlVEZ & FRANCKE— SCORPION PLESIOCHACTAS MITCHELL!

343

Figures 12, 13. — Plesiochactas mitchelli female adult specimen. 12. Retrolateral aspect of right pedipalp chela; 13. Dorsal aspect of right
pedipalp chela. Scale line: 5 mm.

but P. mitchelli lacks the pectinal fulcra. Plesiochactas dilutus
and P. mitchelli share having five trichobothria on the external
terminal series of the patella, as opposed to only 3^ on
Megacormus, and on that basis P. mitchelli was placed in
Plesiochactas by Soleglad (1976). However, the number of
trichobothria on that region of the patella is quite variable in
the family Euscorpiidae Laurie 1896 (Soleglad & Sissom
2001), and thus the similarity between the two species
currently placed in Plesiochactas could be a symplesiomorphy.

Soleglad (1976) and Soleglad & Sissom (2001) also indicate
that Plesiochactas and Megacormus can be separated on the
basis of chelal finger dentition; the former with 45+ small,
inner accessory denticles, arranged row-like; the later with 35+
small outer accessory granules, scattered, irregular row-like. A

direct comparison of the denticle pattern in P. mitchelli
(Fig. 14), against those of Megacormus gerstchi Diaz 1966
(Fig. 15), and P. dilutus (Fig. 16) clearly indicate a stronger
similarity between the first two, which share transverse
“tetrads” along the movable finger length. In P. dilutus,
however, the tetrads tend to disappear after distal group 3 and
only enlarged inner and outer denticles are present, and a
“double row” of median plus inner accessory denticles is quite
distinct on the basal 2/3 of the movable finger. The second
author has undertaken a revision of the tribe Megacormini
Kraepelin 1899 [Megacormus + Plesiochactas], including a
cladistic analysis of the phylogenetic relationships of all
included taxa, which should help resolve the current uncer-
tainty (ms. in preparation).

344

THE JOURNAL OF ARACHNOLOGY

Figures 14—16. — Movable finger dentition of right pedipalp chela. 14. Plesiochactas mitchelli adult female; 15. Megciconnus gertschi adult
female from Hidalgo, Tiangiiistengo; 16. Plesiochactas juvenile female from Oaxaca, Sta. Maria Tlahuitoltepec.

ACKNOWLEDGMENTS

Our thanks to Dr. Carlos R. Beutelspacher Baigts for
donating the specimen to the CNAN, Belen Chavez Galvan
kindly researched the information on Mr. Francis C. Sarg,
Alejandro Valdez Mondragon helped with the photographs,
and Carlos Santibanez Lopez kindly reviewed the manuscript.
Financial support was provided by FOMIX-Chiapas through
the project with key: CHIS-2006-C06-45752.

LITERATURE CITED

Acosta, L.E., D.M. Candido, E.H. Backup & A.D. Brescovit. 2008.
Description of Zahiiis gaiicho (Scorpiones, Buthidae), a new species
from southern Brazil, with an update about the generic diagnosis.
Journal of Arachnology 36:491-501.

Francke, O.F. 1977. Scorpions of the genus Diplocentnis from
Oaxaca, Mexico (Scorpionida, Diplocentridae). Journal of Arach-
nology 2:107-1 18.

Pocock, R.I. 1902. Arachnida: Scorpiones, Pedipalpi and Solifugae.
Ill Biologia Centrali Americana (F.D. Goodman & O. Salvin, eds.).
Taylor and Francis, London.

Sissom, W.D. 1994. Systematic studies on the genus Megacormus
(Scorpiones, Chactidae, Megacorminae), with descriptions of a
new species from Oaxaca, Mexico and the male of Megacormus
segmentatus Pocock. Insecta Mundi 8:265-271.

Soleglad, M.E. 1976. A revision of the scorpion subfamily Mega-
corminae (Scorpionida: Chactidae). Wasmann Journal of Biology
34:251-303.

Soleglad, M.E. & W.D. Sissom. 2001. Phylogeny of the family
Euscorpiidae Laurie, 1896: a major revision. Pp. 25-111. In

ZARATE-GALVEZ & FRANCKE— SCORPION PLESIOCHACTAS MITCHELL!

345

Scorpions 2001. In Memoriam Gary A. Polls. (V. Fet & P.A.
Selden, eds.). British Arachnological Society, Burnham Beeches,
Buckinghamshire, UK.

Stahnke, H.L. 1970. Scorpion nomenclature and mensuration.
Entomological News 81:297-316.

Vachon, M. 1974. Etude des caracteres utilises pour classer les
families et les genres de Scorpions (Arachnides). 1. La trichobo-

thriotaxie en Arachnologie. Sigles trichobothriaux et types de
trichobothriotaxie chez les Scorpions. Bulletin du Museum
National d’Histoire Naturelle, Paris, 3e serie 140 (Zoologie
104);857-958.

Manuscript received 12 January 2009, revised 22 July 2009.

2009. The Journal of Arachnology 37:346-356

New data on the genus Urophonius in Patagonia with a description of a new species of the exochus group

(Scorpiones: Bothriuridae)

Andres Alejandro Ojanguren-Affilastro: Museo Argentine de Ciencias Naturales “Bernardino Rivadavia”, Avenida
Angel Gallardo 470, CP: 1405DJR, CAB A, Buenos Aires, Argentina. E-mail: ojanguren@macn.gov.ar;
andres.ojanguren@gmail.com

German Cheli: Centro Nacional Patagonico (Cenpat), Boulevard Brown 2825, Puerto Madryn - CP: U 9120 ACF,
Chubut, Argentina

Abstract. New data on the distribution and systematics of Patagonian speeies of the scorpion genus Urophonius Pocock
1893 are provided. A speeies of this genus from Peninsula Valdes in central eastern Argentinean Patagonia, Urophonius
nuirtinezi new species, is described. The surface activity period of most of the species of the genus is reviewed and clearly
established. A distribution map as well as a key for the Patagonian species of the genus are provided.

Keywords: Scorpiones, systematics. Neotropics, Patagonia, summer vs. winter activity

Species of the scorpion genus Urophonius Pocock 1893
occur in southern South America, from southern Brazil to
southern Patagonia. Most species inhabit grasslands and
shrub steppes, but some species have been collected in forests
and even in low mountain ranges (Ojanguren-Affilastro 2005).
According to Prendini (2000, 2003), the genus shows an
intermediate position in the phylogeny of the family Bothriur-
idae, being the sister group of the genus Cercophonius Peters
1861 from Australia. One of the most remarkable character-
istics of Urophonius is that most species of this genus have
their surface activity period in the winter, opposite to most of
the Bothriuridae, which have their surface activity period in
the summer (Maury 1968, 1969, 1977, 1979; Ojanguren-
Affilastro 2002, 2005). Only some species of the genus
Vachonia Abalos 1953 apparently share this winter activity
period (Lopez & Magnanelli 2002; Ojanguren-Affilastro
2005); however, the information available on this genus is
still scarce.

In a revision of Urophonius, Acosta (1988) has separated the
species of this genus into three different groups of species: U.
hrachycentrus, U. granulatus, and U. exochus groups. The first
two groups were formerly included in a single group (group B,
Maury 1973), whereas exochus group corresponds to group A
of Maury (1973), which also corresponds to the original
definition of genus lophorus Penther 1913.

Most of the information we have on this genus concerns
species of the hrachycentrus group (Abalos & Hominal 1974;
San Martin & Gambardella 1974; Maury 1977; Acosta 1999),
and information on members of the exochus and granulatus
groups is still scarce (San Martin 1965; San Martin &
Cekalovic-Kuschevich 1968; Maury 1979; Cekalovic-Kusche-
vich 1981; Acosta 2003). The species of the exochus group
have only been collected in Argentina from southern
Patagonia to the central part of the country. Only three
species have been described in this group, Urophonius exochus
(Penther 1913), Urophonius mahuidensis Maury 1973, and
Urophonius eugenicus (Mello-Leitao 1931); all of these species
have been redescribed with modern standards in a recent
monograph on the Argentinean scorpion fauna (Ojanguren-
Affilastro 2005). Besides these species, several specimens of

this group have already been cited from a wide area of central
and southern Patagonia; however, in most cases they were
juveniles or females without clear diagnostic characters for
assigning them to a species; so our previous information about
this genus in Patagonia is scarce and fragmentary. Recently we
have collected additional material of this genus that allowed us
to clarify, at least partially, some aspects of the systematics of
this group. In this contribution we describe a new species of
the exochus group; we provide information about the surface
activity period of most species of this genus, as well as a key
and a distribution map of the Patagonian species of
Urophonius.

METHODS

Descriptive terminology follows Mattoni & Acosta (2005)

for the hemispermatophores, Vachon (1974) for the trichobo-
thria, and Francke (1977) for the metasomal carinae,
abbreviated as follows: DL: dorsolateral; LIM: lateral
inframedian; LSM: lateral supramedian; VSM: ventral sub-
median; VL: ventrolateral; VM: ventromedian. We followed
Francke (1977) for the pedipalp carinae, abbreviated as
follows: DI: dorsal internal; DE: dorsal external; VI: ventral
internal; VE: ventral external; D: digital; E: external; V:
ventral; VM: ventral median; DM: dorsal marginal; DS:
dorsal secondary. Abbreviations of collections are as follows:
AMNH: American Museum of Natural History (New York,
USA); CDA: Catedra de Diversidad Animal 1 (Universidad de
Cordoba, Cordoba, Argentina); MACN-Ar: Museo Argen-
tino de Ciencias Naturales ’Bernardino Rivadavia”, (Buenos
Aires, Argentina); MHNC: Museo de Historia Natural,
Facultad de Ciencias Biologicas, Universidad San Antonio
Abad del Cusco (Cusco, Peru); CENPAT: Centro Nacional
Patagonico (Puerto Madryn, Argentina); MZUC: Museo de
Zoologia de la Universidad de Concepcion, (Concepcion,
Chile). Other abbreviations: NPA-PV: Natural Protected Area
Peninsula Valdes. Illustrations were produced using a Leitz
Wetzlar stereomicroscope and camera lucida. Photographs
where taken using a Digital Camera (Nikon DXM 1200)
attached to a stereomicroscope (Nikon SMZ 1500), the focal
planes composed with Helicon Focus 3.10.3 (Online at http://

346

OJANGUREN-AFFILASTRO & CHELI— NEW DATA ON UROPHONIUS SCORPIONS

347

helicon.com.ua/heliconfocus/). Measurements, taken using an
ocular micrometer, were recorded in mm. Scorpions where
collected manually by ultraviolet collection at night, and by
pitfall traps in Natural Protected Area Peninsula Valdes
(NPA-PV). Traps were placed in shrubby steppes with 40-60%
vegetation cover, where the shrub Chuqidraga avellanedae
Lorentz and the grass Stipa tenuis Philippi were the most
representative species. The traps used were open plastic
containers, 11cm in diameter and 12 cm depth, with
300 cm'* of 30% propylene glycol; the traps were neatly buried
in the soil near Ch. avellanedae bushes. Trap contents were
collected after 15 days, fixed in 70% ethyl alcohol, and taken
to the laboratory for specimen identification.

RESULTS

Family Bothriuridae Simon 1880

Genus Urophonius Pocock 1893
Urophonius Pocock 1893:100-101.

Type species, — Urophonius iheringi Pocock 1893, by original
designation.

Urophonius martinezi new species
Figs. 1, 2, 6-11, 13, 18, 20; Table 1
Urophonius sp. grupo exochus: Ojanguren-Affilastro 2002:185,
186 (“ejemplares del grupo exochus provenientes de
Peninsula Valdes”); Ojanguren-Affilastro 2005:145, 146
(“escorpiones provenientes de Peninsula Valdes [...] muy
relacionados con U. eugenicus”).

Type series. — ARGENTINA: Chuhut: Holotype male
(MACN-Ar 15808) 20 km N from Puerto Madryn

(42°32'49.4"S, 64°48'25"W), 6-10 June 2008, Ojanguren-
Affilastro, Martinez & Cheli.

Paraty’pes'. ARGENTINA: Chuhut Province: 2 d. Lab. Vida
Silvestre, San Jose Gulf (42°27'5L01"S, 64°29'57.15"W), 28
October 1970, J. Dacinde (MACN-Ar 15805); 1 J, 2 9, 4
juveniles. La Falsa, Peninsula Valdes (42°13'26.6"S,
63°51'45.3"W), May-July 2005, G. Cheli (MACN-Ar 15806);
12 (J, 7 9, 12 juveniles, area around Puerto Madryn, (specimens
were collected at different points along the coastal road close
to Puerto Madryn: 42°49T0.8"S, 64°54'00. 1"W; 42°48'02.7"S,
64°57'17.3"W; 42°39'52.7"S, 64°59'37.8''W; 42°36'57.8"S,
64°51'47.4"W; 42°32'49.4"S, 64°48'25"W), 6-10 June 2008,
Ojanguren-Affilastro, Martinez & Cheli (MACN-Ar); 1 cJ, 1 9,
1 juvenile, same data, (CDA); 1 cJ, 1 9, 1 juvenile, same data,
(MHNC); 1 <3, 1 9, 1 juvenile, same data, (AMNH).

Etymology. — This species is named after the entomologist
Juan Jose Martinez (MACN, CONICET), who has collected
most of the type material of this species.

Diagnosis. — Urophonius martinezi is most similar to U.
eugenicus from southern Patagonia. Both species can be
separated by the shape of the telson, which is more globose in
U. eugenicus, with its vesicle more expanded towards the sting
and the posterior part of the telson (Figs. 11-14). This
difference is more conspicuous in males than in females,
because the telson of U. martinezi males is dorsoventrally
compressed (Fig. 11), such that males of both species can be
separated by means of the length/height ratio of the telson: U.
martinezi 3.10-3.52, n = 20, mean = 3.28; U. eugenicus 2.54—
3.03, n = 18, mean = 2.85; (in females the difference in shape

is not reflected by the difference in this ratio). We have not
found any male specimen in which this ratio overlaps;
however, the extremes of variation are very close, so this
possibility cannot be discounted. There are other differences
between these species. Urophonius eugenicus is more densely
pigmented than U. martinezi: the carapace of U. eugenicus has
a broad irregular stripe between the lateral eyes and the ocular
tubercle (Fig. 17), whereas in U. martinezi this stripe does not
exist, or it is reduced to some isolated spots (Fig. 18). The
ventral surface of tergite V and metasoma of U. eugenicus is
more densely granular than in U. martinezi. In tergite V of U.
eugenicus females there are two well developed VL carinae,
and between them there are abundant coarse granules, with
the VSM carinae almost indistinguishable; whereas in U.
martinezi between the VL carinae there are some tiny scattered
granules and two poorly developed VSM carinae (Fig. 6).

Description. — Color: General color yellowish, with dark
brown spots of pigment. Carapace: anterior margin with a
little dark spot in its median area, around the median notch;
ocular tubercle and area around the lateral ocelli dark brown;
and in the median part of the carapace there are two irregular
dark stripes that surround the ocular tubercle, the postocular
furrow, and the anterior longitudinal sulcus. With two lateral
irregular dark spots that connect with the lateral eyes, and
with two little posterolateral dark spots (Fig. 18). Chelicerae:
fixed finger with reticular pigment, especially near the
articulation with the movable finger; movable finger densely
pigmented in the external margin. Tergites: with two lateral
and two paramedian dark stripes, the lateral stripes are very
narrow and occupy almost the entire length of the segment,
from the posterior edge, almost reaching the anterior margin;
the paramedian stripes are triangular shaped, and extend from
the posterior margin to the median part of the segment.
Sternites, sternum, genital opercula and pectines unpigmented.
Metasoma: segment I: ventral surface with two VL stripes
extending the entire length of the segment and two VSM
stripes poorly developed and barely visible; lateral surface
slightly pigmented over the LSM carina; the rest of the
segment unpigmented. Segments II-IV: ventral surface with
two VL and two VSM stripes poorly marked but extending the
entire length of the segment; lateral surface with a dark stripe
over the LSM carinae, which is thicker near the posterior
margin and fuses with the VL stripes; dorsal surface with a
little median triangular spot. Segment V: ventral surface with
two VL and two VSM stripes extending the entire length of the
segment. The VSM stripes are very diffuse and very ramified,
connecting between them and with the VL stripes (Fig. 20);
lateral surface with a poorly marked and very ramified stripe
over the line of LSM setae; dorsal surface slightly pigmented
in the median part and on the lateral margins. Telson: vesicle
with dorsal surface unpigmented in its median area. In males,
the telson gland is light yellow; slightly pigmented near the
lateral margins; ventral and lateral surfaces densely pigment-
ed; aculeus dark brown. Legs: femur and patella densely
pigmented on the internal surface and near the articulation,
the rest unpigmented. Pedipalps: femur and patella slightly
pigmented near the articulations and on the dorsal and
external surfaces, the rest unpigmented; chela, with six
longitudinal stripes over the DI, DM, DS, D, E, V and VM
carinae.

348

THE JOURNAL OF ARACHNOLOGY

Figures 1-9. — 1, 2, 6-9. Urophoniiis martinezi new species; 1. Left hemispermatophore, internal aspect; 2. Left hemispermatophore, external
aspect; 6. Sternite V and metasomal segments I and 11, female, ventral aspect; 7. Left pedipalp chela, female, external aspect; 8. Right pedipalp
chela, male, internal aspect; 9. Right pedipalp chela, male, ventral aspect. 3. Urophoniiis eiigenicus, left hemispermatophore, externa! aspect. 4.
Urophoniiis exochus, left hemispermatophore, external aspect. 5. Urophoniiis mahiiidensis, left hemispermatophore, external aspect. Scale bars:

1 mm.

OJANGUREN-AFFILASTRO & CHELI— NEW DATA ON UROPHONIUS SCORPIONS

349

Figures 10-16. — 10, 11, 13. Urophonius martinezi new species: 10. Metasomal segment V, male, ventral aspect; 1 1. Telson, male, lateral aspect;
13. Telson female, lateral aspect. 12, 14. Urophonius eugenicus: 12. Telson male, lateral aspect; 14. Telson female, lateral aspect. 15. Retrolateral
aspect of a stereotyped left pedipalp femur of a Urophonius belonging to the granulatus group, showing the relative position of e trichobothria
respect to Ml macroseta in the different species of the group. 16. Retrolateral aspect of a stereotyped left pedipalp femur of a Urophonius
belonging to the hrachycentrus group. Scale bars: 1 mm.

350

THE JOURNAL OF ARACHNOLOGY

Figures 17 -20. — 17. Urophonius eitgeniais, carapace, dorsal aspect. 18. Urophonius martinezi, carapace, dorsal aspect. 19. Urophonius
gniniilatiis, metasomal segments IV and V, ventral aspect. 20. Urophonius martinezi, metasomal segments IV and V, ventral aspect. Scale bars:

1 min.

OJANGUREN-AFFILASTRO & CHELI— NEW DATA ON UROPHONIUS SCORPIONS

351

Table 1. — Measurements of a male and a female paratype of Uroplionius martinezi (MACN-Ar 15807).

Measurements (mm)

Uroplionius martinezi

Male paratype

new species

Female paratype

Total length

32.24

33.23

Carapace, length

4.04

4.05

Carapace, anterior width

2.83

2.91

Carapace, posterior width

4.85

4.28

Mesosoma, total length

7.27

11.48

Metasoma, total length

15.27

12.85

Metasomal segment I, length/width/height

2.02/2.59/1.94

1.70/2.67/2.02

Metasomal segment II, length/width/height

2.42/2.51/1.94

2.10/2.59/2.01

Metasomal segment III, length/width/height

2.59/2.51/1.94

2.42/2.51/1.95

Metasomal segment IV, length/width/height

3.39/2.34/1.86

2.83/2.42/1.78

Metasomal segment V, length/width/height

4.85/2.18/1.65

3.80/2.42/1.70

Telson, length

5.66

4.85

Vesicle, length/width/height

4.44/2.34/1.78

3.88/2.34/1.78

Aculeus, length

1.21

0.97

Femur, length/width

3.88/1.21

3.56/1.21

Patella, length/width

3.72/1.37

3.64/1.41

Chela, length/width/height

7.03/2.26/2.66

6.30/1.86/2.18

Movable finger, length

3.55

3.35

Morphology: Measurements of a paratype male and a
paratype female (MACN-Ar 15807) are recorded in Table 1.
Total length in males 28.50-36 mm (n = 14, mean = 32.25),
29-39 in females (n = 12; mean = 34.60). Carapace: tegument
slightly granular on the lateral margins, the rest smooth;
anterior margin with a well developed median notch; anterior
and posterior longitudinal sulci, lateral sulcus and postocular
furrow well developed; ocular tubercle well developed, median
eyes well developed, one diameter apart; with three lateral eyes
on each side of the carapace placed in a small bulge, two of
them placed in the lower part of it, aiming to the front and the
lateral margins, and the third one, 20% smaller than the
others, is placed in the upper part of the bulge, aiming to the
posterolateral margin. Chelicerae: with two well developed
subdistal teeth. Tergites: I-VI completely smooth or slightly
granular near the posterior margin; VII slightly granular in the
posterior half, with four longitudinal carinae (two parame-
dians, two laterals) marked by coarse granulation, the lateral
carinae occupying almost the entire length of the segment,
whereas the paramedian carinae are restricted to the posterior
half of the segment. Sternites I-IV with smooth tegument,
spiracles small and slightly elliptic; sternite V: smooth in the
anterior half, posterior half slightly granular, with four
longitudinal carinae poorly marked. Metasoma: segment I:
ventral surface with two VL carinae well developed, two VSM
carinae formed by scattered coarse granules diverging
proximally almost forming a transversal carinae, with four
ventral macrosetae (n = 10), two over each VSM carina, with
four VL macrosetae (n - 10), two over each VL carina, with
four distal macrosetae {n = 10) (Fig. 6); lateral surface: LIM
carinae granular and well marked in the posterior three
quarters of the segment, with one macroseta over the anterior
third of the LIM carina, LSM and DL carinae well marked
occupying the entire length of the segment; dorsal surface
smooth. Segment II: similar to segment I, but the LIM carina
is restricted to the second half of the segment, and there is a
macroseta over the posterior third of the LSM carina. In some
specimens there is also a DL macroseta. Segment III: ventral

surface, VSM and VL carinae poorly marked, longitudinal,
and occupying the entire length of the segment; lateral
surfaces, LIM carina restricted to the posterior third of the
segment, the rest similar to segment II. Segment IV: ventral
surface smooth, or with poorly developed VL carinae, with
four to six ventral macrosetae {n = 10; median = 6) and three
or four VL macrosetae (n = 10; median = 3); lateral surface,
LSM and DL carinae represented by a slight elevation of the
tegument, and occupying the entire length of the segment,
lateral surface with one to three LSM setae (n = 10; median =
2) and one DL macroseta (n = 10), the rest of the tegument
smooth. Segment V: very elongated, ventral surface with some
coarse scattered granules in the posterior third of the segment,
VM and VL carinae occupying the entire length of the
segment. Very close to the VL carinae there is a group of
granules parallel to the lateral margin (Fig. 10) that appar-
ently correspond to a VSM carina; with six or seven ventral
macrosetae (n = 10; median =7) and five to seven VL
macrosetae on each lateral margin of the segment {n = 10;
median = 6); lateral surface smooth, LSM carina represented
by a row of six to nine macrosetae (n = 10; median = 9), with
one to three DL macrosetae {n = 10; median = 1). Telson:
vesicle low and elongated in males, globose in females
(Figs. 11, 13), ventral surface slightly granular; dorsal surface
smooth, with a well developed median glandular depression in
males; aculeus very short and curved. Legs: smooth tegument,
with two well developed and symmetrical pedal spurs; telotarsi
low and elongated, with a ventromedian row of hyaline setae
of the same length as the VL spines, and with well developed
ventrolateral spines; spinal formula typical of the group:
tarsus I: 1-1 (n = 15); tarsus 11: 2-2 {n = 15), tarsus III: 4-4 in
most specimens {n = 13), but in some cases there is one
additional external spine 4-5 {n = 2), tarsus IV: 4-5 in most
specimens (n = 12), but in some specimens there is an
additional internal spine 5-5 (n = 3); telotarsal ungues
symmetrical, very curved. Pectines: number of pectinal teeth
in males: 16-18 {n = 10; median = 17); in females: 15-17 (n =
10; median = 16). Pedipalps: femur: DE and VE carinae

352

THE JOURNAL OF ARACHNOLOGY

extending the entire length of the segment, slightly granular in
males, blunt in females; DI and VI carinae granular and well
developed, extending the entire length of the segment. Patella:
DI Carina blunt, with some granules near the articulation with
the femur, extending the entire length of the segment, VI
Carina granular, extending the entire length of the segment,
DE and VE carinae blunt, extending the entire length of the
segment, internal median carina represented by some scattered
granules in the median part of the segment. Chela: slightly
elongated, more robust in males, (Figs. 7, 9); on the internal
surface males with a lobular expansion continued by a slight
depression near the articulation with the movable finger
(Figs. 8, 9); with six carinae poorly developed, most of them
barely visible as a slight elevation of the tegument with some
setae, but the ventral carinae is well developed and bears two
rows of setae, probably a fusion of two different carinae: DI
and DS (in the basal half of the segment), DM, D, E and V +
VM, extending the entire length of the segment. Trichohothrial
pattern: neobothriotaxic major type C, with one accessory
trichobothrium in the V series of chela; femur with 3
trichobothria (1 r/, 1 i and 1 e); patella with 19 trichobothria
(3 V, 2 d, 1 /, 3 et, 1 est, 2 em, 2 esb, and 5 eh)', chela with 27
trichobothria ( 1 Est, 5 Et, 5 V, 1 Esb, 3 Eb, 1 Dt, 1 Db, 1 et, 1
est, 1 esb, 1 eb, 1 dt, 1 dst, 1 dsb, 1 db, 1 ib, 1 it).
Hemispernuitophore: basal portion very developed; distal
lamina short and curved, smaller than the basal portion;
distal crest parallel to and almost transversal to the posterior
margin, with a transversal crest; internal lobe with a small
bilobate apophysis in its external surface that is not connected
with the laminar apophysis (Fig. 2); lobe region well
developed but very simple (Fig. 1), capsular concavity not
very deep, restricted to the anterior margin of the basal lobe,
but occupying most of the frontal surface of the lobe region;
basal lobe very simple, without internal structures; internal
lobe also very simple, almost completely covered by the basal
lobe. We have dissected the hemispermatophores of seven
specimens and have observed no conspicuous differences
between them.

Distribution and habitat. — Urophonius martinezi has only
been collected in Peninsula Valdes and in an area very close to
Puerto Madryn in central eastern Argentinean Patagonia
(Fig. 21).

The Natural Protected Area Peninsula Valdes (NPA-PV) is
the largest unit of conservation of arid ecosystems of
Argentina, it consists of a wide plateau of 4000 km^ located
in north-eastern Chubut province (42°05'^2°53'S, 63°35'-
65°04'W) and has been declared Human Patrimony by
UNESCO in 1999. Geologically it is formed by Oligo-
Miocenic marine sediments and a continuous cover of aeolian
sediments with quaternary gravels (Haller et al. 2001). Its
actual landscape configuration was originated in the Pleisto-
cene (~1 myrs) probably by strong periglacial winds that
caused the deflation of the Gulfs of Nuevo and San Jose
(Haller et al. 2001 ). The mean annual temperature is 14° C and
the average annual precipitation is 175 mm in the coastal
zone, with oscillations between 200 and 225 mm in internal
zones (Barros & Rivero 1982). In this area, as in other arid
ecosystems, the vegetation has a patchy structure alternating
with bare ground (Bisigato & Bertiller 1997). The dominant
physiognomy is a shrub steppe of Chiupdraga avelkmedae.

accompanied by Chiupdraga hystrix Don., Condalia micro-
phylla Cav., Lyciiin chilense Miers, Schinus polygamus (Cav.),
and Prosopidastrum globosiim (Gill.). At the grass layer, the
most common species are S. tenuis, Piptochaetium napostaense
(Speg.) Hack., and Poa ligularis Nees (Bertiller et al. 1981). In
the southern portion of NPA-PV the shrub steppe is replaced
by an herbaceous steppe where Sporobolus rigens Desv. as the
most important species, along with patches of Ch. avelkmedae
and Hyalis argentea D. Don (Bertiller et al. 1981; Leon et al.
1998). The NPA-PV has been the subject of several scientific
contributions; however, there is conflict over its biogeograph-
ical identity. Soriano (1956) classified this area as belonging to
the phytogeographical province of Patagonia. Cabrera &
Willink (1973) included NPA-PV within the Monte province,
whereas Leon et al. (1998) and Elissalde et al. (2002) described
this region as an ecotone Monte-Patagonia. From a zoogeo-
graphical perspective its identity is still confusing because
Morrone (2001a, 2001b) placed NPA-PV at the Patagonia
Central province, while Roig-Junent «&: Flores (2001) classified
it as Monte region. The information about the epigean
arthropod fauna from NPA-PV included in these contribu-
tions is mainly based on sporadic samplings and some
references about the presence of a few isolated taxa (e.g.
Cuezzo 1998; Flores 1998; Ceballos 2008; Crespo 2008;
Ocampo 2008). Thus these contradictions about the biogeo-
graphical identity of NPA-PV are a consequence of a
fragmentary knowledge about its epigeal arthropods fauna.
The first ecological study in Peninsula Valdes including an
intensive seasonal sampling effort (from 2003 to 2006) is the
one performed by the second author (Cheli, unpublished
data). The presence of Urophonius martinezi in the Peninsula
Valdes, its closest relative being U. eugenicus (a typical
southern Patagonian species), adds support to Morrone’s
(2001a) proposal of including NPA-PV in the Patagonian
region, but future studies are needed to clarify this matter.

The actual distribution of U. martinezi cannot be accurately
established with the scarce distribution data we have at this
moment. With the information available, we can only
conclude that this species occurs in the coastal area of central
Chubut province. Some 200 km further north in Valcheta,
southern Rio Negro Province, it is replaced by U. exochus, and
400 km south in Caleta Olivia, in northern Santa Cruz
province, it is replaced by U. eugenicus (Fig. 21). Our data
show that U. martinezi has a restricted distribution (compared
to other species of the genus), but we cannot be certain that it
is restricted only to the NPA-PV and the area close to it, or if
it occurs in a wider area of Chubut province.

Comments on the distribution and activity period of
Urophonius in Patagonia. — In a previous contribution on the
genus Urophonius (Ojanguren-Affilastro 2002) the first author
has suggested that the Patagonian species of group exochus, U.
eugenicus, and the species described in this contribution may
be active on the surface during summer, whereas the rest of the
species of the group are active at the surface during the winter.
We based this conclusion on the dates of collection of the
material deposited in different collections. However, the
Patagonian specimens of Urophonius previously studied by
us were old, manually collected material. The fact that these
specimens were collected in the summer does not necessary
imply that they were actually active in that period; most

OJANGUREN-AFFILASTRO & CHELI— NEW DATA ON UROPHONIUS SCORPIONS

353

Brazil

Uruguay

A A

[Chi lev

4-

Argentina

Atlantic

Ocean

A A-

A,

San Jose
Gulf

Puerto
Madryn I

Peninsula

Valdes

) O o

Nuevo Gulf 1

Atlantic

Ocean

0

25 Km

Pacific

Ocean

A Urophonius brachycentrus
Y Urophonius mahuidensis
■ Urophonius granulatus

# Urophonius somuncura
A Urophonius eugenicus

• Urophonius martinezi n. sp.

<0> Urophonius exochus

X Urophonius tregualemuensis
□ Urophonius tumbensis
O Urophonius sp. exochus group

0

300 km

Figure 21. — Map of the southern part of South America, with the known distribution in this area of the species mentioned in this
contribution. In detail, map of Peninsula Valdes, with the known distribution of Urophonius martinezi new species.

probably this only reflects when most collecting trips in that
area are carried out (during the summer), because of the
extreme cold weather of the Patagonian winter. In a recent
summer trip to Patagonia, no specimens of the exochus or
brachycentrus group were observed during UV collecting at
night (C.I. Mattoni pers. com.); the only active specimens of
Urophonius observed at night were Urophonius granulatus
Pocock 1898. In a recent trip to Patagonia in early winter
(June-July 2008) we observed active specimens belonging to
different species of exochus and brachycentrus groups: U.
exochus, U. eugenicus, U. martinezi, and Urophonius brachycen-
trus (Thorell 1876), but no specimens of the granulatus group.

In addition, extensive collections from 2003 to 2006
performed in NPA-PV using pitfall traps by the second author

have shown that the activity period of Urophonius martinezi is
restricted to late autumn and winter (June, July, August),
making it the only species of scorpion active in the area during
this period of the year.

We have recently observed active populations of Urophonius
tregualemuensis Cekalovic-Kuschevich 1981 in southern Chile
during the spring and summer; this species of the granulatus
group apparently has the same activity period as the
Argentinean species of the granulatus group, U. granulatus
and Urophonius somuncura Acosta 2003, which also have a
spring-summer surface activity period (Maury 1979; Acosta
2003; Ojanguren-Affilastro 2005, 2007). The activity period of
the other known species from southern Chile, Urophonius
tumbensis Cekalovic-Kuschevich 1981, is still not known; we

354

THE JOURNAL OF ARACHNOLOGY

could not collect, nor observe, any active specimen of this
species in areas near the type locality during collection trips in
the summer. The type material of this species could not be
found in its depository at the MZUC (Raul Briones Parra and
Jaime Pizarro Araya pers. com.) and no other specimen has
been collected since the original description of the species.
Unfortunately the original description of U. tumbensis is a
little vague, and we can not assign it to any of the groups of
the genus. Our results show that all known species of exochiis
and braclivceiitnis groups have a winter surface activity
period, whereas all known species of gnmulatiis group have
a spring-summer surface activity period. Taking into consid-
eration that almost all bothriurid species show a spring-
summer surface activity period, the winter activity pattern
could be a synapomorphy for the exochus and brachycentrus
group (C. Mattoni pers. com).

Recently we have collected specimens of U. exochus in a
wide area of northern Patagonia (Neuquen and Rio Negro
provinces) belonging to the Monte phytogeographic province
and to the ecotone between Monte and Patagonia phytogeo-
graphic provinces (Fig. 21). The known distribution of this
species was restricted to Mendoza (in a slightly different
environment) and to some probable records from Neuquen
(Ojanguren-Affilastro 2005). These new records considerably
expand the distribution of these species. We have observed
slight morphological differences between specimens from
different areas, but we prefer to consider them as intraspecific
variation.

Acosta (1988) mentions the presence of a probable
undescribed entity of the exochus group from Perito Moreno,
in western Santa Cruz province, Argentina. We have examined
a male specimen from this locality (probably the one studied
by Acosta) and consider it to be an undescribed species. This
species is closely related to U. eiigeuiciis, but it has different
morphometric proportions; is smaller and more densely
pigmented. We have examined other specimens from south-
eastern Santa Cruz province, and they could also belong to
this new species; however, they are poorly preserved, so we
cannot assure this. Apparently this undescribed species
inhabits areas close to the Andes mountain chain, whereas
U. eugenicus occurs in the eastern part of the province.

New records for Urophonius species from Patagonia. —
Urophonius exochus: Neuquen Province: 6 c?, 12 ?, 18 juveniles,
30 km SW Zapala, (39°01 '04.4"S, 70°09'21.2"W), 2 June 2008,
Ojanguren-Affilastro, Compagnucci & Martinez (MACN-Ar);
7 >?, 16 ?, 23 juveniles, 20 km NE Piedra del Aguila
(39°58'46.5"S, 70°02'20.8"W), Ojanguren-Affilastro, Compag-
nucci & Martinez (MACN-Ar). Rio Negro Province: 1 ?, 1
juvenile, Pajalta, (40°45'0"S, 66°2'60"W), 18 August 1967,
Maury (MACN-Ar); 1 juvenile, 60 km N Nahuel Niyeu,
(40°30'04"S, 66°32'58.9"W), Bachmnn (MACN-Ar); 3 4 ?,

10 juveniles, 15 km N Valcheta, (40°43'05.7"S; 66°1 1 '56.2"W),

Ojanguren-Affilastro, Compagnucci & Martinez, 4 June 2008,
(MACN-Ar); 1 ?, 4 9, 8 juveniles, 8 km W. Choele-Choel,
General Roca Monument (39°14'19"S, 65°40'48.7"W), Ojan-
guren-Affilastro, Compagnucci & Martinez, 3 June 2008,
(MACN-Ar); 3 c?, 5 9, 14 juveniles, 15 km S. Paso Cordova
(General Roca), (39°07'29.7"S, 67°40'37.4"W), 31 May 2008,
Ojanguren-Affilastro, Compagnucci & Martinez, (MACN-
Ar); 1 9, Cerro Villegas, Estancia San Ramon, San Carlos de
Bariloche, (4r04T0.86"S, 71°0.8'52.41"W), 17 July 1968,
Muller, (MACN-Ar).

Comments. — Maury (1973) records the presence of Uropho-
nius mahuiclensis Maury 1973 in the locality of Paja Alta, in the
base of the Somuncura plateau, based on a female and a
juvenile specimen; Ojanguren-Affilastro (2002, 2005) repeats
this. Unfortunately the identification of the specimens of this
group is very difficult and in some cases it is only possible to
identify the male specimens. We have been able to study more
material from that area (including males) and we conclude
that the specimens studied by Maury actually belong to U.
exochus, or at least to a species very closely related to it, but
not to U. mahuiclensis. According to our results U. mahuiclensis
is endemic to the Tandilia and Ventania mountain chains in
southern Buenos Aires province, Argentina, the same as
Bothriurus voyatti Maury 1973 (Maury 1973).

Urophonius eugenicus: Santa Cruz Province: 8 9, 16 juveniles,
15 km S Caleta Olivia, near Cahadon Seco (46°30'36.6"S,
67°27'48"W), 7 June 2008, Ojanguren-Affilastro & Martinez,
(MACN-Ar); 5 9, 7 juveniles, 20 km W. Las Heras
(46°33'35"S, 69°02'23.2"W), 8 June 2008, Ojanguren-Affilastro
& Martinez, (MACN-Ar); 17 S, 5 9, 20 juveniles, 8 km N.
Puerto Deseado, (47°42'42,43"S, 65°50T5.68"W), 6 June 2008,
Ojanguren-Affilastro & Martinez, (MACN-Ar).

Comments. — We have collected this species in the Eastern
part of Santa Cruz province. Specimens from Western Santa
Cruz, previously mentioned as belonging to U. eugenicus
(Ojanguren-Affilastro 2002, 2005) belong to an undescribed
species.

Urophonius brachycentrus: Rio Negro Province: 1 c?, 3 9, 8
juveniles, 30 km E Choele-Choel, (39°15'27.7"S, 65°35'59.3"W),
3 June 2008, Ojanguren-Affilastro, Compagnucci & Martinez,
(MACN-Ar).

Comments. — This is the first time this species has been
collected in Rio Negro province; however, it was previously
collected in nearby areas from the surrounding provinces of
Buenos Aires and La Pampa. In the locality of Cholel-Choel,
this species has been collected very close to U. exochus;
however, both species have different habitat preferences; U.
exochus apparently prefers slopes and areas with some rocks,
whereas U. brachycentrus prefers plains. In the nearby locality
of Paso Cordova, where U. brachycentrus is not present, U.
exochus is present in both types of environment, but it is more
abundant on slopes than in plains.

KEY TO THE PATAGONIC SPECIES OF UROPHONIUS

1. Ventral surface of metasoma with two VL and a VM stripe (Fig. 19) granulatus group

2

Ventral surface of metasoma with two VL and two VSM stripes (Fig. 20) 4

2. e trichobothria of pedipalp femur placed proximally with respect to Ml macroseta (Fig. 15) Urophonius granulatus

e trichobothria of pedipalp femur placed distally or in the same axis with respect to Ml macroseta (Fig. 15) 3

OJANGUREN-AFFILASTRO & CHELI— NEW DATA ON UROPHONIUS SCORPIONS

355

3. e trichobothria of pedipalp femur placed on the same axis or slightly distally with respect to Ml macroseta (Fig. 15)

Urophonius somimcura

e trichobothria of pedipalp femur placed clearly more distally than Ml macroseta (Fig. 15) Urophonius tregualemuensis

4. Pedipalp femur with two macrosetae (Ml and M2) associated with d and e trichobothria (Fig. 16) hracliycentrus group

Urophonius brachycentrus

Pedipalp femur with one macroseta (Ml) associated with d and e trichobothria (as in granulatus group) (Fig. 15) .... exochus group

5

5. Bilobate protuberance of the hemispermatophore connected to the distal lamina (Fig. 5) Urophonius muhuidensis

Bilobate protuberance of the hemispermatophore not connected to the distal lamina (Figs. 2, 3, 4) 6

6. Bilobate apophysis of the hemispermatophore very close to the distal lamina, but not forming a part of it; distal lamina slender,

almost straight, anterior margin slightly curved (Fig. 4) Urophonius exochus

Bilobate apophysis of the hemispermatophore clearly separated from the distal lamina; distal lamina stout, anterior margin
strongly curved (Figs. 2, 3) 7

7. Vesicle of the telson globose, highly developed toward the anterior margin (Figs. 12, 14), length/height ratio of telson in males:
2.54-3.03, n = 18, median = 2.85. Pigment pattern of prosoma occupying most of the area between the ocular tubercle and the

lateral eyes (Fig. 17) Urophonius eugenicus

Vesicle of the telson slender, especially in males, in females it is globose but not highly developed towards the anterior margin
(Figs. 11, 13), length/height ratio of telson in males: 3.10-3.52, n = 20, mean = 3.28. Pigment pattern of prosoma poorly
developed, area between the ocular tubercle and the lateral eyes almost unpigmented (Fig. 18) Urophonius marlinezi

ACKNOWLEDGMENTS

We are grateful to Luis Compagnucci for his help during
field work and to Camilo 1. Mattoni for the information on
summer Patagonian species of Urophonius. We are grateful to
Paula Cushing, Mark Harvey, Camilo Mattoni and Oscar
Francke for their helpful comments on the manuscript. We are
also grateful to Raul Briones Parra and Jaime Pizarro Araya
for the information provided about the type specimen of U.
tumbemis. This research was supported by a postdoctoral
grant from CONICET to the first author, by a doctoral grant
from CONICET to the second author, and by a CONICET
grant (PIP 6502) to the first author. Finally we deeply thank
Idea Wild Foundation for providing field equipment for the
first author and part of the optica! equipment used in this
contribution by the second author.

LITERATURE CITED

Abalos, J.W. & C. Hominal. 1974. Urophonius achalensis, nueva
especie de Bothriuridae. Acta Zoologica Lilloana 31:19-26.
Acosta, L.E. 1988. Contribucion al conocimiento taxon6mico del
genero Urophonius Pocock 1893 (Scorpiones, Bothriuridae).
Journal of Arachnology 16:23-33.

Acosta, L.E. 1998. Urophonius transandinus sp. nov. (Bothriuridae), a
scorpion from Central Chile. Studies on Neotropical Fauna and
Environment 33:157-164.

Acosta, L.E. 2003. Description of a new Patagonian species of
Urophonius Pocock (Scorpiones, Bothriuridae), from Meseta de
Somuncura, Argentina. Zootaxa 187:1-12.

Barros, V.R. & M. Rivero. 1982. Mapas de probabilidad de
precipitaci6n en la provincia de Chubut, Centro Nacional
Patagonico (CONICET), Puerto Madryn, Argentina.

Bertiller, M.B., A.M. Beeskow & M.P. del Irisarri. 1981. Caracteres
Fisionomicos y Floristicos de la Vegetacion del Chubut. 2. La
Peninsula Valdes y el Itsmo Ameghino. Contribucidn No. 41.
CONICET, Centro Nacional Patagonico, Puerto Madryn.
Bisigato, A.J. & M.B. Bertiller. 1997. Grazing effects on patchy
dryland vegetation in northern Patagonia. Journal of Arid
Environments 36:639-653.

Cabrera, A.L. & A. Willink. 1973. Biogeografia de America Latina.

OEA Monografias 13. Serie Biologia.

Ceballos, A. & B. Rosso de Ferradas. 2008. Pseudoscorpiones.
Pp. 105-116. In Biodiversidad de Artropodos Argentines. Volu-

men 2. (L.E. Claps, G. Debandi & S. Roig-Junent, eds.). Sociedad
Entomologica Argentina, Mendoza, Argentina.

Cekalovic-Kuschevich, T. 1981. Dos nuevas especies y un nuevo
registro del genero Urophonius para Chile (Scorpiones, Bothriur-
idae). Boletin de la sociedad Biologica de Concepcidn 52:195-201.

Crespo, F.A. & A. Valverde. 2008. Blattaria. Pp. 167-180. In
Biodiversidad de Artropodos Argentinos. Volumen 2. (L.E. Claps,
G. Debandi & S. Roig-Junent, eds.). Sociedad Entomologica
Argentina, Mendoza, Argentina.

Cuezzo, F. 1998. Formicidae. Pp. 452-462. In Biodiversidad de
Artrdpodos Argentinos. Volumen 1. (J.J. Morrone & S. Coscaron,
eds.). Ediciones Sur, La Plata, Argentina.

Elissalde, N., J. Escobar & V. Nakamatsu. 2002.
Inventario y evaluacion de pastizales naturales de la zona
arida y semiarida de la Patagonia. Instituto Nacional de
Tecnologia Agropecuaria. Centro Regional Patagonia sur. EEA
Chubut, Trelew.

Flores, G.E. 1998. Tenebrionidae. Pp. 232-240. In Biodiversidad de
Artropodos Argentinos. Volumen 1. (J.J. Morrone & S. Coscaron,
eds.). Ediciones Sur, La Plata, Argentina.

Francke, O.F. 1977. Scorpions of the genus Diplocentrus Peters from
Oaxaca, Mexico. Journal of Arachnology 4:145-200.

Haller, M., A. Monti & C. Meister. 2001. Hoja Geologica 4363-1,
Peninsula Valdes, Provincia del Chubut. Boletin 266. Secretaria de
Energia y Mineria, Servicio Geol6gico Minero Argentino, Buenos
Aires.

Leon, R.J.C., D. Bran, M. Collantes, J.M. Paruelo & A. Soriano.
1998. Grandes unidades de vegetacion de la Patagonia extra-
andina. Ecologia Austral 8:125-144.

L6pez, E.C. & M.A. Magnanelli. 2002. Nueva localidad y biologia de
Vachonia sp. Abalos (Scorpiones, Bothriuridae), 356 pp., V
Congreso Argentino de Entomologia, Buenos Aires. Abstract.

Mattoni, C.I. & L.E. Acosta. 2005. A new species of Bothriurus from
Brazil (Scorpiones, Bothriuridae). Journal of Arachnology
33:735-744.

Maury, E.A. 1968. Aportes al conocimiento de los escorpiones de la
Repiiblica Argentina. I. Observaciones biologicas sobre Uropho-
nius brachycentrus (Thorell 1877) (Bothriuridae). Physis
27:407^18.

Maury, E.A. 1969. Observaciones sobre el ciclo reproductivo de
Urophonius brachycentrus (Thorell 1877) (Scorpiones, Bothriuri-
dae). Physis 29:131-139.

Maury, E.A. 1973. Los escorpiones de los sistemas serranos de la
provincia de Buenos Aires. Physis, Sec. C 32:351-371.

356

THE JOURNAL OF ARACHNOLOGY

Maury, E.A. 1977. Comentarios sobre dos especies de escorpiones del
genero Urophoniiis (Bothriuridae). Revista del Miiseo Argentine de
Ciencias Naturales “Bernardino Rivadavia” (Buenos Aires)
5(7):143-160.

Maury, E.A. 1979. Escorpiofauna patagonica. II. Urophonius
granulatus Pocock 1898 (Bothriuridae). Physis, Sec. C 38:57-68.

Morrone, J.J. 2001a. Biogeografia de America Latina y el Caribe.
Zaragoza, Manuales y Tesis SEA 3.

Morrone, J.J. 2001b. Review of the biogeographic provinces of the
Patagonian subregion. Revista de la Sociedad Entomologica
Argentina 60:1-8.

Ocampo, F.C. & E.R. Manzanos. 2008. Scarabaeidae. Pp. 535-558.
In Biodiversidad de Artropodos Argentines. Volumen 2. (L.E.
Claps, G. Debandi & S. Roig-Junent, eds.). Sociedad Entomolo-
gica Argentina, Mendoza, Argentina.

Ojanguren-Affilastro, A. A. 2002. Nuevos aportes al conocimiento del
genero Urophonius Pocock 1893 (Scorpiones, Bothriuridae).
Revista Iberica de Aracnologia 6:181-186.

Ojanguren-Affilastro, A. A. 2005. Estudio Monografico de los
escorpiones de la Reptiblica Argentina. Revista Iberica de
Aracnologia 11:74—246.

Ojanguren-Affilastro, A. A. 2007. A new endemic scorpion species
from the Somuncura Plateau, in northern Patagonia (Scorpiones;
Bothriuridae). Zootaxa 1466:47-56.

Pocock, R.I. 1893. A contribution to the study of neotropical
scorpions. Annals and Magazine of Natural History (6) 12:77-103.

Prendini, L. 2000. Phylogeny and classification of the superfamily
Scorpionoidea Latrielle 1802 (Chelicerata, Scorpiones): an exem-
plar approach. Cladistics 16:1-78.

Prendini, L. 2003. A new genus and species of bothriurid scorpion
from the Brandberg Massif, Namibia, with a reanalysis of
bothriurid phylogeny and a discussion of the phylogenetic position
of Lisposoma Lawrence. Systematic Entomology 28:149-172.

Roig-Juiient, S. & G.E. Flores. 2001. Historia geografica de las
areas aridas de America del Sur Austral. Pp. 257-266. In
Introduccion a la Biogeografia en Latinoamerica: Teorias, Con-
ceptos, Metodos y Aplicaciones. (J.L. Busquets & J.J. Morrone,
eds.). Las prensas de Ciencias, Facultad de Ciencias, UNAM,
Mexico, Distrito Federal.

San Martin, P.R. 1965. Escorpiofauna Argentina. 1. Bothriuridae.
Redescripcion del holotipo y descripcion del alotipo hembra de
Urophonius eugenicus. (Mello-Leitao, 1931). Physis 25:283-290.

San Martin, P.R. & T. Cekalovic-Kuschevich. 1968. Escorpiofauna
Chilena. I. Bothriuridae. Una nueva especie de Urophonius para
Chile. Investigaciones Zoologicas Chilenas 13:81-100.

San Martin, P.R. & L.A. Gambardella. 1974. Redescripcion de
Urophonius iheringi Pocock 1893 y consideraciones sobre morfo-
logia, bioecologia y distribucion. Boletin de la Sociedad Biologica
de Concepcion 47:93-1 19.

Soriano, A. 1956. Los distritos floristicos de la provincia Patagonica.
Revista de Investigaciones Agricolas 10:321-357.

Vachon, M. 1973. Etude des caracteres utilises pour classer les
families et les genres de scorpions (Arachnides). 1. La trichobo-
thriotaxie en arachnologie. Sigles trichobothriaux et types de
trichobothriotaxie chez les scorpions. Bulletin du Museum
National d’Histoire Naturelle (Paris), Ser. 3, 140:857-958.

Manuscript received 17 September 2008, revised 14 January 2009.

2009. The Journal of Arachnology 37:357-362

Foraging strategies and diet composition of two orb web spiders in rice ecosystems

Hafiz Muhammad Tahir; Department of Zoology, University of the Punjab, Lahore, Pakistan.

E-mail: hafiztahirpkl@yahoo.com

AMda Butt: Department of Zoology, University of the Punjab, Lahore, Pakistan
Sher Muhammad Sherawat: Department of Agriculture, District Sheikupura, Pakistan

Abstract. We conducted a field study in September 2007 and 2008 to analyze the foraging activity, natural diets, and
predatory efficacy of Tetragnatha javana (Thorell 1890) (Araneae: Tetragnathidae) and Neoscona theis (Walckenaer 1842)
(Araneae: Araneidae) on selected prey. The relationship between body measurements (carapace width, leg length, total
body length, and body weight) and web dimensions (capture area, capture thread length, number of radii, number of
spirals, and mesh height) of both species was also investigated. Most of the observed T. javana constructed their webs
between two adjacent rice plants, while N. theis placed theirs at the top of rice plants. Both species required approximately
an hour to complete a web, which differed significantly from each other in height, diameter, and capture area. Both species
constructed only a single web per day. Web building activity of both species was intense from 17:00 to 18:00, while prey-
handling activity was high from 19:00 to 20:00. In both species, peaks of feeding were recorded just after the peaks of prey
handling (21:00). The main prey orders caught in the webs of both species were Lepidoptera, Diptera, Homoptera,
Coleoptera, Hymenoptera, and Orthoptera. The time required to reach and capture lepidopteran (adults of stem borer and
leaf folder) and homopteran prey was similar for both species. However, the time required to reach and capture
orthopteran (grasshopper nymphs) prey was significantly longer for T. javana than for N. theis. Capture area increased with
carapace width, and capture thread length increased with carapace width and body weight, while leg length and body length
did not relate to either of these web variables. The number of radii, number of spirals, and mesh height did not correlate
with any of the body size measurements. We concluded that both species can be used effectively to reduce insect pests of
rice fields.

Keywords: Araneidae, agroecosystem, pest suppression, biological control

Spiders are among the most abundant predatory groups in
rice ecosystems (Sebastian et al. 2005; Takashi et al. 2006;
Tahir & Butt 2008). Most of them are polyphagous predators,
able to feed on various insect pests of agricultural crops (Lang
et al. 1999; Hanna et al. 2003; Schmidt et al. 2004; Takashi et
al. 2006). Several studies clearly describe their role in reducing
insect pests in rice fields (Xu et al. 1987; Ye & Wang 1987;
Tanaka 1989; Jalaluddin et al. 2000). Spiders use a variety of
methods to capture prey. Hunting spiders may actively pursue
or ambush prey, while web building spiders present a unique
case of “sit-and-wait” predation (Heiling 1999; Park et al.
1999). Orb web spiders are characterized by the use of a web to
capture prey (Turnbull 1960; Kajak 1965). Prey capture
success of web building spiders is also influenced by spider
size (Eberhard 1990), web size (Sherman 1994), and web
placement (Chacon & Eberhard 1980; Rypstra 1985) as well as
specific web parameters such as strength (Lubin 1986),
adhesiveness (Opell 1994), extensibility (Vollrath 1992), and
mesh size (Rypstra 1982; Eberhard 1986).

Orb web spiders may trap more insects than they can
consume. Silk of some orb web weavers attracts herbivorous
insects that would normally be drawn to flowers and new
leaves (Craig et al. 1996). Up to 1000 insects may be present in
a web at a given moment, and many are left in the web to be
eaten later (Nyffeler et al. 1994a). Small pests, such as thrips,
midges, and aphids, may be caught and die in the webs of large
spiders, only to be ignored by the spiders (Nentwig 1987;
Landis et al. 2000). Web weaving spiders must anchor their
prey-capture devices to the appropriate substratum in order to
increase the effectiveness of their webs; complex habitats

provide appropriate sites for different sizes and types of webs
in prey capture (Rypstra et al. 1999).

Orb webs have developed as an efficient means of capturing
flying insects. The optical properties of these webs tend to
reduce their visibility, especially in low-light and varying
background conditions (Craig 1986), making detection and
avoidance difficult (Robinson & Mirick 1971). There are
numerous reports concerning prey captured by web building
spiders (Heiling 1999; Ibarra-Nunez et al. 2001; Ceballos et al.
2005).

We designed the present study to understand the role of two
nocturnal orb web spiders, Tetragnatha javana (Thorell 1890)
and Neoscona theis (Walckenaer 1842), in the suppression of
insect pests in rice fields. These species were selected because
of their abundance in the rice ecosystems of central Punjab,
Pakistan (Tahir & Butt 2008). The objectives of the study were
to record the differences, if any, in web building, prey
handling, and feeding activities of both orb web weavers, to
study the relationship of body size measures (carapace width,
leg length, total body length, and body weight) with various
web characteristics (capture area, capture thread length,
number of radii, number of spirals, and mesh height), and to
record the difference in time elapsed to reach and capture prey
blown experimentally into their webs. This study will help to
understand the impact of these orb web spiders in the
suppression of insect pests of rice.

METHODS

Study site. — The study was conducted in September 2007
and 2008 in rice fields at the agricultural research farm,

357

358

THE JOURNAL OF ARACHNOLOGY

Sheikupura (31°43'N, 73°59'E). The rice variety grown was
super basmati. At the time of the experiment, the average
height of the plants in the fields was 131 ± 11 cm). During the
course of the study, the temperature fell to approximately 27
± 4° C at night, and rose to about 41 ± 6° C during the day.
The relative humidity was highly variable (65-85%).

Field observations. — We identified T. javana and N. theis by
consulting Barrion & Litsinger (1995). We conducted field
observations on three different days in the third and fourth
week of September (2007 and 2008), respectively, starting each
day at 16:00 h. During the study period, sunset occurred
between 18:00 and 18:30 h and sunrise between 05:00 and
06:00 h. To check the activities of spiders and their webs, we
walked through the field every hour during 24 h to cover a 50
X 50 m plot (the walk required 20 to 25 min). At night, we
used a fiashlight covered with dark red plastic, because it
neither attracted insect prey nor disturbed the spiders’ natural
photoperiod (Herberstein & Elgar 1994; Helling 1999;
Ceballos et al. 2005). On each walk we recorded adult female
spiders present on rice plants with or without a web. Each
spider’s position was marked individually with a numbered
piece of white plastic tied to the nearest twig. Spider activities
recorded were resting, building a web, handling or eating a
prey. Prey counted included both those fed upon at the hub
and ones tangled in the web but not being fed upon. Prey items
in the webs of spiders were identified to order.

Information regarding web diameter, web position (height
from the ground and location on the rice plant), capture area,
capture thread length, mesh size, number of radii, number of
spirals, and time required to complete a web were also
collected (n = 50 each year). To record the data, we removed
spiders from the webs and sprayed their webs with a fine mist
of water and cornstarch, using a knapsack hand sprayer
(THS-1 19428) to improve the resolution. Similarly, carapace
width, length of leg IV (coxa to tarsus), total length, and
weight were also recorded. Since the data did not differ during
the two years (for either species), they were pooled for
statistical analysis.

To estimate the number of prey items per m“ of rice plants,
we quickly covered plants with two plastic bags, and then cut
all plants just above the roots. Arthropod prey were sampled
every two hours (three replicates). Each entire cut rice stem
was brought to the laboratory and carefully examined for
insects. Total webs per ni" were also counted every two hours
(three replicates).

The normality distribution of the data was analyzed with
Kolmogorov-Smirnov tests before conducting further statisti-
cal analysis. Student /-tests were applied to normally
distributed data. Relationships between body size (carapace
width, leg length, body length, and weight) and web size
(capture area, capture thread length, mesh size, number of
radii and number of spirals) were analyzed using Pearson
correlations (Minitab 13.3). Data are presented as mean ± 1
SD.

Prey capture efficiency. — We conducted an experiment to
record prey capture events in the field during the third week of
September 2007. In order to record the prey-capture efficiency
of web weaving spiders, we used four experimental prey types
that are major rice insect pests in the study area: adults of
whitebacked planthoppers Sogatella furcifera (Horvath), leaf

Tinte [hours]

Figure 1. — Relative frequency of web building, prey handling, and
feeding individuals of Tetragnatha javana each hour during 24 h in
rice fields.

rollers Cnaphalocrocis medinalis (Guenee), white stem borers
Scripophaga innotata (Walker), and nymphs of grasshoppers
Hieroglypinis banian (Fabricius). Twenty-five individuals of
each prey type (plant hoppers: 3.4 ± 0.6 mm, grasshoppers: 10
± 2.3 mm, stem borers 9 mm ± 2.4 mm, leaf folders 10 ±
3.4 mm) were collected with a sweep net (42 X 80 cm, with a
90 cm handle) and a suction device (SIEMENS VK 20C0I).
Webs of adult females with no signs of prey were used in the
experiment (Sebastian et al. 2005). Each prey animal was gently
blown into the web with an inverted aspirator from 10 cm
away. An average of twenty replicates was used for each type of
prey, using a different web for each item. All prey were alive and
undamaged before and after their introduction into the web.
Once the prey contacted the web, we measured how long the
spider needed to approach and capture the prey. Prey handling
time began when the spider bit the prey, continued as the prey
was manipulated, and ended when the spider took the prey to
the hub. The predatory efficiency of both web-weaving spiders
on four experimental prey types was compared using two-way
ANOVA (SPSS 13). Subsequently, we conducted a Tukey HSD
test separately for the time it took the spider to reach the prey
and to capture it, with one factor being the spider species (2
levels) and the other being the prey species (4 levels).

RESULTS

Tetragnatha javana. — A total of 214 spiders (123 in 2007
and 91 in 2008) and 135 webs (77 in 2007 and 58 in 2008) was
observed during experimental periods of three days and
nights. Although T. javana built webs at all hours of the day
and night, their web building activity was most intense
between 16:00 and 20:00 h (Fig. 1). Of the total observed,
59% completed web building between 17:00 and 18:00 h. T.
javana constructed their webs between two adjacent rice plants
and required 1 ± 0.3 h to complete a web. Webs averaged 109
± 7.3 cm high and 29 ± 3.9 cm wide, n — 50 both years).
After constructing a web T. javana occupied the center of the
web and quickly attacked prey that attempted to escape.

Prey handling of T. javana was most intense between 19:00
and 20:00 h and decreased until 02:00 h, just after a second
peak of web building. Feeding activity of T. javana was highest
at 21:00 h, with a smaller peak between 03:00 and 04:00 h
(Fig. 1). Combining both years, we observed 135 webs that
contained 993 prey items (mean = 7.4 prey/web: Lepidoptera
(41%), Diptera (24%), Homoptera (15%), Coleoptera (6%),
Hymenoptera (3%), Orthoptera (3%), Araneae (3%) and
unidentified prey (5%) (Table 1).

TAHIR ET AL.— ORB WEB SPIDERS OF RICE ECOSYSTEMS

Table 1. — Relative frequency of pests recorded from the webs of
Tetragnatha Javana {n = 993) and Neoscona theis (n = 849).

Common name

Scientific name

T. javana

N. theis

Lepidoptera

Yellow stem borer

Scripophaga incertidas

14

11

White stem borer

(Walker)

Scripophaga innota (Walker)

12

13

Pink stem borer

Sesamia inferem (Walker)

4

3

Stripped borer

Chilo suppressalis (Walker)

3

2

Sorghum stem borer

Chilo partellus (Swinhoe)

1

-

Leaf folder

Cnaphalocrocis medinalis

5

4

Leaf folder

(Guenee)

Mythimna separata (Walker)

2

3

Homoptera

Whitebacked

Sogatella furcifera (Horvath)

7

9

planthopper
Green leafhopper

Nephotettix nigripictus (Stal.)

4

4

White leafhopper

Cofana spectra (Distant)

4

3

Diptera

Rice gall midge

Pachydiplosis oryzae (W.-M.)

14

14

Rice shoot fly

Atherigona oryzae (Mall.)

6

4

Rice shoot fly

Atherigona soccata (Rond.)

3

3

Mosquito

Culex spp.

1

-

Density of potential prey increased after 16:00, reached a
maximum at 20:00, and then declined until 04:00. Frequency
of prey handling increased with density of potential prey (r =
0.63; P < 0.01, Fig. 1).

Neoscona theis. — We observed 141 spiders (73 in 2007 and
68 in 2008) and 97 webs (57 in 2007 and 40 in 2008). Of these,
63% completed web building between 17:00 and 18:00 h. Web
building activities decreased but did not cease throughout the
night (Fig. 2). Most N. theis (67%) constructed their webs at
the top of rice plants and required 1 ± 0.4 h to build their
webs. The hub of the web averaged 128 ± 7.0 cm above the
ground, and the average diameter of the web was 34 ± 4.7 cm
(« = 50 both years).

Figure 2. — Mean density (± SD) of potential prey per 1 of rice
fields recorded at each two-hour period during 24 h (combined for
both years).

359

Time [hours]

Figure 3. — Relative frequency of web building, prey handling, and
feeding individuals of Neoscona theis each hour during 24 h in
rice fields.

Prey handling activity of N. theis was most intense between
19:00 and 20:00 h, and increased slightly again around 02:00 h,
just after the second highest frequency of web building.
Feeding peaked at 21:00 h, 1 h after peak prey handling
activity. A second minor peak in feeding occurred between
03:00 and 04:00 h. Web building, prey handling, and feeding
activities of N. theis decreased before 06:30 (Fig. 3).

A total of 849 insects was recorded from 97 webs, an
average number of 8.8 prey/web over the two years combined,
consisting of Lepidoptera (36%), Diptera (21%), Homoptera
(16%), Coleoptera (8%), Orthoptera (8%) Hymenoptera (4%),
Araneae (4%), and unidentified prey (3%) (Table 1). Frequen-
cy of prey handling increased with density of potential prey (r
= 0.59; P < 0.01, Fig. 3).

Predatory efficacy. — The time required for the two spiders
to reach prey differed (two-way ANOVA: F\^ 23 = 29.54, P <
0.01, Table 2). Time to reach prey also differed among four
prey types (^3^ 23 = 359.93, P < 0.001). Similarly, the time to
capture prey differed between the two species {F\_ 23 = 209.33,
P < 0.001), as well as among the four prey types {F-^^ 23 =
296.41, P < 0.001, Table 3). Differences were due to the
handling of orthopteran prey (Tukey HSD test).

Relationship of body size and web design. — The web design
of adult females of both species correlated differently with the
various body size measurements (Table 4). Capture areas of
both species’ webs increased significantly with carapace width
(r = 0.51, P < 0.01 for T. javana; r = 0.54, P < 0.01 for N.
theis), and capture thread length increased significantly with
carapace width (r = 0.55, P < 0.01 for T. javana; r = 0.62, P <
0.01 for N. theis), and body weight (r = 0.60, P < 0.01 for T.
javana; r = 0.68, P < 0.01 for N. theis). Leg length and body
length did not correlate with these two web variables. Neither
did the number of radii, number of spirals, and mesh height
correlate with any of the four measurements of body size (P >
0.05). Comparison of web height, diameter, and capture area
of the two species differed significantly (148 = 4.24, P < 0.01;
As = 3.87, P < 0.01; 148 = 11.9, P < 0.001, respectively).
However, the number of spirals and number of radii in the
webs of these two orb web spiders did not differ significantly
(As = 0.42, P > 0.05; t^^ = 0.72, P > 0.05, respectively).

DISCUSSION

This study suggests that most individuals of both species
started to build their webs just after sunset, and kept the
activities of web building, prey handling, and prey eating to a
minimum after sunrise. Most individuals of both species fed

360

THE JOURNAL OF ARACHNOLOGY

Table 2. — Mean time (s, ± SE) to reach four prey types by
Tetragnatlui javana and Neoscomi theis. Row-wise comparisons were
done by Tukey HSD test. * P < 0.005; ns = non significant.

Prey type

T. javana

N. theis

Comparison

Scripophaga innotata (Walker)

9.5 ± 1.0

12.1 ± 1.3

ns

Cnaphalocrocis medinalis

(G lienee)

8.1 ± 2.1

11.7 ± 3.1

ns

Sogatella fiircifera (Horvath)

9.1 ± 2.2

11.4 ± 2.0

ns

Hieroglyphus banian (Fabricius)

31 ± 1.4

14.2 ± 3.5

*

throughout the observation

period;

they seem

to have a

strategy to ’’build, catch, and eat” in a short period (Ceballos
et al. 2005). Prey handling rates were highest at the beginning
of the night in both species due to the high level of prey
activity at this time (Kraker et al. 1999). Prey handling
decreased after 20:00, which might be due to decreased activity
of the prey after 20:00.

Both spiders took less time to reach and capture adult stem
borers, leaf folders, and planthoppers than grasshopper
nymphs in the prey capture efficacy experiment. This difference
might be due to lower efficiency of their venom. Small prey were
usually paralyzed more quickly than larger ones.

Although the main prey items of both web weavers were
Lepidoptera, they both also fed on Diptera, Homoptera,
Hymenoptera, Coleoptera, and Orthoptera. A comparable
composition (insect orders) of potential prey for the web
building spiders Araneus diadematus Clerck 1757 and Argiope
hnieimichi (Scopoli 1772) was described by Ludy (2007). These
insect orders make up the majority of prey of spiders in rice
agroecosystems. A varied diet creates an optimal, balanced
nutrient composition needed for survival and reproduction
(Greenstone 1979; Toft 1995). However, prey groups were not
caught in the spider webs in proportions to their availability in
the habitat. For example, we recorded 180 plant hoppers from
1 m^ of experimental rice field during a high abundance period
in the last week of September, but only 21 (11.7%) from the
webs in this area.

Members of insect orders with good vision and maneuver-
ability in flight, such as Diptera and Hymenoptera (Land
1997), may detect and evade webs, resulting in an under-
representation there. However, good vision and maneuver-
ability in flight is not important in the present study because
both of the species captured prey mainly during night when
visibility was low. Small and slow-flying insects with relatively
large surface areas may be caught in spider webs most easily

Table 3. — Mean time (s, ± SE) to capture four prey types by
Tetragnatlui javana and Neoscona theis. The prey capture started from
the first contact with the prey and ended when the spider took the
prey to the hub. Row-wise comparisons were done by Tukey HSD
test. * P < 0.005; ns = non significant.

Prey type

T. javana

N. theis

Comparison

Scripophaga innotata (Walker)

19.2 ± 3.0

24.4 ± 1.7

ns

Cnaphalocrocis medinalis

(Guenee)

23.4 ± 2.6

27.0 ± 0.0

ns

Sogatella furcifera ( Horvath)

19.7 ± 2.1

17.0 ± 4.3

ns

Hieroglyphus banian

(Fabricius)

84.2 ± 12.8

42.4 ± 5.1

*

Table 4. — Summary of web characteristics and body measure-
ments (mean ± SE) of adult females of Tetragnatha javana and
Neoscona theis.

Characteristic

T. javana

N. theis

Web height (cm)

109.7

H-

7.3

128.0 ± 7.0

Web diameter (cm)

29.3

-F

3.9

34.0 ± 5.0

Capture area (cm^)

91.1

H-

9.4

126.7 ± 16.3

Number of radii

15.0

-i-

3.0

17.0 ± 4.0

Number of spirals

22.0

-F

4.0

27.0 ± 4.0

Mesh height (mm)

1.6

-F

0.6

2.7 ± 0.3

Carapace width (mm)

1.0

-F

0.3

1.9 ± 0.4

Leg IV length (mm)

12.0

-F

1.4

11.0 ± 1.5

Total length (mm)

11.5

-F

0.5

7.4 ± 1.3

Wet weight (mg)

13.0

-F

7.34

110.7 ± 39.0

(Kajak 1965; Nentwig 1982, 1985). In the present study, more
than 70% of the prey caught in the webs of both spiders belonged
to the Lepidoptera, Diptera and Homoptera. More than 90% of
the prey items recorded from the webs of both species were
insects - the remaining 10% were spiders and unidentified prey.
The difference in abundance of prey at different times is due to
the difference in activity of insects at those times.

The webs of orb weaving spiders vary greatly in design;
scientists have interpreted this variation as specialization for
the capture of specific prey types (Walker 1992). Much of the
interspecific variation in web architecture is related to factors
other than prey types, including amount or shape of available
space, presence of conspecifics, lack of previous experience at
a website, presence or absence of water immediately below the
orb, amount of silk available in the glands, and time of day
(Eberhard 1990). Webs of smaller spiders, which are generally
made with thinner threads and less adhesive, have reduced
abilities to capture large prey (Eberhard 1990). Variation in
web design (position, height, capture area, number of radii,
hub position, number of spirals) of both species, as well as
variation within species, was also recorded in this study. The
general web architecture is thought to be genetically deter-
mined (Foelix 1992). Capture area and capture thread length
increased significantly with carapace width in this study, a
result also reported by Heiling et al. (1998). A large capture
area results in high prey interception (Chacon & Eberhard
1980), and by increasing the distance between sticky spirals,
spiders may enlarge the overall capture area without
increasing their energy expenditure (Herberstein et al. 2000).
Our result is in accordance with previous studies that also
found a positive relationship between carapace width and web
size (Eberhard 1988; Heiling et al. 1998). Mesh height did not
relate to any of the body measurements in the present study,
contrasting with the results of Eberhard (1988), who found leg
length to be a good indicator of mesh height. Numerous field
studies have also failed to find a consistent relationship
between mesh height and prey size (Herberstein & Elgar 1994;
Herberstein & Heiling 1998). A narrow mesh may facilitate the
retention of larger prey, as more threads are in contact with
the item (Eberhard 1990). However, more spiral turns also
reflect more light, thus increasing the visibility of the web to
the prey (Craig 1986; Craig & Freeman 1991). Mesh height
may therefore indicate a compromise between prey retention
and web visibility.

TAHIR ET AL.— ORB WEB SPIDERS OF RICE ECOSYSTEMS

The possible role of these spiders in pest control can be
estimated with simple calculations. The average number of
pests and webs from 1 of rice fields of Punjab were 140 and
3.5, respectively. The average number of pests recorded from a
single web of T. javana was 7.4, while the average number of
pests collected from a single web of N. theis was 8.8. Thus,
these two web builders {T. javana and N. theis) can produce up
to a 22% reduction of the total pest population per day.
Furthermore, many other spider species and other natural
predators also contribute to the suppression of insect pests in
the rice ecosystem. Both spiders studied can reduce the
populations of insect pests in rice fields and may be useful in
biological control of rice insect pests in Pakistan. However, in
order to use them as biological control agents on a broader
scale, further knowledge of their feeding habits, web
construction behavior, reproductive strategies, prey preferenc-
es, and response to insecticides and herbicides is needed.

ACKNOWLEDGMENTS

The research was supported by the Department of Zoology,
University of the Punjab, Lahore, Pakistan. Stano Pekar and
two anonymous reviewers provided very useful and relevant
comments on an earlier version of manuscript.

LITERATURE CITED

Barrion, A.T. & J.A. Litsinger. 1995. Rice land spiders of South and
Southeast Asia. CAB International, Wallingford, Oxfordshire, UK.
Ceballos, L., Y. Henaut & L. Legal. 2005. Foraging strategies of
Eriophora edax (Araneae, Araneidae): a nocturnal orb-weaving
spider. Journal of Arachnology 33:509-515.

Chacon, P. & W.G. Eberhard. 1980. Factors affecting numbers and
kinds of prey caught in artificial spider webs, with consideration of
how orb webs trap prey. Bulletin of the British Arachnological
Society 5:29-38.

Craig, C.L. 1986. Orb-web visibility: the influence of insect flight
behaviour and visual physiology on the evolution of web designs
within the Araneoidea. Animal Behaviour 34:54-68.

Craig, C.L. & C.R. Freeman. 1991. Effect of predator visibility on the
prey encounter: a case study on aerial web weaving spiders.
Behavioral Ecology and Sociobiology 29:249-254.

Craig, C.L., R.S. Weber & G.D. Bernard. 1996. Evolution of
predator prey systems: spider foraging plasticity in response to
the visual ecology of prey. American Naturalist 147:205-229.
Eberhard, W.G. 1986. Effect of orb-web geometry on the prey
interception and retention. Pp. 70-100. In Spiders: Webs, Behavior
and Evolution. (W. Shear, ed.). Stanford University Press, Stanford,
California.

Eberhard, W.G. 1988. Behavioral flexibility in orb web construction:
effect of supplies in different silk glands and spider size and weight.
Journal of Arachnology 16:295-302.

Eberhard, W.G. 1990. Function and phylogeny of the spider webs.

Annual Review of Ecology and Systematics 21:341-372.

Foelix, R.F. 1992. Biologic der Spinnen. Edition 2. Thieme, Stuttgart.
Greenstone, M.H. 1979. Spider feeding behaviour optimizes dietary
essential amino acid composition. Nature 282:501-503.

Hanna, R., G.F. Zalom & J.W. Roltsch. 2003. Relative impact of
spider predation and cover crop on population dynamics of
Erythroneura variabilis in a raisin grape vineyard. Entomologia
Experimentalis et Applicata 107:177-191.

Heiling, A.M. 1999. Why do nocturnal orb-web spiders (Araneidae)
search for light? Behavioral Ecology and Sociobiology 46:43-49.
Heiling, A.M., M.E. Herberstein & G. Spitzer. 1998. Calculation of
capture thread length in orb webs: evaluation of a new formula.
Annals of the Entomological Society of America 91:135-138.

361

Herberstein, M.E. & M.A. Elgar. 1994. Foraging strategies of
Eriophora transmarimi and Nephila phmnpes (Araneae): nocturnal
and diurnal orb-weaver spiders. Australian Journal of Ecology
19:451^57.

Herberstein, M.E., A.C. Gaskett, D. Glencross, S. Hart, S. Jaensch &
M.A. Elgar. 2000. Does the presence of potential prey affect web
design in Argiope keyserlingi (Araneae, Araneidae)? Journal of
Arachnology 28:346-350.

Herberstein, M.E. & A.M. Heiling. 1998. Does mesh height influence
prey length in orb-web spiders? European Journal of Entomology
95:367-371.

Ibarra-Nunez, G., J.A. Garcia, J.A. Lopez & J.P. Lachaud. 2001.
Prey analysis in the diet of some ponerine ants (Hymenoptera:
Formicidae) and web-building spiders (Araneae) in coffee planta-
tions in Chiapas, Mexico. Sociobiology 37:723-756.

Jalaluddin, S.M., R. Mohan, R. Velusamy & S. Sadakathulla. 2000.
Predatory behaviour in rice varieties under sodic soil conditions.
Entomology 25:347-350.

Kajak, A. 1965. An analysis of food relations between the spiders
Araneus cornutus Clerck and Araneus quadratus Clerck and their
prey in meadows. Ekologia Polska Series A 13:717-764.

Kraker, J.D., V.A. Huis, L.K. Heong, J.C.V. Van Lenleren & R.
Robbinge. 1999. Population dynamics of rice leaffolder and their
natural enemies in irrigated rice fields in Philippines. Bulletin of
Entomological Research 89:411-421.

Land, M.F. 1997. Visual acuity in insects. Annual Review of
Entomology 42:147-177.

Landis, D.A., S.D. Wratten & G.M. Gurr. 2000. Habitat manage-
ment to conserve natural enemies of arthropod pests in agriculture.
Annual Review of Entomology 45:175-201.

Lang, A., J. Filser & J.R. Henschel. 1999. Predation by ground beetles
and wolf spiders on herbivorous insects in maize crop. Agriculture,
Ecosystems & Environment 72:189-1999.

Lubin, Y.D. 1986. Web building and prey capture in the Uloboridae.
Pp. 132-171. In Spiders: Webs, Behavior and Evolution. (W.
Shear, ed.). Stanford University Press, Stanford, California.

Ludy, C. 2007. Prey selection of web building spiders (Araneidae) on
field margins. Agriculture, Ecosystems & Environment 1 19:368-
372.

Nentwig, W. 1982. Why do only certain insects escape from a spider’s
web? Oecologia 53:412-417.

Nentwig, W. 1985. Prey analysis of four species of tropical orb-
weaving spiders (Araneae: Araneidae) and a comparison with
araneids of the temperate zone. Oecologia 66:580-594.

Nentwig, W. 1987. The prey of spiders. Pp. 249-263. In Ecophysi-
ology of Spiders. (W. Nentwig, ed.). Springer-Verlag, Berlin.

Nyffeler, M., W.L. Sterling & D.A. Dean. 1994. How spiders make a
living. Environmental Entomology 23:1357-1367.

Opell, B.D. 1994. Increased stickiness of prey threads accompanying
web reduction in the spider family Uloboridae. Functional Ecology
8:85-90.

Park, T.S., J. Jeon, H.K. Lim, S. Lee & J.C. Choe. 1999. Web
orientation in a golden orb-web spider Nephila clavata (Araneae;
Tetragnathidae). Korean Journal of Biological Science 3:161-165.

Robinson, M.H. & H. Mirick. 1971. The predatory behavior of the
golden-web spider Nephila clavipes (Araneae: Araneidae). Psyche
78:123-139.

Rypstra, A.L. 1982. Building a better insect trap: an experimental
investigation of prey capture in a variety of spider webs. Oecologia
52:31-36.

Rypstra, A.L. 1985. Aggregations of Nephila clavipes (L.) (Araneae,
Araneidae) in relation to prey availability. Journal of Arachnology
13:71-78.

Rypstra, A.L., P.E. Carter, R.A. Balfour & S.D. Marshall. 1999.
Architectural modifications of agricultural habitats and their impact
on the spider inhabitants. Journal of Arachnology 27:371-377.

362

THE JOURNAL OF ARACHNOLOGY

Schmidt, M.H., C. Thies & T. Tscharntke. 2004. Landscape context
of arthropod biological control. Pp. 55-63. In Ecological
Engineering for Pest Management: Advances in Habitat Manip-
ulation for Arthropods. (G.M. Gurr, S.D. Wratten & M.A. Altier,
eds.). CSIRO Press, Collingwood, Australia.

Sebastian, A.P., M.A. Thew, B.S. Pathummal, J. Joseph & R.C. Biju.
2005. The spider fauna of the irrigated rice ecosystem in central
Kerala, India across different elevational ranges. Journal of
Arachnology 33:247-255.

Sherman, P.M. 1994. The orb-web: an energetic and behavioural
estimator of spider’s dynamic foraging and reproductive strategies.
Animal Behaviour 48:19-34.

Tahir, H.M. & A. Butt. 2008. Activities of spiders in rice fields of
central Punjab, Pakistan. Acta Zoologica Sinica 54:701-711.

Takashi, M., E Chikara, T. Motonori, M.T. Mihoko, T. Ayame & K.
Yasuhisa. 2006. Effects of tillage practices on spider assemblage in
rice paddy fields. Applied Entomology and Zoology 41:371-381.

Tanaka, K. 1989. Movements of spiders in arable land. Plant Pro-
tection 43:34-39.

Toft, S. 1995. Value of the aphid Rhopalosiplium padi as food for
cereal spiders. Journal of Applied Ecology 32:552-560.

Turnbull, A.L. 1960. The prey of the spider Linyphia triangularis
(Clerck) (Araneae: Linyphiidae). Canadian Journal of Zoololgy
38:859-873.

Vollrath, F. 1992. Spider webs and silks. Scientific American 266(3):
46-52.

Walker, J.R. 1992. What do orb webs catch? Bulletin of the British
Arachnological Society 9:95-98.

Xu, J.S., Z.F. Chen & R.L. Zhu. 1987. Study and application of spiders
in rice fields in Zhejiang Province. Natural Enemy 9:140-144.

Ye, Z.X. & D.D. Wang. 1987. Population dynamics of spiders in rice
fields of Jiangxi Province. Bulletin of Biological Control 3:11-14.

Manuscript received 19 October 2008, revised 2 May 2009.

2009. The Journal of Arachnology 37:363-364

BOOK REVIEW

Dominican Amber Spiders: a Comparative Palaeontological-Neontoiogicai Approach to Identification,
Faunistics, Ecology and Biogeography. By David Penney. 2008. Siri Scientific Press, Manchester, UK.
176 pp. ISBN 978-0^9558636^0=8. £40.

David Penney is a Visiting Research Fellow at the
University of Manchester, UK and an expert on fossil spiders
preserved in amber. This beautifully illustrated book has in
excess of 330 figures, including more than 80 high-quality
color photos. The author uses a unique, integrated approach,
that combines and compares information derived from both
fossil and living spiders. There are eight main sections, the first
being a general introduction to amber, classification of spiders
and their fossil record, as well as information on other
important fossil localities, modes of preservation, the timing of
important radiations in spider evolutionary history, co-
evolution with their insect prey and the effects of mass
extinction events. Much of this information is neatly
illustrated in an accurate and comprehensive figure of the
spider evolutionary tree.

Chapter 2 provides thorough coverage of the current state
of knowledge with regard to Dominican amber, including
geological age and botanical origin, with a fully referenced
discussion of competing ideas. This is followed by sections on
chemical and physical properties, how to identify fakes, and a
comprehensive coverage of tissue preservation in amber,
including whether or not it is possible to extract DNA. A
full description of the journey of amber from mine to museum
is presented, based on the author’s own experience of visits to
the amber mines. This is followed by an extremely useful
section on the methods of amber preparation and study,
including general and advanced photography and microscopy,
with coverage and amazing images of new techniques such as
those provided by x-ray computed tomography. In particular,
this section is broadly applicable to any amber and brings
together information from a great deal of literature, which is
often hidden in obscure specialist journals. This chapter ends
with short discussions on the major world collections of
Dominican amber inclusions, how to conserve and curate
amber collections and the current state of knowledge of
biodiversity of all fossils preserved in Dominican amber. This
is fully up-to-date, and readers are directed to the most
important publications on Dominican amber biodiversity,
including the most recent, which lists all 1,404 fossil species
known to occur in Dominican amber.

Chapters 3 to 5 are more spider orientated, the first
providing an interesting historical account of Hispaniolan
spider research with regard to both the extant and fossil
faunas, which are treated separately. All publications describ-
ing new species or providing new records for the island are
included. This is followed by a full checklist by family,
including both fossil and extant species (495 species in 52

families), with the fossil species clearly indicated. It is worth
noting at this point that no new species are described in this
book, but one new combination is proposed. Chapter 4 is a
fully illustrated key to Hispaniolan spider families, with a
strong emphasis on those characters that are likely to be
observed in fossils. The key is somewhat limited in scope
because it refers only to Hispaniolan species (although that is
the purpose of the book). For example, the author distin-
guishes Nephilidae from Tetragnathidae by the presence of
humps on the carapace of the former. Whilst this holds true
for the Neotropical N. clavipes it is not the case for all species
in this genus. In my opinion, there are a few other incorrect
statements within the key, but it is impossible to avoid such
inaccuracies in rather generalized dichotomous keys. The key
is preceded by suites of highly distinctive characters with
regard to general body morphology, which will negate the
need to use the key in many instances and allow the reader to
skip directly to the next chapter, which consists of the family
descriptions. Each family entry contains the following sub-
headings: Dominican amber, extinct taxa; Hispaniola, extant
taxa; Identification; Natural history; Relevant publications;
Additional notes. Again, this chapter is beautifully illustrated
throughout with photographs and illustrations of both fossil
and living spiders. Whilst the general aim is to permit
identification only to family level, in many cases it will allow
identification to genus and species. The correct assignment of
some species attributed to various families by earlier workers
is questionable in my opinion, but in most instances this is not
discussed, presumably to maintain a balance for non-
arachnologists who are interested in amber spiders.

Chapter 6 covers various aspects of palaeoecology and
historical biogeography, with particular reference to spiders,
but is of much broader significance with regard to the fossil
amber fauna and Caribbean biogeography in general. Topics
covered include the site of resin secretion, the entrapment
process, whether different ambers trap organisms in the same
way, bias in the amber fauna, comparison of fossil and extant
faunas. All information is easily accessible and graphically
portrayed, using excellent color figures. An extensive analysis
of the biogeographic origins of Hispaniolan spiders is
published here for the first time, based on large datasets of
spider distribution and the current knowledge of the
geological origins of the Caribbean. Once again, the scope of
this chapter extends far beyond the amber spider world. This
chapter ends with predictions that can be made for the
Hispaniolan spider fauna based upon what is known about the
fossils. Chapter 7 provides a useful referenced checklist of all

363

364

THE JOURNAL OF ARACHNOLOGY

other arachnids described from Dominican amber, including
18 families of Acari, in addition to the orders Amblypygi,
Opiliones, Pseudoscorpiones, Scorpiones and Solifugae. The
book ends with an extensive bibliography of more than 350
entries, providing an invaluable resource for anybody
interested in amber or Caribbean spiders, followed by an
index. Unfortunately, the index does not list spiders by genera,
although families are listed and it is not too much effort to
determine whether or not a genus is included from the
Hispaniolan spider checklist in Chapter 3 or from the relevant
family page in Chapter 5. The page number for Megarachne
has also been accidentally omitted. In addition there are
several typos in the text and very minor inconsistencies. For
example, the orientation of the spider evolutionary tree
relative to the page differs on pages 18 and 138.

In summary, this book will be of considerable interest
beyond the Dominican amber spider world and represents a
very important contribution to studies on Caribbean bioge-
ography and palaeobiogeography, the literature on amber, the
fauna of Hispaniola (both fossil and extant), and an

identification aid for workers in the Caribbean region. David
Penney is undoubtedly the world expert in this field and has
compiled a comprehensive synthesis with beautiful illustra-
tions on almost every page, making it a pleasure to the eye.
The information is sound and reliable, the bibliography
extensive and complete, and the text is authoritative. There
is no other work available quite like it. Despite the minor
shortcomings mentioned above, I evaluate this book very
highly and would recommend it to be on the bookshelves of
academic libraries and museums, in addition to those of
people with a general interest in spiders or amber, both
amateur and professional.

The book is available from the publisher (http://www.
siriscientificpress.co.uk, email: ) or from the author via e-mail
(david.penney@manchester.ac.uk).

Yuri M. Marusik: Institute for Biological Problems of the
North, Russian Academy of Science (IBPN RAS),
Portovaya Str. 18, Magadan 685000, Russia. E-mail:
yurmar@mail.ru

2009. The Journal of Arachnology 37:365-367

SHORT COMMUNICATION

Habitat selection and potential antiherbivore effects of Peucetia flava (Oxyopidae) on

Solaniim thomasiifoliiim (Solanaceae)

Giuliano B. JacobuccI: Institute de Biologia, Universidade Federal de Uberlandia, CEP 38400-902, Uberlandia, MG,
Brazil. E-mail: jacobucci@inbio.ufu.br

Lenice Medeiros: Departamento do Patrimonio Genetico, Secretaria de Biodiversidade e Florestas, Ministerio do Meio
Ambiente e da Amazonia Legal, CEP 70818-900, Brasilia, DF, Brazil

Joio Vasconcellos-Neto: Departamento de Zoologia, Institute de Biologia, Universidade Estadual de Campinas, CEP

13083-970, CP 6109, Campinas, SP, Brazil

Gustavo Q. Romero: Departamento de Zoologia e Botanica, Instituto de Biociencias, Letras e Ciencias Exatas (IBILCE),
Universidade Estadual Paulista, CEP 15054-000, Sao Jose do Rio Preto, SP, Brazil

Abstract. Several spider species use plants as shelter and foraging sites, but the relationships among these organisms are
still poorly known. Lynx spiders of the genus Peucetia do not build webs, and many species live strictly in plants bearing
glandular trichomes. Peucetia flava Keyserling 1877 inhabits Solamtm thomasiifoHum in southeastern Brazil and usually
preys on herbivores and other small insects adhered to the glandular trichomes of its host plant. To evaluate the potential
anti-herbivore protection of this spider species for S. thomasiifoHum, we glued termites used as herbivore models on
trichomes of S. thomasiifoliiim and on neighboring plants lacking glandular trichomes. Leaf miner damage and spider
density were recorded for S. thomasiifolium plants in July 1997. There was a positive relationship between plant size and
spider density. The removal of termites in S. thomasiifolium by P. flava was higher than in plants without glandular
trichomes. The leaf miner damage was negatively related to spider density. Our results suggest that P. flava may be an
important plant bodyguard in the defense of S. thomasiifolium from its natural herbivores.

Keywords: Animal-plant interactions, host plant specificity, lynx spider, plant protection

It is widely known that several spider species use plants as shelter
and foraging sites (Foelix 1996), but only in the last few decades have
strict associations of spiders with particular plant types or species
been described in detail. A variety of spider-plant associations has
been described in the Neotropical regions (Barth et al. 1988; Dias &
Brescovit 2004; Romero 2006). The associations typically occur
because plants have morphological traits that provide suitable
foraging, mating, and egg-laying sites for the spiders, shelter for
adults and immatures, and nurseries for spiderlings (Romero &
Vasconcellos-Neto 2005), thereby improving the probability of the
spiders living on them. One of these associations includes jumping
spiders (Salticidae) that are strictly associated with Bromeliaceae
(Romero & Vasconcellos-Neto 2004a, 2004b, 2005; Romero 2006).
For one of these associations, Romero et al. (2006) demonstrated
through experiment that the jumping spider Psecas chapoda Peckham
& Peckham 1894 contributed to 18% of the total nitrogen of its host
terrestrial bromeliad Bromelia halansae, and in general, plants with
spiders produced leaves 15% longer than plants from which the
spiders were excluded.

Other associations involve spiders that forage on plants with
glandular trichomes. For instance, at least 10 lynx spider species of
the genus Peucetia (Oxyopidae) live strictly in several plant families and
species bearing glandular trichomes in many distinct vegetation types in
Neotropical, Nearctic, Palearctic, and Afrotropical regions. The main
plant families used by these spiders are Solanaceae, Asteraceae, and
Melastomataceae (Vasconcellos-Neto et al. 2007). The specialization of
the Peucetia species for plants bearing glandular trichomes may have
evolved because insects adhering to these sticky structures may be used
as prey by the spiders (Vasconcellos-Neto et al. 2007). Although
glandular hairs have probably evolved as a defense against herbivores

and pathogenic fungi (Levin 1973, Duffey 1986), they can also mediate
mutualistic interactions between plants and spiders or other predators
(Dolling & Palmer 1991; Romero et al. 2008).

In South America, Peucetia rubroliueata Keyserling 1877 and P.
flava Keyserling 1877 are the most common representatives of the
genus. Both species are widely distributed throughout Brazil and
occur sympatrically in several localities (Santos & Brescovit 2003;
Vasconcellos-Neto et al. 2007). In a forest reserve in the state of
Espirito Santo, southeastern Brazil, P. flava is frequently observed in
association with Solanum thomasiifolium Sendtn. 1846, a solanaceous
plant bearing glandular hairs (Vasconcellos-Neto et al. 2007). While
investigating the occurrence of spiders on 70 individuals of S.
thomasiifolium, we observed that all individuals were inhabited by P.
flava, while none of their neighboring plants (« = 80) had spiders.

To understand this spider-plant interaction better, we addressed the
following questions: 1) Is P. flava abundance related to S.
thomasiifolium size? 2) Are insect predation rates by P. flava on S.
thomasiifolium higher than predation rates by other predators (e.g.,
ants, other spiders) on neighboring plants without glandular hairs?

This study was performed in the Reserva Natural da Vale do Rio
Doce ( 19°26'S, 40°03'W), 30 km north of the city of Linhares, state of
Espirito Santo, a forest reserve covered mainly by Atlantic Forest
vegetation. The reserve comprises 22,000 ha of forest, elev. 28-65 m
above sea level, and a soil rich in sand. The weather is seasonal, with
rainfall varying monthly from 30 mm in the dry season (June) to
226 mm in the wet season (January). Mean annual rainfall is
1320 mm. Temperature varies from 10° to 30° C (Jesus 1988).
Samplings were done in July 1996 and July 1997, in a 4-ha area of
common, nativo vegetation on sandy soil, dominated by cacti,
bromeliads, herbs, and small shrubs (Peixoto & Gentry 1990).

365

THE JOURNAL OF ARACHNOLOGY

6 '

s ■

I ^ '

e

I''

|2-

I

366

A

Volume (mS)

B

Volume (in3)

Figure 1. — Relation between Peucetia jJava abundance and Sola-
niim thonuisiifoUiim size. A. 1 996 (« = \l),y = — 1 .543 + 0.840 .v, R' =
0.671, F= 30.569, P < 0.001; B. 1997 (h = 59), j; = 5.710+ 1.810.T,
R- = 0.428, F = 42.618, P < 0.001.

To verify whether the abundance of Peucetia jiava correlates to the
size of Solanum thomasiifolium plants, we estimated the size of plants
in a 50-m random transect, in each year by measuring the maximum
height of branches with leaves, the maximum canopy diameter, and
the perpendicular length to this diameter. These measures were then
multiplied to estimate plant size in cubic meters. The number of
spiders was recorded on each plant by inspecting branches, leaves,
and stems. This relationship was tested using linear regression.

We assessed the effect of spiders as bodyguards on S. thomasiifo-
lium by using termite workers as herbivore models. We used termites
instead of leaf-mining larvae because the former are thicker and less
tender and, thus, easier to manipulate. In addition, once we attached
termites to the surface they were unable to fly; thus, we could
accurately judge predation rate. Fifteen plants of S. thomasiifolium
and an equal number of similarly-sized neighboring plants without
glandular trichomes were randomly selected in the study area. Each
plant received ten termites randomly positioned on leaves of different
branches, and after 30 min we recorded the number of individuals

0 +- — > I

Solanum Other plants

Figure 2. — Number of termites removed (mean ± SD) from
Sokimmi thomasiifolium and other neighboring plants without
glandular hairs.

removed. Termites were affixed with non-toxic white glue (Cascolar®)
to plants without glandular hairs to prevent them from falling off.
This glue had no influence on spider behavior (unpublished data,
GQR). Additionally, previous studies have used similar methods for
the same purpose and shown that this type of glue does not interfere
with predatory (e.g., ant) behavior (e.g., Oliveira et al. 1987). All the
plants were checked once during the experiment in an attempt to
record predation events. The removal rate of termites from both plant
types was compared using a ?-test for independent samples.

To estimate the relationship between P. Jiava and herbivory on S.
thomasiifolium, we calculated spider density and estimated leaf miner
damage in July 1997. We evaluated leaf miner herbivory because it is
the commonest kind of damage caused by herbivores on S.
thomasiifolium in the study area (personal communication). Plants
were randomly selected, and four branches of each individual were
evaluated for leaf miner damage. The numbers of intact and damaged
leaves were recorded for 20 plants, and the ratio between damaged
and total leaves was calculated. Spider density was estimated as the
ratio between the number of spiders and the plant size in cubic meters.
The relationship was tested using linear regression. Data normality
and homoscedasticity were verified. Logarithm transformations were
applied when necessary prior to the analyses (Zar 1999).

Voucher specimens of the spiders collected (males and females)
were deposited in the Arachnological Collection of the Laboratorio
de Artropodes Pe^onhentos, Institute Butantan, Sao Paulo (accession
numbers: IBSP 12982, IBSP 12940, IBSP 12891, and IBSP 12887).
Solanum thomasifolium exsiccates were deposited at Universidade
Estadual de Campinas Herbarium (UEC-Herbarium).

In 1996, we examined 17 Solanum thomasiifolium plants and found
57 Peucetia flava individuals. In 1997, we recorded 694 spiders on 59
plants. Linear regressions indicated that spider abundance was
positively correlated with plant size during both sampling periods
(Figure 1). Figure 2 displays the results of the termite removal
experiment on the 30 observed plants, with significantly higher
predation rates on S. thomasiifolium than for other plants (/ = 3.88; df
= 28; P < 0.001). Peucetia flava accounted for 11 of the 17 termite
predation events observed on S. thomasiifolium (Figure 3). The other
six events were recorded on plants without glandular trichomes.
Those termites were removed by ants (« = 4) and salticid spiders (n =
2).

Larger plants had more spiders, probably because they provided
more suitable habitat sites for the spiders. In a study of salticid-
bromeliad association, Romero et al. (2007) showed that Coiyphasia
monteverde Santos & Romero 2007 inhabited large rosettes of

JACOBUCCI ET AL.— ANTIHERBIVORE AFFECTS OF PEUCETIA FLA VA

367

Figure 3. — Pence tia flava preying on termite placed on Solarium
thomasiifoUum (Photo G. B. !.)•

Aechmaea distichantha and suggested that these spiders may actively
select their microhabitats based on host plant size. Additionally, large
plants may also represent a more suitable resource for many insects that
constitute the main prey of the spiders. This hypothesis was proposed by
Romero & Vasconcellos-Neto (2004a), who reported that another
bromeliad-dwelling salticid, Eustiromastix nativo Santos & Romero
2004, had a similar microspatial distribution on two bromeliad species,
possibly because larger plants have a higher probability of being visited
by insects as a result of their large surface area.

Although we have no replicates or exclusion experiments demon-
strating the influence of spiders on leaf miner density, observational
and correlative data associated with the experiment using termites as
herbivore models suggest that P. flava might act as an important
plant bodyguard. The higher predation rates of the termite prey
models on S. thomasiifoUum compared to species without glandular
trichomes suggest that P. flava is a very active insect predator,
apparently surpassing the performance of sympatric ants and salticid
spiders.

The role of Peucetia species in the reduction of the herbivore
population and its destructive impact has already been already
demonstrated in studies with other plants. For instance, Louda (1982)
has shown that F. viridans (Hentz 1 832) can decrease herbivore damage
to its host plant, Haplopappus venetus (Asteraceae). In addition,
Romero et al. (2008) recently showed that P. flava and P. rubrolineata
maintain mutualistic relationships with their host plant, Trichogoniopsis
adenantha (Asteraceae), the spiders removing herbivores from their host
plants and the glandular hairs of the plants improving the spiders’
growth by facilitating predation on adhering insects.

In this study, we have provided evidence that the relationship
between P. flava and S. thomasifolium might be mutualistic. However,
more definite conclusions should be obtained in the future by using
spider exclusion experiments in the field.

ACKNOWLEDGMENTS

The authors thank Prof. Dr. Marcelo de Oliveira Gonzaga and
Prof. Dr. Adalberto Jose dos Santos for their comments and help in
preparing this manuscript. This paper was supported by Curso de
P6s-Graduagao em Ecologia, UNICAMP and CAPES.

LITERATURE CITED

Barth, F.G., E.A. Seyfarth, H. Bleckmann & W. Schiich. 1988.

Spiders of the genus Cupiennius Simon 1891 (Araneae, Ctenidae). I.

Range distribution, dwelling plants, and climatic characteristics of
the habitats. Oecologia 77:187-193.

Dias, S.C. & A.D. Brescovit. 2004. Microhabitat selection and co-
occurrence of Pachistopelma rufonigrum Pocock (Araneae, Ther-
aphosidae) and Nothroctenus fuxico sp. nov. (Araneae, Ctenidae)
in tank bromeliads from Serra de Itabaiana, Sergipe, Brazil.
Revista Brasileira de Zoologia 21:789-796.

Dolling, W.R. & J.M. Palmer. 1991. Pwmem/ea (Hemiptera: Miridae):
predaceous bugs specific to the highly viscid plant genus Roridula.
Systematic Entomology 16:319-328.

Duffey, S.S. 1986. Plant glandular trichomes: their partial role in
defense against insects. Pp. 151-172. In Insects and the Plant Surface.
(B. Juniper & R. Southwood, eds.). Edward Arnold, London.

Foelix, R.F. 1996. Biology of Spiders. Second edition. Oxford
University Press, Oxford, UK.

Jesus, R.M. 1988. A reserva florestal da CVRD. Anais do VI
Congresso Florestal Estadual, Volume 1, pp. 59-112.

Levin, D.A. 1973. The role of trichomes in plant defense. Quarterly
Review of Biology 48:3-15.

Louda, S.M. 1982. Inflorescence spiders: a cost/benefit analysis for
the host plant, Haplopappus venetus Blake (Asteraceae). Oecologia
55:185-191.

Oliveira, P.S., A.T. Oliveira-Filho & R. Cintra. 1987. Ant foraging on
ant-inhabited Triplaris (Polygonaceae) in western Brazil: a field
experiment using live termite-baits. Journal of Tropical Ecology
3:193-200.

Peixoto, A.L. & A. Gentry. 1990. Diversidade e composipao floristica
da Mata de Tabuleiro na Reserva Florestal de Linhares (Espirito
Santo, Brasil). Brazilian Journal of Botany 13:19-25.

Romero, G.Q. 2006. Geographic range, habitats and host plants
of bromeliad living jumping spiders (Salticidae). Biotropica 38:
522-530.

Romero, G.Q. & J. Vasconcellos-Neto. 2004a. Spatial distribution
patterns of jumping spiders associated with terrestrial bromeliads.
Biotropica 36:596-601.

Romero, G.Q. & J. Vasconcellos-Neto. 2004b. Beneficial effects of
flower-dwelling predators on their host plant. Ecology 85:446-457.

Romero, G.Q. & J. Vasconcellos-Neto. 2005. The effects of plant
structure on the spatial and microspatial distribution of a
bromeliad-living jumping spider (Salticidae). Journal of Animal
Ecology 74:12-21.

Romero, G.Q., P. Mazzafera, J. Vasconcellos-Neto & P.C.O.
Trivelin. 2006. Bromeliad-living spiders improve host plant
nutrition and growth. Ecology 87:803-808.

Romero, G.Q., A.J. Santos, E.H. Wienskoski & J. Vasconcellos-
Neto. 2007. Association of two new Coryphasia species (Araneae,
Salticidae) with tank-bromeliads in southeastern Brazil: habitats
and patterns of host plant use. Journal of Arachnology
35:181-192.

Romero, G.Q., J.C. Souza & J. Vasconcellos-Neto. 2008. Anti-
herbivore protection by mutualistic spiders and the role of plant
glandular trichomes. Ecology 89:3105-3115.

Santos, A.J. & A.D. Brescovit. 2003. A revision of the Neotropical
species of the lynx spider genus Peucetia Thorell 1869 (Araneae:
Oxyopidae). Insect Systematics and Evolution 34:95-116.

Vasconcellos-Neto, J., G.Q. Romero, A.J. Santos & A.S. Dippenaar-
Schoeman. 2007. Associations of spiders of the genus Peucetia
(Oxyopidae) with plants bearing glandular hairs. Biotropica
39:221-226.

Zar, J.H. 1999. Biostatistical Analysis. Fourth edition. Prentice Hall,
Upper Saddle River, New Jersey.

Manuscript received 14 July 2008, revised 11 April 2009.

2009. The Journal of Arachnology 37:368-370

SHORT COMMUNICATION

Reducing scorpion fluorescence via prolonged exposure to ultraviolet light

Carl T, Kloock; Department of Biology, California State University, Bakersfield, 9001 Stockdale Highway, Bakersfield,
California, 93311, USA. E-mail: ckloock@csub.edu

Abstract. A simple technique is presented for reducing the Ouorescence of living scorpions by prolonged exposure to UV
light. Scorpion’s fluorescence peak can be eliminated by a 1-mo exposure to low intensity UV light. Although the
fluorescence peak returns within 1 wk after removal from UV light exposure, the magnitude remains reduced. This
technique potentially opens up new options for testing a variety of hypotheses about possible functions of scorpion
fluorescence including potential effects on cuticle strength, visual responses, predation, cannibalism, and mating.

Keywords: Spectra, spectroscopy

There is no known function of scorpion lluorescence (Brownell
2001; Kloock 2008). Although it is certainly possible that fluorescence
has no function, it is only by testing and falsifying potential functions
that they can be eliminated from consideration. In order to test
potential functions of scorpion lluorescence, having scorpions with
reduced lluorescence could be a powerful tool. Several methods for
reducing lluorescence exist, but those currently available are
problematic. Kloock (2005) removed fluorescence from preserved
scorpions by applying a varnish that blocked ultra-violet light (UV),
and fluorescence can also be eliminated or reduced either by not
supplying UV light or blocking it with filters. Use of coatings limits
experiments because live, non-fluorescing scorpions cannot be used.
In addition, coating scorpions introduces experimental complications,
given the different chemicals needed to either block or allow
fluorescence and secondary coatings to remove these effects dim the
fluorescence of controls. Detecting lluorescence under natural
conditions, if possible, requires very sensitive visual senses that
scorpions may possess (Kloock 2008), but any dimming could prevent
detection, and thus affect experimental attempts to demonstrate
detection. Eliminating or blocking UV light with filters prevents
fluorescence of living specimens, but also introduces a necessary
experimental complication: effects of fluorescence and UV light
cannot be separated by techniques that simply eliminate UV light.
Given that scorpion eyes (Machan 1968) and their extra-ocular light
sense (Zwicky 1970) are sensitive to both UV light and light near the
fluorescence peak (—500 nm), this is a serious complication.
Modifying scorpions so that they can be exposed to UV light without
fluorescing can remove this problem, although possible behavioral
and physiological side effects of any such manipulation could
introduce new difficulties.

Anecdotal accounts suggest that long-term exposure to ultraviolet
radiation may reduce scorpion fluorescence (Wankhede 2004). This
has not previously been quantitatively demonstrated. I present here
evidence that long-term exposure to UV light significantly reduces
scorpion fluorescence.

Paniroctonus hecki (Gertsch & Allred 1965) (Vaejovidae) were
collected July-Septcmbcr 2008 in Kern County, California (voucher
specimens deposited at the California Academy of Sciences), and
housed in small, foam-plugged plastic vials. Sixteen females were
randomly chosen from this population for the experimental manip-
ulation and placed in small open-topped plastic arenas (13 cm long X
10 cm wide X 7 cm high) with 50 ml of soil collected from the same
site as the scorpions. The soil helped to absorb and retain moisture,
but was not deep enough to allow scorpions to bury themselves to
escape irradiation. Scorpions were exposed 24 hrs/day for 32 days to
two 40 W lluorescent blacklights (GE F40BLB) at a constant distance

of 75 cm. This resulted in scorpions constantly receiving 1 1 pW/cm^
of UV light energy (measured with a Mannix UV-340 light meter:
range = 290-390 nm). Control scorpions were kept without
significant UV light exposure. They were exposed to light from white
fluorescent lights (< 1 pW/cm" UV) daily. Once a week, each scorpion
was fed a single mealworm larva (Tenebrio sp.) and provided water by
spraying the soil surface with a mister. I selected only females to
simplify analysis and because females were more readily available. I
see no reason to expect major differences between genders, but this
should be tested in the future.

Within 2 wk, UV-exposed scorpions exhibited visibly reduced
fluorescence, particularly on the dorsal surfaces. The thinner, more
flexible portions of the exoskeleton (i.e., carapace and mesosoma)
experienced a larger fluorescence reduction than the thicker regions
(pedipalps, metasoma). After 32 days, fluorescence on the dorsal
surfaces was no longer visible, though ventral surfaces and chelicerae
still fluoresced dimly, in a pattern consistent with the effects of
shading. At this point, the 12 surviving UV-exposed scorpions and 12
randomly selected control scorpions were measured with a reflectance
spectrometer using a UV light source to stimulate fluorescence.

Emission spectra measurements were taken with a BW-Tek
BRClllA CCD Spectrometer using BWSpec version 2.24 software
with an integration time of 100 ms and 20 averages per sample. A
single UV LED with peak emission at 390 nm and transmitted via a
fiber-optic cable directly to the reflectance probe supplied excitation
energy. A light-absorbing fabric (Edmund scientific #3060068) served
as a dark reference, and I used the dark subtraction method for all
measurements. The spectra were analyzed in raw form, without
smoothing techniques (e. g., Fourier or running averages) applied to
remove noise from the spectra.

All spectrum measurements occurred inside a light-tight box
enclosing both the scorpion and reflectance probe. A squeeze cage,
as described in Kloock (2008), immobilized scorpions during
measurement. The reflectance probe was lowered until it touched
the carapace at a point just behind the medial eyes. The order of
measurements was randomized to reduce potential bias.

Fig. 1 provides a visual demonstration of the results of the
manipulation. More importantly, spectrometer measurements showed
large effects (Fig. 2, Table 1). The peak seen in each spectrum at
—400 nm (Fig. 2) results from reflection of the UV light source (peak
emission at 390 nm) and is not the result of fluorescence. To avoid
this artifact, all measurements were made over the range of 450-
700 nm.

Thirty-two days of UV exposure produced a significant reduction
in peak power (Table 1; t-test assuming unequal variance, / = 8.88,
P = 2.39 X 10“^ clf= 12) and a significant increase in the wavelength

368

KLOOCK— REDUCING SCORPION FLUORESCENCE

369

Figure 1. — Photograph (grayscale) of two scorpions under black-
light against a non-fluorescent white background. The scorpion on
the left received prolonged exposure to UV light, the scorpion on the
right did not. Preserved specimens were photographed due to the long
exposure times required under these lighting conditions. Photo
courtesy of Jason-Marc Mohamed. Used with permission.

of peak power (Mest assuming unequal variance, t = 4.48, P =
0.0007, df = 12). The emission spectra for the UV-exposed scorpions
show no clear peaks (Fig. 2). An F-test for variance ratios (Zar 1996)
shows that the variance in the peak wavelength of the UV-exposed
scorpions far exceeded the variance for control scorpions (UV-
exposed scorpions / = 655, control scorpions / = 19.0, F = 34.5, P
= 6.46*10""'^). This and the wide range of values (Table 1) indicate
that the wavelength of peak power was essentially random for the
UV-exposed scorpions.

After measurement, both UV exposed and control scorpions were
placed in covered containers with soil and a shelter (a fragment of
terra cotta pot) and maintained on a 13 hours white light: 1! hours
dark cycle, with feeding and watering schedules maintained as above.
One week later, spectra were remeasured. Table 1 shows that in one
week without UV exposure, significant recovery of fluorescence had
occurred, with consistent peaks again evident in the spectra and no
difference between groups in wavelength of peak power (/-test
assuming unequal variance, t = 1.32, P = 0.204, df = 12). Although
the relative intensity of the fluorescence was still significantly reduced
compared with controls (/-test assuming unequal variance, / = 4.60, P
= 6.06 X df = 12.), fluorescence intensity increased significantly
within the UV-exposed group over the one-week recovery period (/-
test assuming unequal variance, / = 3.56, P = 1.56 X 10““*, df = 12).
Significant differences in the variance between controls and UV-
exposed scorpions still existed for peak intensity (F = 35.2, P = 5.79
X 10“'^) and for peak wavelength {F = 3.28, P = 0.030). Preserved
UV-reduced specimens showed no signs of recovery, indicating that
active metabolic processes are responsible for recovery. Determining
just what those processes are will require further study.

Photobleaching, the loss of fluorescence due to prolonged exposure
to excitation wavelengths, is commonly encountered in fluorescence
microscopy, which focuses on methods of preventing it (Deschenes &
Vanden Bout 2002). Photobleaching is probably caused by a variety

Wavelength (nm)

Figure 2. — Representative spectra in the visible range (400-
700 nm) of a control scorpion (gray line) and a scorpion after 32
days of exposure to 1 1 pW/cm^ UV light (black line). The intensity
peak in both spectra as they approach 400 nm is caused by reflection
from the light source, an UV LED with peak emission at 390 nm.

of different mechanisms (Georgakoudi et al. 1997), so more detailed
study will be needed to determine a mechanism of the photobleaching
observed here. Two molecules responsible for scorpion fluorescence
have been identified: B-carboline (Stachel et al. 1999) and 4-methyl-7-
hydroxycoumarin (Frost et al. 2001), which may aid future
investigations into this phenomenon. Similarly, the mechanism of
recovery has not yet been investigated, and possible side effects of the
treatment, including the potential of retinal or other tissue damage
and effects on behavior of long-term exposure, need to be investigated
and controlled in any future experiments using this technique.

With this important caveat, the ability to reduce fluorescence
potentially opens up a broad variety of new experiments. For
example, Camp & Gaffin (1999) and Blass & Gaffm (2008) suggest
that fluorescence may function as a light amplifier, or aid in detecting
UV light. If true, then scorpions with reduced fluorescence should
display an altered light avoidance response. Tests can also be designed
to test ecological hypotheses of function (summarized in Kloock
2008). For example, experiments could be designed to test whether
scorpions with reduced fluorescence experience different levels of
predation, cannibalism, prey capture or mating success than
fluorescent scorpions under different lighting conditions in both
natural and laboratory settings.

It is also possible that fluorescence is a byproduct of a molecule
whose primary function is unrelated to fluorescence itself For
example, Stachel et al. (1999) suggested that the fluorescent molecule
functions in sclerotization. To test this hypothesis, experiments can be
designed to test the effects of fluorescence reduction on cuticle
strength. Similarly, the fluorescent molecule may function in reducing
water loss (Lourenqo & Cloudsley-Thompson 1996), and experiments
could be designed to determine the effect of fluorescence reduction on
the rate of water loss.

Caution must be exercised in experimental design because we have,
as yet, no information on side effects of the technique of fluorescence

Table 1. — Data from fluorescence spectra of control and UV-exposed scorpions after 1 mo of exposure, and repeated after 1 wk of recovery.
n = 12 for each treatment. Peak values determined over the range 450-700 nm.

Measurement

Treatment

Mean peak relative intensity (SE)

Mean wavelength of peak
intensity, nm (SE)

Range: wavelength of peak
intensity, nm

After exposure

Control

608 (120)

501 (1.26)

496-511

UV-exposed

72.1 (10.2)

535 (7.39)

497-576

After recovery

Control

482 (77.2)

501 (2.16)

492-519

UV-exposed

122 (13.0)

508 (3.91)

490-533

370

THE JOURNAL OF ARACHNOLOGY

reduction itself on scorpion behavior or physiology, which could
complicate future experiments. The challenge of experimental design
using this technique will be to adequately control for potential side
effects of lluorescence reduction. One way to accomplish this will be
to cross UV presence and lluorescence reduction: if fluorescence
reduction has an effect unrelated to lluorescence, similar differences
should be observed between fluorescence-reduced and control
scorpions regardless of the presence of UV light. Of course, the
details of any experiment may require more complicated controls to
be developed. Provided adequate controls are used, this technique
makes possible future experiments designed to determine whether or
not scorpion tluorescence (or the fluorescent molecules) serves specific
functions. Even if fluorescence serves no function, such experiments
can enhance our understanding of scorpion ecology, physiology, and
behavior.

ACKNOWLEDGMENTS

A CSUB University Research Council Grant supplied funding. J.
Mohamed provided the photograph and Dr. P. Mickler allowed use
of his photographic equipment. I thank the reviewers for their
comments, which improved the manuscript.

LITERATURE CITED

Blass, G.R.C. & D.D. Gaffin. 2008. Light wavelength biases of
scorpions. Animal Behaviour 76:365-373.

Brownell, P. 2001. Sensory ecology and orientational behaviors.
Pp. 159-183. In Scorpion Biology and Research. (P. Brownell &
G.A. Polis, eds.). Oxford University Press, Oxford, UK.

Camp, E.A. & D.D. Gaffin. 1999. Escape behavior mediated by
negative phototaxis in the scorpion Paniroctomis utahensis
(Scorpiones, Vaejovidae). Journal of Arachnology 27:679-674.
Deschenes, L.A. & D.A. Vanden Bout. 2002. Single molecule
photobleaching: increasing photon yield and survival time through
suppression of two-step photolysis. Chemical Physical Letters
365:387-395.

Frost, L.M., D.R. Butler, D. O’Dell & V. Fet. 2001. A coumarin as a
fluorescent compound in scorpion cuticle. Pp. 365-368. In
Scorpions 2001: in Memoriam, Gary A. Polis. (V. Fet & P.A.
Seldon, eds.). British Arachnological Society, Burnham Beeches,
Buckinghamshire, UK.

Gertsch, W.J. & D.M. Allred. 1965. Scorpions of the Nevada test site.
Brigham Young University Science Bulletin Biological Series
6:1-15.

Georgakoudi, L, M.G. Nichols & T.H. Foster. 1997. The mechanism
of Photofrins® photobleaching and its consequences for photody-
namic dosimetry. Photochemistry and Photobiology 65:135-144.

Kloock, C.T. 2005. Aerial insects avoid fluorescing scorpions.
Euscorpius 21:1-7.

Kloock, C.T. 2008. A comparison of fluorescence in two sympatric
scorpion species. Journal of Photochemistry and Photobiology B:
Biology 91:132-136.

Lourenqo, W.R. & J.L. Cloudsley-Thompson. 1996. The evolutionary
significance of colour, colour patterns and fluorescence in
scorpions. Revue Suisse de Zoologie, Vol., hor serie II: 449^58.

Machan, L. 1968. Spectral sensitivity of scorpion eyes and the
possible role of shielding pigment effect. Journal of Experimental
Biology 49:95-105.

Stachel, S.J., S.A. Stockwell & D.A. Van Vranken. 1999. The
fluorescence of scorpions and cataractogenesis. Chemistry and
Biology 6:531-539.

Wankhede, R.A. 2004. Extraction, isolation, identification and
distribution of soluble fluorescent compounds from the cuticle of
scorpion (Hadnirus cirizonensis). M.S. Thesis, Marshall University,
Huntington, West Virginia. 61 pp.

Zar, J.H. 1996. Biostatistical Analysis, 3rd edition. Prentice Hall,
Upper Saddle River, New Jersey. 662 pp.

Zwicky, K.T. 1970. The spectral sensitivity of the tail of Urodacus, a
scorpion. Specialia 15:317.

Manuscript received 15 October 2008, revised I April 2009.

2009. The Journal of Arachnology 37:371-372

SHORT COMMUNICATION

Presence of Vaejovis franckei in epiphytic bromeliads in three temperate forest types

Demetria Mondragon and Gabriel Isaias Cruz Ruiz: Centro Interdisciplinario de Investigacion para el Desarrollo Integral
Regional (CIIDIR) Unidad Oaxaca, Calle Hornos No. 1003. Santa Cruz Xoxocotlan, Oaxaca, Mexico. E-mail:
dmondragon@ipn.mx

Abstract. Reports of scorpions on epiphytic bromeliads in temperate forests are scarce. Here we present some ecological
aspects of this animal-plant interaction in three different types of temperate forests (pine, pine-oak and oak forest) in
Oaxaca, Mexico. From 2005 to 2007, we collected 373 bromeliads belonging to 10 species, and each plant was defoliated in
search of scorpions. We found 35 individuals of Vaejovis franckei Sissom 1989 in 19 bromeliads: 22 specimens in Tillandsia
carlos-hankii with 21% occupancy and an average abundance of 2.1 ± 1.9 individuals/plant; 12 specimens in T. prodigiosa
(10% occupancy, average abundance = 1.6 ± 0.6) and one specimen in T. calothyrsus (3% occupancy, average abundance
= 1 ± 0.0). Pine-oak forest had 29 individuals; pine forest, 4 individuals; and oak forest, 2 individuals. Percentage of
occupancy differed among localities, while average abundance remained the same. Vaejovis franckei preferred T. carlos-
hankii and pine-oak forest, which was correlated with the percentage of occupancy but not with the average abundance.

Keywords: Phytotelmata, Mexico

The presence of scorpions in tank-type bromeliads has been widely
reported (Lucas 1975; Richardson 1999; Santos et al. 2006); however,
most studies of scorpions on bromeliads have taken place in tropical
forests, whereas research on bromeliads in temperate forests is scarce
(Lucas 1975; Ochoa et al. 1993; Sissom 2000).

The present study sought to evaluate different ecological aspects of
scorpions living in tank-bromeliads in three types of temperate
forests. The study was carried out in Santa Catarina Ixtepeji in the
state of Oaxaca, Mexico, located at 17°09'-17°t I'N and 96°36'-
96°39'W. Its climate varies with the altitude, and ranges from
temperate to cold sub-humid with summer rains. The mean annual
temperature and precipitation are 14° C and 1000 mm, respectively
(INEGI 1998). Three sampling sites were selected: Pena Prieta
(2870 m), a pine forest; La Petenera (2547 m), a pine-oak forest; and
El Cerezal (2300 m), an oak forest. From 2005 to 2007, we sampled
373 bromeliads belonging to ten species. Bromeliads were transported
to the laboratory where each plant was defoliated, and each leaf was
carefully inspected. Scorpion specimens were preserved in 70%
alcohol. We used the taxonomic keys of Hoffmann (1931) and
Stockwell (1992) to identify the scorpions. Collected specimens have
been placed in the Coleccion Nacional de Aracnidos in the Institute
de Biologia (IB-UNAM) in Mexico City.

A total of 35 Vaejovis franckei Sissom 1989 was found in 19 of the
373 sampled bromeliads. We only found scorpions on Tillandsia
prodigiosa (Lem) Baker 1889, Tillandsia carlos-hankii Matuda 1973
and Tillandsia calothyrsus Mez 1896. No specimens were found in
Viridantha plumosa (Baker) Espejo 2002 {n = 22), Catopsis
berteroniana (Schult. & Schult.f.) Mez 1896 (n = 17), Tillandsia
macdoiigallii L.B. Smith 1949 {n = 39), T. oaxacana L.B. Smith 1949
(n = 50), T. magnusiana Wittm. 1901 (n = 10), T. bourgaei Baker 1887
(n = 30), or T. violaceae Baker, 1887 (n = 38).

The greatest number of scorpions was found in T. carlos-hankii (24
individuals), followed by T. prodigiosa (10 individuals) and T.
calothyrsus (1 individual) (see Table 1). The percentage of occupancy
differed among species (X^2 ~ 7.60, P = 0.022, n = 167), whereas the
average abundance of scorpions per plant was the same ( T. prodigiosa
= 1.6 ± 0.6, T. carlos-hankii = 2.1 ± 1.9 and T. calothyrsus = 1.0,
ANOVA, F2,i7 = 1.03, P = 0.3).

Pine-oak forest (La Petenera) produced the greatest abundance of
V. franckei, with 29 individuals, followed by pine forest (Pena Prieta)
with 4 individuals and oak forest (El Cerezal) with 2 individuals. The

percentage of occupancy differed among sampling localities (8% in
Pena Prieta, 24% in Petenera and 3% in El Cerezal; X~2 = 14.88, P =
0.006, n = 167), while the average abundance of scorpions per
bromeliad did not differ significantly among sites (Pena Prieta = 1.3
± 0.6, Petenera = 2.1 ± 1.3 and Cerezal 1.0; ANOVA, = 1.1, P
= 0.4). The average abundance of scorpions by bromeliad (r = 0.18,
P = 0.9) was not correlated with the size of the bromeliad, although
the percentage of occupancy was significantly related to it (Kendall
Tau = 0.154, P = 0.018, n = 107)

Most studies of arthropods living inside bromeliads do not show
specificity for particular bromeliad species (Richardson 1999; Ospina-
Bautista et al. 2004; Liria 2007), even though some arthropods show a
strong preference for certain species of bromeliads (Quevedo &
Vasconcellos-Neto 2005; Osses et al. 2007). In our case, V. franckei
preferred T. carlos-hankii. This preference could be related to the
architecture and/or color of the plant; e.g., T. carlo.s-hankii has green
leaves with a purple base. In contrast, T. prodigio.sa possesses green
pale leaves. These differences could promote microclimatic conditions
that favor the presence of scorpions, although further research is
required to reach such a conclusion.

Although bromeliad size did not correlate with scorpion abun-
dance, there may be a minimum size of bromeliad that can support
scorpions, since scorpions were absent in three small species ( V.
plumosa, T. macdoiigallii and T. magnusiana) that do not form a water
tank. Size limits the amount of water and leaf litter that accumulates
inside these bromeliads, resulting in a decrease in arthropod species
richness and abundance (Benzing et al. 2000), including many species
that scorpions prey upon. Other studies have also shown that small
bromeliads have lower abundance and species richness of arthropods
that scorpions feed on, potentially resulting in lower scorpion
abundances in these bromeliads (Ospina-Bautista et al. 2004; Franco
2008). Scorpions were not found in T. oaxacana either, probably
because of the plant’s small size (17-30 cm in height) and its small
tank, which retains a maximum of 300 ml of water, compared to
1400 ml for T. carlos-hankii (Franco 2008). Bromeliad size might also
directly limit the presence of scorpions, as these arthropods are
relatively large, measuring up to 6 cm in length. Space within the
bromeliads might thus be inadequate to provide a refuge for
scorpions. Accordingly, we observed scorpions only on the larger
bromeliad species (between 50 and 75 cm in height) such as T.
prodigio.sa, T. carlos-hankii and T. calothyrsus.

371

372

THE JOURNAL OF ARACHNOLOGY

Table 1. — Presence of Vaejovis franckei Sissom 1989 on different bromeliad species in three temperate forests in Santa Catarina Ixtepeji. a =
number of sample plants, b = number of plants with scorpions, c = scorpion abundance.

Bromeliad species

Pine forest

Pine-oak

Oak

a

b

c

a

b

c

a

b

c

TiUandsia prodigiosa (Lem) Baker, 1889

0

0

0

40

5

9

37

1

1

TiUandsia carlos-hankii Matuda, 1973

40

3

4

18

9

20

0

0

0

TiUandsia calothyrsus Mez, 1896

0

0

0

0

0

0

32

1

1

Scorpions show high specificity for certain environmental condi-
tions (Hoffmann 1931; Lourengo & Sissom 2000; Florez 2001;
Prendini 2001). Physical conditions in the pine-oak forest site are
probably the most ideal for V. franckei, given that it was most
common in this forest type (29 individuals). The low abundance of V.
franckei in the pine forest may explain why this scorpion was not
found in T. violaceae in this forest type, despite the plant’s suitability
for the establishment of scorpions. Nonetheless, at least one other
species of scorpion, C. Jlavopictus, has been found in this plant species
in a cloud forest in Chiapas (Lucas 1975).

Although tank-type bromeliads appear to be a very attractive
habitat for scorpions because they represent a potential source of
food and refuge, in this study they colonized only a small percentage
of the bromeliad specimens (9%). This result agrees with observations
by Santos et al. (2006), who also reported a low percentage of
occupancy (13%) for Tityiis neglectns Mello-Leitao in four tank-type
bromeliad species. Such low occupancy levels may be due to overall
low scorpion densities, as reported across most types of vegetation
(Bradley & Brody 1984).

ACKNOWLEDGMENTS

This project was supported by SEP-CON AC YT: SEP-2004-C01-
48316. We thank Megan Lore and Sheeva Sreenivasan for help with
the English in this paper.

LITERATURE CITED

Benzing, D.H., H.E. Luther & B. Bennett. 2000. Relationships with
fauna. Pp. 405-462. In Bromeliaceae: Profile of an Adaptive
Radiation. (D.H. Benzing, ed.). Cambridge University Press,
Cambridge, UK.

Bradley, R.A. & A.J. Brody. 1984. Relative abundance of three
vaejovid scorpions across a habitat gradient. Journal of Arachnol-
ogy 11:437-440.

Florez, D.E. 2001. Escorpiones de la familia Buthidae (Chelicerata:

Scorpiones) de Colombia. Biota Colombiana 2:25-30.

Franco, M.A.D. 2008. Diversidad de macroartropodos TiUandsia
carlos-hankii Matuda y TiUandsia oaxacana L. B. Smith en un
bosque de pino-encino de Oaxaca. Master’s Thesis, CIIDIR-
OAXACA, Oaxaca, Mexico. 101 pp.

Hoffmann, C.C. 1931. Los Scorpiones de Mexico. Primera parte:
Diplocentridae, Chactidae, Vejovidae. Annales del Instituto de
Biologia, Mexico 2:291^08.

INEGI (Instituto Nacional de Estadistica, Geografia e Informatica).
1998. Carta del uso de suelo y vegetacion. Escala 1 : 250 000. Oaxaca
E14-9. Carta de precipitacion, escala 1: 250 000. Oaxaca E14-9.

Liria, J. 2007. Fauna fitotelmata en las bromelias Aechmea fendleri
Andre y Hohenbergia stellata Schult de! Parque Nacional San
Esteban, Venezuela. Facultad de Ciencias Biologicas Universidad
Nacional Mayor de San Marcos, Venezuela. Revista Peruana de
Biologia 14:33-38.

Lourenqo, W.R. & W.D. Sissom. 2000. Scorpiones. Pp. 115-135. In
Biodiversidad, taxonomia y biogeografia de artropodos de Mexico:
Hacia una sintesis de su conocimiento. (J. Llorente-Bousquet, E.
Gonzales-Soriano & N. Papavero, eds.). Universidad Nacional
Autonoma de Mexico, Mexico, D.E.

Lucas, K.E. 1975. Tank bromeliads and their macrofauna from cloud
forest of Chiapas, Mexico. M.A. Thesis, San Francisco State
University, California, USA. 82 pp.

Ochoa, M.G., M.C. Lavin, F.C. Ayala & A.J. Perez. 1993.
Arthropods associated with Bromelia hemisphaerica (Bromeliales,
Bromeliaceae) in Morelos, Mexico. Florida Entomologist
76:616-621.

Osses, F., E.G. Martins & G. Machado. 2007. Oviposition site
selection by the bromeliad-dweller harvestman Boiirguyia hanuita
(Arachnida: Opiliones). Journal of Ethology 26:233-241.

Ospina-Bautista, F., J.V. Estevez-Varon, J. Betacur & E. Realpe-
Rebolledo. 2004. Estructura y composicion de la comunidad de
macro invertebrados acuaticos asociados a TiUandsia turneri Baker
(Bromeliaceae) en un bosque Alto Andino Colombiano. Acta
Zoologica Mexicana (n.s.) 20:153-166.

Prendini, L. 2001. Further additions to the scorpion fauna of
Trinidad and Tobago. Journal of Arachnology 29:173-188.

Quevedo, G. & J. Vasconcellos-Neto. 2005. Spatial distribution and
microhabitat preference of Psecas cliapoda (Peckham & Peckham)
(Araneae, Salticidae). Journal of Arachnology 33:124-134.

Richardson, B.A. 1999. The bromeliad microcosm and the assessment
of fauna diversity on a Neotropical Forest. Biotropica 31:321-336.

Santos, R.L., E.A. de Almeida, M.G. Almeida & M.C. Serra. 2006.
Biogeography of bromeliad-dwelling scorpion TUyus neglectns
Mello-Leitao (Buthidae) in Rio Grande does Norte, Brazil. Journal
of Bromeliad Society 56:201-207.

Sissom, W.D. 2000. Family Vaejovidae Thorell, 1876. Pp. 503-553. In
Catalog of the Scorpions of the World (1758-1998). (V. Fet, W.D.
Sissom, G. Lowe & M.E. Braunwalde, eds.). New York
Entomological Society, New York.

Stockwell, S.A. 1992. Systematic observations on North American
Scorpionida with a key and checklist of the families and genera.
Journal of Medical Entomology 29:407^22.

Manuscript received 10 October 2008, revised 5 February 2009.

2009. The Journal of Arachnology 37:373-374

SHORT COMMUNICATION

Taxonomic notes on the genus MicmfiMstata (Araneae: Filistatidae), with a description of a new species

from Turkmenistan

Sergei L. Zonsteln: Department of Zoology, The George S. Wise Faculty of Life Sciences, Tel Aviv University,

Tel Aviv 69978, Israel. E-mail: znn@post.tau.ac.il

Abstract. MicmfiMstata ovckinaikovi new species (Araneae: Filistatidae), the second member of this formerly monotypic
Central Asian genus, is described from Kyzyi-Dzhar Ravine, Southern Turkmenistan. The genus is redescribed and
referred to the Filistatidae insertae sedis.

Keywords: spider, Filistatidae, taxonomy, Central Asia

The genus Microfilistata was based on a single incomplete male
specimen from Uzbekistan (Zonstein 1990). According to Gray
(1995), palpal characters of the genus show some resemblance to
those of the filistatine genera but relationships of Microfilistata
remain unresolved. The occurrence of the second congener in
Turkmenistan, represented also by females, seems to be helpful to
provide the genus with a complete description. The type series,
including the holotype of this new species, is kept in Department of
Zoology, Tel-Aviv University, Israel (TAUI). Measurements are
expressed in millimeters, except eye diameters/interdistances shown as
the ratio of microscope scale units at lOOx magnification. Abbrevi-
ations: AME, ALE, PLE, PME = anterior median and lateral,
posterior lateral and median eyes; ALS, PMS, PLS = anterior lateral,
posterior median and lateral spinnerets, respectively.

Microfilistata Zonstein 1990

Microfilistata Zonstein 1990:51; Gray 1995:80.

Type species. — Microfilistata tyshchenkoi Zonstein 1990 by original
designation and monotypy.

Diagnosis. — Males belonging to Microfilistata can be distinguished
from other male filistatids by their enclosed palp. Conspecific females
are characterized by unusually (for female filistatids) long and slender
palps and legs covered with a reduced number of long setae.

Redescription. — Small pale-colored filistatids with body length of
1.6-1. 7 mm in males and 2. 5-2.7 mm in females; legs and abdomen
without pattern. Carapace domed, broad-oval, wide-rounded anterior-
ly. Thoracic fovea absent. Clypeus short and steeply inclined with few
stout erect bristes. Eye tubercle low. ALE > PLE ^ PME > AME.
Chelicerae small, downward-directed; cheliceral furrow and fang very
short; prolateral lamina small but distinct. Sternum subcircular, sigillae
not evident (Figs. 3, 6). Labium slightly wider than long. Maxillae
trapezoidal. Pedipalps and legs long and slender, both in males and
females. Male palpal tibia long, cylindrical. Cymbium long, with apical
extension entirely covering the tegulum. Embolus sharpened, more or
less curved. Leg formula: 1423. Sexual dimorphism in leg length weakly
developed. Legs covered with rather sparse, long setae. All femora with
one weak dorsoproximal spine. Leg tarsi ascopulate, long and entire,
without pseudosegmentation. Short calamistrum represented by a few
curved and flattened setae on raised keel. Paired tarsal claws narrow
and curved, with a single row of few weak teeth. Unpaired claw weakly
curved, edentate. Female spermathecae divided. Spinneret group small
and located closer to abdomen tip. Cribellum small, bipartite,
trapezioidal. ALS and PLS with thickened setae, PMS with one
probably paracribelar gland spigot (PS).

Notes, — Like the filistatine genera, Microfilistata possesses a short
and narrow calamistrum confined to the metatarsal crest in females

(shown as a filistatine synapomorphy by Ramirez & Grismado, 1997)
and a long cylindrical cymbium with a highly coiled ejaculatory duct in
males. On the other hand, the genus resembles in some aspects the
members of the Prithinae: in both cases the thoracic fovea appears to be
undeveloped, only one PS is present (indicated by Ramirez & Grimaldo,
op. cit., as a prittine synapomorphy), and male tarsi are entire, not
pseudosegmented, cracked or bent; although at least some of the
mentioned characters in Microfilistata could be explained rather by a
miniaturize size of the congeners. Hence, the characters of Microfilistata
are found to be doubtful, and the genus is referred to the Filistatidae
incertae sedis until a further knowledge on Asian filistatids is attained.

Microfilistata ovcMimikovi new species
(Figs. 1-8)

Types. — Male holotype and 3 female paratypes from Kyzyl-Dzhar
Ravine, 580-620 m above sea level, SE border of Badkhyz Plateau,
Turkmenistan (35°48'N, 61°53'E), 11 April 1993, S.V. Ovchinnikov
(TAUI).

Etymology. — The specific name is a patronym in honor of the late
Central Asian arachnologist Mr. Sergei Ovchinnikov, the collector of
the species and a good friend of mine.

Diagnosis. — Differs from M. tyshchenkoi by a different eye ratio
(AME:ALE 1:2 vs. 1:3 in the latter species) and by having a more
thin, tapering and spirally twisted embolus. Any structures resembling
palpal conductor absent.

Description. — Male (holotype): Total length 1.61; whole spider pale
yellowish-brown save for the dark brown eye tubercle and prolateral
lamina, and brownish setae, chelicerae and tarsal claws; abdomen light
yellowish-gray. Carapace (Fig. 1): 0.68 long, 0.59 wide. Eye tubercle
(Fig. 2) low. Ratio of AME, ALE, PLE, PME: 5, 10, 7, 6. Interdistances:
AME-AME 2, ALE-AME 1, ALE-PLE 1, PLE-PME 1, PME-PME
3. Labium and sternum as shown on Fig. 3. Measurements (length):
palp: femur 0.66, patella 0.16, tibia 0.57, cymbium 0.28; legs IM:
femora: 1.22, 0.91, 0.86, 1.04; patellae: 0.25, 0.23, 0.22, 0.23; tibiae: 1.24,
0.77, 0.62, 1.17; metatarsi: 1.12, 0.83, 0.78, 1.07; tarsi: 0.70, 0.52, 0.45,

0. 53. Palps lack spines. Leg spination: femora: I dl-0-0, p 0-0-1, r 0-0-

1, II-IV dl-0-0; tibiae: I-II dl-0-0, p 0-0-1, v 0-1-0, r 0-0-1; III-IV
dl-0-0, p 0-0-1, r 0-0-1; metatarsi: I p 0-1-1, r 0-1-1; other segments
lack spines. Palpi as shown on Figs. 3, 4. Paired tarsal claws with 3-4
weak teeth, unpaired claw bare.

Female (paratype): as in male, except as noted. Total length 2.70.
Carapace (Fig. 5): 0.94 long, 0.84 wide. Ratio of AME, ALE, PLE,
PME: 5.5, 10, 9, 8. Interdistances: AME-AME 2, ALE-AME 2, ALE-
PLE 1, PLE-PME 1, PME-PME 3. Measurements (length): palp: femur
0.65, patella 0.22, tibia 0.48, cymbium 0.53; legs IM: femora: 1.17, 0.95,
0.91, 1.02; patellae: 0.33, 0.32, 0.23, 0.33; tibiae: 1.23, 0.86, 0.79, 1.04;

373

374

THE JOURNAL OF ARACHNOLOGY

Figures 1-8. — Microfilistata ovchhmikovi sp. n., male (1^) and female (5-8). 1, 5. Dorsal view of carapace; 2. Lateral view of carapace,
clielicera and palp; 3, 6. Ventral view of chelicerae, labium, sternum and maxillae; 4. Lateral view of male palpus; 7. Lateral view of calamistrum;
8. Ventral view of spermathecae. Scale for 1-3, 5-7 = 0.25 mm; for 4 and 8 = 0.1 mm.

metatarsi; 1.10,0.81,0.76, 1.00; tarsi: 0.74, 0.53, 0.48, 0.51. Femora I-IV
with one basodorsal spine. Labium and sternum, calamistrum and
spermathecae are as shown in Figs. 6, 7, and 8, respectively.

Distribution and habitat. — Known only from the type locality.
Spiders inhabit cavities and cracks in rock cliffs and escarpments of
the ravine.

ACKNOWLEDGMENTS

Dr. Victor Fet (Marshall University, Huntingon, West Virginia,
USA) kindly checked the manuscript. I am also thankful to the
anonymous reviewers for their valuable comments. The study was
completed with financial help generously provided by the Ministry of
Absorption, Israel.

LITERATURE CITED

Gray, M.R. 1995. Morphology and relationships within the spider
family Filistatidae (Araneae: Araneomorphae). Records of the
Western Australian Museum, Supplement 52:79-89.

Ramirez, M.G. & C.J. Grismado. 1997. A review of the spider family
Filistatidae in Argentina (Arachnida, Araneae), with a cladistic re-
analysis of filistatid genera. Entomologica Scandinavica 28:319-349.

Zonstein, S.L. 1990. A synopsis of species of the spider family
Filistatidae (Aranei) of the USSR with description of a new genus
and a new species from Western Tien-Shan. Zoologicheskiy
Zhurnal 69 (10);50-53 [in Russian, English summary].

Manuscript received 28 July 2008, revised II July 2009.

2009. The Journal of Arachnology 37:375-378

SHORT COMMUNICATION

Evidence for multiple paternity in broods of the green lynx spider Peucetia viridans (Araneae: Oxyopidae)

Martin G. Ramirez', Elizabeth C. Wight', Victoria A. Chirikian', Evelyn S. Escobedo^, Lauren K. Quezada^ Antu

Schamberger^, Jodi A. Kagihara' and Carolyne L. Hoey'; 'Department of Biology, Loyola Marymount University, Los
Angeles, California 90045, USA. E-mail; mramirez@lmu.edu; ^14815 Fonthill Avenue, Hawthorne, California 90250,
USA; '’Department of Biology, Saint Joseph’s University, Philadelphia, Pennsylvania 19131, USA; "TCC Salud Family
Health Center, 5359 West Fullerton Avenue, Chicago, Illinois 60639, USA

Abstract. In the green lynx spider Peucetia viridans (Hentz 1832), the two openings of a mated female’s epigynum are
often sealed by copulatory plugs, sometimes with the two-pronged distal portion of the paracybium of a male palpus
inserted in each opening and embedded in the plugs. The presence of copulatory plugs and paracymbia may prevent further
mating by the female. However, not all mated females exhibit these structures, perhaps allowing some P. viridans females to
mate with more than one male, despite the assertion of Whitcomb & Eason (1965) that females only mate once. We
investigated this possibility by surveying the extent of multiple paternity in field-collected P. viridans broods from southern
California. For adult females and their egg sacs, we determined the aspartate aminotransferase genotype for each mother
and her spiderlings using allozyme electrophoresis in order to assess whether the progeny data best fit with a single male as
the father. Two broods exhibited clear evidence of multiple paternity, verifying that multiple mating by females is possible
in this species. Although most mothers of single paternity broods had one or both epigynal orifices blocked, some had no
blockage at all, while the two mothers of multiple paternity broods had some kind of blockage to one or both orifices,
suggesting that neither plugs nor inserted paracymbial processes are associated with a reduction in female remating.

Keywords: Multiple mating, polyandry, molecular marker, copulatory plug, paracymbial process

The green lynx spider Peucetia viridans (Hentz 1832) is the largest
and commonest member of the family Oxyopidae, with a distribution
throughout the southern United States, Mexico, and Central America
(Brady 1964). It is a cursorial hunter that forages on prey commonly
found on plants (Arango et al. 2000). Although little studied up to
1960, P. viridans has been the sole or partial focus of at least 25 papers
since then, making it one of the best-characterized hunting spiders in
North America.

While much is now known about the reproductive biology of P.
viridans, one question that remains unresolved is whether adult
females ever remate in the wild. This is especially significant given that
P. viridans and P. longipalpis F.O. Pickard-Cambridge 1902 are the
only oxyopids known to produce copulatory plugs (Suhm et al. 1996),
structures which are commonly thought to delay and/or reduce the
probability of female remating (Eberhard 1996). Brady (1964) was the
first to note the presence of plugs in P. viridans females, as he found
that the two openings of a mated female’s epigynum were usually
plugged with a hard, black material, often with the two-pronged distal
portion of the paracybium of a male palpus inserted in each opening
and embedded in the material. Brady stated that the black material
must be deposited during or immediately after insemination, a
suggestion possibly corroborated by Whitcomb & Eason’s (1965)
observation during a laboratory study of a large drop of shiny liquid
on the epigynum of a female immediately following copulation that
later disappeared.

Brady (1964) reasoned that the plugging of the female epigynum
and the loss of the male paracymbial process should prevent further
matings by both female and male (although since males possess two
palpi, an individual male could potentially mate twice). However, in
their study of mating behavior in P. viridans, Whitcomb & Eason
(1965) found that each mating episode involves numerous copulations
between female and male with both palps being inserted alternately
into the epigynal openings. They also found that males mated freely
on successive days, with one male having mated with three different
females over consecutive days. In contrast, an individual female

would only mate with one male and would actively reject subsequent
male suitors. Of course, if the copulatory plug in P. viridans is indeed
a device for impeding access to subsequent males as suggested by
Brady (1964), perhaps assisted in this role by broken-off male
paracymbia, why it is necessary if females never remate is unclear.
However, since females of many species may be more reluctant to
remate in captivity than in nature (Eberhard 1996), Whitcomb &
Eason’s (1965) assertion that female P. viridans mate only once may
not be universally true, as it was based on laboratory observations.

Whitcomb & Eason (1965) also reported on the frequency of the
copulatory plug, as did Exline & Whitcomb (1965), who also provided
data on the frequency of inserted paracymbia. Whitcomb & Eason
(1965) found plugs in all mated females they examined, but not in any
virgin females. In contrast, Exline & Whitcomb (1965) stated that not
all mated females exhibit this covering, noting that it can be easily
removed and is probably sometimes lost during egg-laying. They
found that in a sample of approximately 20 mated females, 10 had at
least one male paracybium embedded in the plug. Among the
remaining 10, the plug was missing altogether or did not contain a
paracybium. Like Exline & Whitcomb (1965), we have found that the
presence of plugs and paracymbia in the epigyna of mated P. viridans
females is variable in both laboratory-mated and wild females, with
almost half of a field-collected sample of females with egg sacs from
southern California (n = 54, 2004) having neither plugs nor
paracymbial processes in their epigynal orifices (unpublished data).
Thus, since plugs and paracymbia are often absent as obstacles in the
epigyna of mated P. viridans females, some females may be physically
capable of mating with more than one male. Given this possibility, the
purpose of this study was to search for cases of multiple paternity in
field-collected P. viridans broods using a genetic approach, since such
broods would result from individual females having had more than
one male sexual partner.

From October through December 2007, we collected 29 adult P.
viridans females with their respective egg sacs from six sites in
southern California (population abbreviation and sample size are

375

376

THE JOURNAL OF ARACHNOLOGY

indicated in parentheses): Los Angeles Co. — Kenneth Hahn State
Recreation Area (HSR, 18), Yvonne Burke Sports Complex (HBF,
2), Ernest Debs Regional Park (DEB, 3), Robert Bernard Biological
Field Station, Claremont (BPS, 1); San Diego Co. — Crest Canyon
Preserve, Del Mar (CC, 4), Carmel Valley Road, Del Mar (CVR, 1).
In the laboratory, we assigned adult females and their respective egg
sacs unique identification numbers. Each female was microscopically
examined for the presence of epigynal plugs, their condition noted,
and any retained male paracymbial processes were recorded. Females
were then frozen at -85° C pending genetic analysis. The egg sacs
were maintained separately in small plastic tumblers until spiderlings
had emerged from each sac, typically within 2-4 weeks of collection.
Up to 60 randomly chosen spiderlings from each brood were then
placed individually into 1.5 ml microcentrifuge tubes marked to
match the identification number of their respective mothers. These
brood samples were also then frozen at -85° C.

We used allozyme electrophoresis as our molecular paternity
assessment technique, given its cost-effectiveness for large samples.
Procedures for horizontal starch gel electrophoresis generally
followed Ramirez (1990) and used gels that were 12.5% starch
(StarchArt). We homogenized individual spiderlings directly in their
1.5 ml microcentrifuge tubes along with 20 pi of deionized water
using a hand-held microcentrifuge tube pestle. This homogenate
material was centrifuged at 484 G’s at 0-4° C for 5 min to separate
extracted proteins from cellular debris and was then frozen at -85° C
until needed for gel loading. The adult females were homogenized
individually in 15 ml Corning centrifuge tubes with an approximately
equal volume of deionized water using a motorized grinding pestle.
These homogenates were centrifuged at 12,100 G’s at 0-4° C for
1 5 minutes. Following centrifugation, the relatively greater volume of
liquid supernatant for each female was transferred into multiple
0.5 ml microcentrifuge tubes, which each contained sufficient
material for one gel run; these tubes were then frozen at -85° C
pending gel loading. Since our study involved destructive processing
of the specimens, no vouchers were retained.

During a related study, we found that five loci are regularly
polymorphic in P. viridcms populations (aspartate aminotransferase,
AAT-1,-2, E.C. 2. 6. 1.1; glucosephosphate isomerase, GPI, E.C.
5.3. 1.9; lactate dehydrogenase, LDH, E.C. 1.1.1.27; phosphogluco-
mutase, PGM, E.C. 2.7.5. 1) (Commission on Biochemical Nomen-
clature 1979). Unfortunately, three of these loci (AAT-2, GPI, PGM)
presented problems of poor resolution of multiple alleles and/or
inconsistent staining during test gel runs, and so were not used. The
two remaining loci (AAT-1, LDH) are both diallelic systems that
resolve well in adults, but since LDH could not be consistently scored
in spiderlings during preliminary testing, probably due to a low level
of enzyme concentration in their homogenates, only AAT-1 (a
cationic locus) was used in this study. The recipe for AAT was based
on Manchenko (1994), and it was resolved using the Continuous Tris-
Citrate 1 buffer system of Selander et al. (1971). Females and their
respective brood samples were run side-by-side on the same gels to aid
in recognizing band homologies during genotype assignment.
Agreement between observed genotypic proportions and Hardy-
Weinberg expectations was evaluated for the adult females by
calculation of exact significance probabilities using the BIOSYS-2
(Black 1997) computer program.

To determine whether a single male was unlikely to account for the
genetic diversity of a female’s brood, we used a significant deviation
from a Mendelian ratio among progeny genotypes as our criterion. We
identified the paternal alleles present in each brood by inspection of the
progeny and maternal genotypes and then used this single paternal
genotype along with the maternal genotype to determine an expected
Mendelian progeny distribution. We tested the observed genotype
numbers against the expected numbers using a Chi-square test with
Yates continuity correction for small sample size (YY), as implemented
in the E-Z Stat 1.0.1 (Towner 1999) statistical analysis program.

The frequencies of the A and B alleles at the AAT-1 locus were

O. 862 and 0.138, respectively, for the adult females and genotype
numbers did not differ significantly from Hardy-Weinberg expecta-
tions. The 29 broods of P. viridcms yielded 1337 spiderlings, which
were genotyped at the AAT-1 locus along with their respective
mothers (mean brood sample = 46, range = 15-60). In 17 broods, the
mothers and spiderlings were all of the same genotype (AA), making
it impossible to test for deviations from Mendelian ratios among
progeny genotypes. The 12 remaining broods contained two or three
offspring genotypes and among these, two (HSR-160, HSR-179)
showed significant deviations from expected Mendelian ratios
(Table 1), indicating that a single male was unlikely to account for
the observed ratios in each case. Since multiple mating episodes
involving males of the same genotype cannot be detected using a
single diallelic locus, the frequency of multiple mating (2/12 = 16%)
reported here may be an underestimate, especially given the limited
number of broods analyzed. Nonetheless, we have verified that P.
viridcms females sometimes mate with more than one male, contrary
to Whitcomb & Eason’s (1965) assertion that females do not remate.

Among the 12 females whose broods contained multiple genotypes
suitable for paternity assessment, the presence and state of the
epigynal plug was quite variable, as was the presence of male
paracymbial processes (Table 1), consistent with the findings of
Exline & Whiteomb (1965) and our prior observations in southern
California. Of the 10 females whose broods provided no evidence of
multiple paternity, 7 (CC-17, DEB-330, DEB-343, HSR-141, HSR-
143, HSR-144, HSR-176) possessed a complete copulatory plug in
one or both of their orifices, sometimes accompanied by a
paracymbial process, while the remaining three (CC-15, HSR- 148,
HSR- 177) had neither plugs nor paracymbial processes in their
orifices (Table 1). As for the two females whose brood genotypes
indicated multiple paternity (HSR-160, HSR-179), HSR-160 had
complete plugs in both orifices and HSR-179 had a partial plug in one
orifice and no plug in the other, while neither possessed paracymbia
in their epigynal orifices (Table 1). Thus, although most females
classified as “singly inseminated” had one or both of their orifices
blocked, three had no blockage at all, while both of the multiply
inseminated females had some kind of blockage to one or both
orifices.

In this study, we found clear evidence for the occurrence of multiple
paternity in P. viridcms offspring from southern California. At a
minimum, 16% of the broods that were suitable for paternity
assessment had been multiply sired. The fact that some level of
female remating and multiple paternity is possible in this species
indicates that sperm competition is a component of sexual selection in

P. viridcms. A widespread adaptive response for paternity assurance
given sperm competition is the formation of copulatory plugs (Wigby
& Chapman 2004), so their production in P. viriclans is understand-
able. Moreover, since parts of the male palp often break off within the
female during mating in several groups of spiders (Eberhard 2004;
Huber 2005), where they can act as impediments to sperm transfer by
subsequent males (e.g., Fromhage & Schneider 2006), the paracym-
bial process of P. viridcms males may similarly serve as a copulatory
plug when lodged in a female. However, both copulatory plugs and
palpal structures in the female genital tract are less than 100%
effective at preventing female remating in many spiders (references in
Huber 2005; Schneider et al. 2005), perhaps partly due to female
efforts to counteract male monopolization if they can benefit from
polyandry (Hosken et al. 2009). More generally, plugs of any sort are
not expected to be absolute barriers since selection would then favor
male avoidance of nonvirgin females and plugging would be selected
against (Eberhard 1996). As for P. viridcms, while we cannot know
how many of the plugs and paracymbial processes we observed were
the result of first matings, the fact that both multiply inseminated
females had some kind of blockage to their epigynum and three
females classified as singly inseminated had no blockage at all

RAMIREZ ET AL.— MULTIPLE PATERNITY IN PEUCETIA BROODS

377

Table 1. — Distribution of AAT-1 genotypes among Peucetia virkkms spiderlings from 12 brood samples, along with maternal epigynal
configurations. The expected brood genotype distributions (in parentheses) assume Mendelian inheritance of two alleles (A, B) at the AAT-1
locus and appropriate genotypes for the unknown male parents based on the paternal alleles evident in each brood. The Chi-square test with
Yates correction (Y^) evaluates the hypothesis that a single male inseminated each female; superscript designates females determined to be
multiply inseminated. Scoring of the female epigyna is the same for both the left and right epigynal orifices (LO, RO):0 = copulatory plug
absent; § = complete plug present; % = partial plug present; i or } - male paracymbial process in left or right orifice.

Eemale

genotype

Presumed

male

genotype

Spiderling genotypes

Epigyna

Female

AA

AB

BB

P

LO

RO

CC-15

AB

AA

32

(30)

28

(30)

—

0.150

0.699

0

0

CC-17

AA

AB

21

(20)

19

(20)

—

0.025

0.874

i

DEB-330

AA

AB

16

(15.5)

15

(15.5)

—

0.000

1.000

f

0

DEB-343

AB

AB

18

(15)

28

(30)

14

(15)

0.546

0.761

If

0

HSR-141

AA

AB

34

(29.5)

25

(29.5)

—

1.085

0.298

f

f

HSR- 143

AB

AA

10

(12.5)

15

(12.5)

—

0.641

0.423

i

ft

HSR- 144

AB

AA

33

(30)

27

(30)

—

0.417

0.519

ft

HSR-148

AB

AA

32

(30)

28

(30)

—

0.150

0.699

0

0

HSR- 160^

AB

AB

30

(15)

26

(30)

4

(15)

22.046

<0.001

f

f

HSR-176

AB

AA

25

(24)

23

(24)

—

0.021

0.885

0

ft

HSR-177

AB

AB

3

(4)

10

(8)

3

(4)

0.586

0.746

0

0

HSR-179^

AA

AB

53

(30)

7

(30)

—

33.759

<0.001

i

0

suggests that neither plugs nor inserted paracymbial processes are
associated with a reduction in female remating.

While the frequency of multiple paternity in P. viridans populations
may be greater than that indicated by our limited data set, it is possible
that the frequency may not be considerably greater due in part to
changes in mate availability associated with a seasonal shift toward a
female-biased sex ratio, as documented for a P. viridans population in
Merida, Mexico (Arango et al. 2000). Thus, females reaching
adulthood later in the year at this site presumably had access to fewer
male suitors, a pattern also seen with the orb-weaving spider Nephila
clavipes (Linnaeus 1767) due to a similar sex ratio shift (Higgins 2000).
If a seasonal, female-biased sex ratio shift also occurs in southern
California P. viridans populations, this may be one reason why even
more broods did not exhibit evidence of multiple paternity in this
study, particularly those whose mothers possessed neither plugs nor
paracymbial processes (CC-15, HSR-148, HSR-177, Table 1).

The analysis of mating and parentage in P. viridans would greatly
benefit from the future use of one or more hypervariable molecular
markers applied to females, their spermathecal contents, and broods
(e.g., Simmons et at. 2007) because such an approach would estimate
both female remating rates and patterns of sperm utilization in
natural populations. In addition, since copulatory plugs are often
externally visible when they are present in the epigynum, if females

were inspected for the presence of plugs on a daily basis (e.g., Kim &
Choe 2007) prior to such genetic analyses, it might also improve our
understanding of the frequency, persistence, and influence of this
structure on opportunities for multiple mating by females.

ACKNOWLEDGMENTS

We are grateful to L. Towner for providing a copy of the E-Z Stat
for Windows computer program. For providing access to the Robert
Bernard Biological Field Station, we thank S. Dreher, Claremont
University Consortium. We similarly thank J. Chapman, R. Delgado,
M. Renaker, P. Sun, R. Valdivia and E. Yanez of the Audubon
Center at Debs Park for facilitating collecting at Ernest Debs
Regional Park. Financial support was provided by Loyola Mary-
mount University (William F. McLaughlin Chair in Biology).

LITERATURE CITED

Arango, A.M., V. Rico-Gray & V. Parra-Tabla. 2000. Population
structure, seasonality, and habitat use by the green lynx spider
Peucetia viridans (Oxyopidae) inhabiting Cnidoscolus aconitifolius
(Euphorbiaceae). Journal of Arachnology 28:185-194.

Black, W.C.B., IV. 1997. BIOSYS-2: A Computer Program for the
Analysis of Allelic Variation in Genetics. Colorado State
University, Ft. Collins, Colorado.

378

THE JOURNAL OF ARACHNOLOGY

Brady, A.R. 1964. The lynx spiders of North America, north of
Mexico (Araneae: Oxyopidae). Bulletin of the Museum of
Comparative Zoology 131:429-518.

Commission on Biochemical Nomenclature. 1979. Enzyme Nomen-
clature, 1978. Academic Press, New York.

Eberhard, W.G. 1996. Female Control: Sexual Selection by Cryptic
Female Choice. Princeton University Press, Princeton, New Jersey.

Eberhard, W.G. 2004. Why study spider sex: special traits of spiders
facilitate studies of sperm competition and cryptic female choice.
Journal of Arachnology 32:545-556.

Exline, H. & W.H. Whitcomb. 1965. Clarification of the mating
procedure of Peucetia viridans (Araneida: Oxyopidae) by a
microscopic examination of the epigynal plug. Florida Entomol-
ogist 48:169-171.

Fromhage, L. & J.M. Schneider. 2006. Emasculation to plug up
females: the significance of pedipalp damage in Nephila fenestrata.
Behavioral Ecology 17:353-357.

Higgins, L. 2000. The interaction of season length and development
time alters size at maturity. Oecologia 122:51-59.

Hosken, D.J., P. Stockley, T. Tregenza & N. Wedell. 2009.
Monogamy and the battle of the sexes. Annual Review of
Entomology 54:361-378.

Huber, B.A. 2005. Sexual selection research on spiders: progress and
biases. Biological Reviews 80:363-385.

Kim, K.W. & J.C. Choe. 2007. Sex ratio and approximate date of
fertilization of the subsocial spider Amaurobius ferox Walckenaer
(Araneae: Amaurobiidae). Journal of Ecology and Field Biology
30:277-280.

Manchenko, G.P. 1994. Handbook of Detection of Enzymes on
Electrophoretic Gels. CRC Press, New York.

Ramirez, M.G. 1990. Natural history, population genetics, system-
atics and biogeography of the spider genus Lutica (Araneae:
Zodariidae). Ph.D. dissertation. University of California, Santa
Cruz, California.

Schneider, J.M., L. Fromhage & G. Uhl. 2005. Copulation patterns in
the golden orb-web spider Nephila madagascariensis. Journal of
Ethology 23:51-55.

Selander, R.K., M.H. Smith, S.Y. Yang, W.E. Johnson & J.B.
Gentry. 1971. Biochemical polymorphism and systematics in the
genus Peromyscus. I. Variation in the old-field mouse {Peromyscus
polionotus). Studies in Genetics VI. University of Texas Publication
7103:49-90.

Simmons, L.W., M. Beveridge & W.J. Kennington. 2007. Polyandry
in the wild: temporal changes in female mating frequency and
sperm competition intensity in natural populations of the
tettigoniid Recpiemi verticalis. Molecular Ecology 16:4613^623.

Suhm, M., K. Thaler & G. Alberti. 1996. Glands in the male palpal
organ and the origin of the mating plug in Amaurobius species
(Araneae, Amaurobiidae). Zoologische Anzeiger 234:191-199.

Towner, H. 1999. E-Z Stat for Windows, ver. 1.0.1. Trinity Software,
Inc., Plymouth, New Hampshire.

Whitcomb, W.H. & R. Eason. 1965. The mating behavior of Peucetia
viridans (Araneida: Oxyopidae). Florida Entomologist 48:163-167.

Wigby, S. & T. Chapman. 2004. Sperm competition. Current Biology
14:R100-R103.

Manuscript received 26 January 2009, revised 27 July 2009.

2009. The Journal of Arachnology 37:379-382

SHORT COMMUNICATION

A new approach to examining scorpion peg sensilla: the mineral oil flood technique

Elizabeth D. Knowlton and Douglas D. Gaffin: Department of Zoology, University of Oklahoma, Norman, OK
73019-0235, USA. E-mail; eknowlton@ou.edu

Abstract. All scorpions possess jointed, ventral appendages called pectines. These organs have chemosensory, peg-shaped
sensilla that detect substrate-borne chemicals. Previous physiological studies show that neurons within peg sensilla respond
to an assortment of volatile organic chemical stimulants blown across the sensillar opening. We developed an improved
method of chemical stimulant delivery called the mineral oil flood technique to further investigate the neural circuitry of
scorpion pectines. The new mineral oil flood technique allows us to deliver chemical stimulants directly to individual
sensilla by introducing a polar, liquid substance under non-polar mineral oil. Unlike previous methods of stimulant
delivery, the mineral oil flood technique allows for precise control over the duration of direct contact between a liquid
stimulant of known concentration and a sensillum.

Keywords: Pectines, chemosensory, electrophysiology, stimulant, Scorpiones

Pectines are movable sensory appendages that extend from the mid-
ventral surface of all scorpions (CIoudsley-Thompson 1955). They are
similar in function to the antennae of mandibulate arthropods that
detect airborne chemicals (Kaissling 1987; Itagaki & Hildebrand
1990), except that pectines are ground-directed and respond to
substrate-borne chemicals. When scorpions move in their environ-
ment, the pectines sweep intermittently against the substrate to detect
food (Krapf 1986; Skutelsky 1995) and pheromones (Gaffin &
Brownell 1992, 2001).

The primary sensory elements on pectines are hundreds of minute
structures called peg sensilla, which adorn the ground-facing surface
of each pectinal tooth (Carthy 1966, 1968; Ivanov & Balashov 1979;
Foelix & Miiller-Vorholt 1983). Each peg sensillum has a single slit-
like pore that allows chemical stimulants access to receptor neurons
inside the peg shaft. Approximately 10 sensory neurons innervate
each peg (Foelix & Miiller-Vorholt 1983), and synaptic contacts exist
between neurons within a sensillum (Gaffin & Brownell 1997a).

Peg sensilla respond with different neural activity patterns to
alcohols, aldehydes, ketones, and esters (Gaffin & Brownell 1997b).
The sensilla were stimulated indirectly through puffs of stimuli blown
across entire peg fields (Gaffin & Brownell 1997b) or by static clouds
of volatile organic compounds brought near the peg tips (Gaffin &
Walvoord 2004). Although these methods of stimulus delivery elicit
neural responses, they have limitations. Most importantly, it is
impossible to introduce a stimulant to a single peg sensillum without
stimulating its neighbors. There is also no way, as such, to tell if
stimulating a neighboring sensillum influences the response of the
recorded sensillum. In addition, the concentration of stimulant
reaching the sensillar pore is unknown, and removal of the stimulant
from the peg field is uncontrollable.

To overcome these limitations, we developed an improved method
of chemical stimulus delivery called the mineral oil flood technique,
which uses non-polar mineral oil as a medium for delivering polar,
liquid stimulants to an individual sensillum. In this study, we describe
our new method and compare spontaneous and chemically-induced
neural activity of peg sensilla in air and under oil.

Mature female Paruroctoniis utahensis (Williams, 1968) (Scor-
piones: Vaejovidae) collected from Crane County (31°28'59"N,
102°40'38"W), Texas, in March of 2008 were the animals used for
this study. We individually housed each scorpion in 3.8 1 glass jars
containing approximately 250 ml of sand from the scorpion’s
collection site. The animals were on a 15:9 h L:D cycle and kept in
a room with a steady temperature and relative humidity (22° C, RH

55-60%). We fed each scorpion one early instar cricket biweekly and
watered the sand of individual containers with 5 ml of deionized
water twice a week. At the conclusion of our study, we deposited a
voucher specimen in the Sam Noble Oklahoma Museum of Natural
History (OMNH-16279).

For preparing a scorpion for electrophysiological study, we briefly
anesthetized a live animal by cooling it for two minutes inside a
freezer at —5° C. Then we immobilized the animal with modeling clay
on a microscope slide (7.62 X 2.54 cm), ventral side up to expose the
pectines. We positioned a modified cover glass (5 X 18 mm) with
walls of wax approximately 1 mm high caudal to where the pectines
join the body (Fig. la). A notch in the wax wall allowed access of a
pecten to the chamber (Fig. lb). We secured the pecten spine and
teeth to the cover glass with double-sided adhesive tape and fine
application of less than 5 pi of quick-drying adhesive glue (Instant
Krazy Glue^''^). Next, we applied additional wax to the area where
the pecten spine crossed the edge of the cover glass to complete the
wall of the chamber. We then placed one drop of mineral oil (~5 pi)
over the pecten with a 0.25 ml syringe (Fig. Ic). We secured the
animal onto an adjustable platform and located peg fields with a high-
powered compound microscope equipped with epi-illumination
(Olympus BX50-WI). Lastly, we used an electrolytically-sharpened
tungsten electrode to make extracellular recordings of chemosensitive
neurons within individual sensilla (Gaffin & Brownell 1997b).

To record peg neurons, we digitized electrical activity with an analog
to digital converter (1401 -plus digitizing hardware, CED, Cambridge,
England) and analyzed the record using Spike 2 laboratory software
(CED, Cambridge, England). Impulses (“spikes”) from each sponta-
neously active chemosensitive neuron were identified and separated into
three classes (Al, A2, and B), based on the characteristic spike
waveforms of the impulses from each neuron (Gaffin & Brownell
1997b). We used auto-correlation analysis to determine the purity of
each spike class and cross-correlation analysis to detect synaptic
interactions among spike classes (for details on auto- and cross-
correlation analyses, see Gaffin & Brownell 1997a; Eggermont 1990).

To stimulate peg sensilla chemically, we filled a glass capillary tube
(stimulant pipette) with an approximately 10 pm diameter tip with
95% ethanol (Gaffin & Walvoord 2004). We used a nonmetallic
syringe needle (World Precision Instruments, Inc. MicroFiF*^
MF34G) to transfer the ethanol to the stimulant pipette. We then
placed the stimulant pipette into a glass electrode holder positioned
on a mechanical micromanipulator (Sutter Instrument Corp. 1 140).
For fluid stimulant introduction, we maneuvered the pipette tip with

379

380

THE JOURNAL OF ARACHNOLOGY

Figure 1. — Scorpion pecten configuration, a. The right pecten (outlined) as positioned for electrophysiological examination, b. Close-up view
of the right pecten in a stimulation chamber. A piece of the wax barrier (dashed outline) is removed for pecten placement, c. Left, an expanded
field of view of patches of peg sensilla (ps). Right, a mineral oil overlay of an expanded field of view of a peg field. A microelectrode (e) is inserted
at the base of a single peg, under oil, to record baseline neural activity in the presence or absence of a chemical stimulant (cs).

the micromanipulator so that it touched the pore of the recorded
sensillum. Fig. Ic shows the general configuration of the peg field,
microelectrode, and chemical stimulus delivery device during chem-
ical stimulation.

In general, we were able to record spontaneous neural activity
under oil for extended periods, some longer than six hours. Because
the stability of a recording often depended on the animal’s inability to
move its pecten, we improved the method of adhering the pecten to
the cover glass. The most effective method was careful application of
quick-drying adhesive to the pecten spine and the distal-lateral
surface of each pectinal tooth. The least effective adhesives were
paraffin wax and silicone gel.

Because the spread of mineral oil beyond the pecten and onto the
animal’s body induced pecten movement, we assembled a barrier
between an isolated pecten and the rest of the animal. The most useful
barriers were about one millimeter high, which still allowed easy
access of an electrode and stimulant pipette to the sensillum, while
preventing the spread of mineral oil beyond the pecten. The amount
of mineral oil applied to the pecten also affected the quality of a
preparation. Volumes of oil greater than or equal to 50 pi were not
contained within the barrier. Excessive oil also blurred the field of
view. In contrast, 5 pi of oil actually improved the resolution of the
peg field; we could discern individual peg sensillar shafts, which is
difficult when viewing sensilla in air. In addition, 5 pi of oil remained

within the barrier and provided sufficient overlay for stimulant
introduction.

The presence of mineral oil on peg sensilla did not affect baseline
neural activity. Fig. 2 compares the spontaneous firing pattern of a
peg sensillum in air with that of another peg sensillum under oil.
Cross-correlation analyses reveal the same synaptic interactions in
each record: when spike B fired, it inhibited spikes A1 and A2 for
approximately 0.1 s.

Using the mineral oil flood technique, we confined chemical
stimulation to a single peg sensillum and controlled the onset and
removal of the stimulant. For example, the right panel of Figure 3
shows the introduction and removal of liquid ethanol to a peg
sensillum under oil for durations of one, two, and three seconds.
Stimulations were consecutive and spaced approximately 20 s apart,
which produced receptor adaptation in the third response. In
contrast, the left panel of Fig. 3 shows a prolonged neural recovery
after introduction of a drop of ethanol to a peg sensillum in air. The
time-expanded view of stimulant introduction shows the extent of
record disturbance; recorded electrical activity was inconsistent across
all stimulations.

Our study represents the first account of selective chemical
stimulation of individual scorpion peg sensilla with a known
concentration of aqueous stimulant. Because the presence of mineral
oil over a sensillum did not affect the baseline neural activity, we used

KNOWLTON & GAFFIN— NEW METHOD FOR SCORPION PECTEN RESEARCH

381

Figure 2. — Processed recordings of spontaneous neural activity. Shown are recordings of chemosensitive neurons (B, Al, and A2) from a
sensillum in air (left panel) and a different sensillum under oil (right panel). Each line represents a neural impulse within the parsed 50 s record.
The enlarged figures at the left of each recording indicate the B, Al, and A2 impulse waveforms. Cross-correlograms reveal that the firing of B
(white arrow) inhibits the activity of Al and A2 (black arrow) both in air and under oil.

oil as a medium through which to directly stimulate individual pegs
with 95% ethanol for controlled durations.

One limitation of the mineral oil flood technique is that we can only
use polar liquids as stimulants. Non-polar stimulants would mix with
non-polar mineral oil. Therefore, future studies on peg sensillar
function will use varying concentrations of polar solutions, such as
salts and organic compounds that contain fewer than three carbons as
stimulants.

In forthcoming experiments, we will use the mineral oil technique
to further our understanding of peg sensillar function. This new
method should generate the quantifiable data necessary for compar-
ing sensillar neural responses. For example, we aim to stimulate many
peg sensilla individually to compare response intensities to varying
concentrations of stimulants. This will help us determine if all peg
sensilla are functionally equivalent and if they follow dose-dependent
response patterns. Additionally, no studies to date have tested for
possible peg-to-peg interactions. We plan to test for synaptic
interactions between neurons of neighboring sensilla by stimulating
one peg sensillum while recording electrophysiologically from a
neighboring sensillum. If synaptic interactions extend beyond neurons
of an individual sensillum, we should observe a change in neural

activity of the recorded sensillum as we stimulate its neighbor. Such a
situation would provide evidence of lateral inhibition, which is a form
of peripheral processing seen most commonly in the vertebrate retina.

ACKNOWLEDGMENTS

We thank Drs. Marielle Hoefnagels, Don Wilson, and Ari
Berkowitz for their helpful suggestions and use of equipment. We
also thank Nataliya Popokina for assistance in collecting and
maintaining our animals. Finally, we thank the Life Fund of the
University of Oklahoma Foundation and the OU Zoology Adams
Scholarship Fund for providing support for this work.

LITERATURE CITED

Carthy, J.D. 1966. Fine structure and function of the sensory pegs on
the scorpion pecten. Experientia 22:89-91.

Carthy, J.D. 1968. The pectines of scorpions. Symposium of the
Zoological Society, London 23:251-261.

Cloudsley-Thompson, J.L. 1955. On the function of the pectines of
scorpions. Annals & Magazine of Natural History 8:556-560.
Eggermont, J.J. 1990. The Correlative Brain: Theory and Experiment
in Neural Interaction. Springer Verlag, Berlin.

Ill

qplipi

Figure 3. — Unprocessed recordings of neural activity during chemical stimulation. In air (left panel), fluid ethanol was introduced to a peg
sensillum at three successive occasions (arrows). To the right of each record is a time-expanded view of the exact moment of stimulant contact
with the recorded sensillum. Under oil (right panel), fluid ethanol was reversibly applied to a peg sensillum for 1, 2, or 3 s (horizontal bars).

382

THE JOURNAL OF ARACHNOLOGY

Foelix, R.F. & G. Muller-Vorholt. 1983. The fine structure of
scorpion sensory organs. II. Pecten sensilla. Bulletin of the British
Arachnological Society 6:68-74.

Gaffin, D.D. & P.H. Brownell. 1992. Evidence of chemical signaling
in the sand scorpion, Parwoctomis mesaensis (Scorpionida:Vaejo-
vidae). Ethology 91:59-69.

Gaffin, D.D. & P.H. Brownell. 1997a. Electrophysiological evidence
of synaptic interactions within chemosensory sensilla of scorpion
pectines. Journal of Comparative Physiology A 181:301-307.

Gaffin, D.D. & P.H. Brownell. 1997b. Response properties of
chemosensory peg sensilla on the pectines of scorpions. Journal
of Comparative Physiology A 181:291-300.

Gaffin, D.D. & P.H. Brownell. 2001. Chemosensory behavior and
physiology. Pp. 184-203. In Scorpion Biology and Research. (P.H.
Brownell & G.A. Polls, eds.). Oxford University Press, New York.

Gaffin, D.D. & M.E. Walvoord. 2004. Scorpion peg sensilla: are they
the same or are they different? [Proceedings of the 3'^‘^ Scorpiology
Symposium (American Arachnology Society Meeting, Norman,
OK, 24 June 2004)]. Euscorpius 17:7-15.

Itagaki, H. & J.G. Hildebrand. 1990. Olfactory interneurons in the
brain of the larval sphinx moth Manduca sexta. Journal of
Comparative Physiology A 167:309-320.

Ivanov, V.P. & Y.S. Balashov. 1979. [The structural and functional
organization of the pectine in a scorpion Buthus eupeus Koch
(Scorpiones, Buthidae) studied by electron microscopy.] In Fauna i
ekologiya paukoobraznykh. [The Fauna and Ecology of Arachni-
da.] (Y.S. Balashov, ed.). Trudy Zoologicheskogo Instituta
Akademii Nauk SSSR, Leningrad 85:73-87 (in Russian).

Kaissling, K.-E. 1987. R.H. Wright Lectures on Insect Olfaction.
Simon Fraser University, Burnaby, British Columbia, Canada.

Krapf, D. 1986. Contact chemoreception of prey in hunting scorpions
(Arachnida: Scorpiones). Zoologischer Anzeiger 217:119-129.

Skutelsky, O. 1995. Flexibility in foraging tactics of Buthus occilanus
scorpions as a response to aboveground activity of termites.
Journal of Arachnology 23:46-49.

Manuscript received 30 September 2008, revised 4 May 2009.

2009. The Journal of Arachnology 37:383-387

SHORT COMMUNICATION

New data on Theridion itaHeme, with description of the unknown female

loan Duma: Department of Biology, Faculty of Chemistry-Biology-Geography, West University of Timisoara, Pestalozzi
Street no. 16, Timisoara, Judetul Timis, Romania. Postal Code: 300115. E-mail: ioan.duma@gmail.com

Abstract. I describe the female and redescribe the male of Theridion italiense Wunderlich 1995 from live and alcohol-
preserved material and provide notes on the ecology, distribution, and affiliations with the very similar Theridion uhligi
Martin 1974 and Theridion petraeum L. Koch, 1872.

Keywords: Araneae, Theridiidae, Carpathian Mountains

After recent taxonomists have transferred some species to other
genera (Ko^ak & Kemal 2008, Wunderlich 2008), genus Theridion
contains 47 species in Europe (Platnick, 2009). About a third of these
species are poorly known, ten being described only from females and
four known only from males. All of the poorly-known species have
been discussed only in their original descriptions, so little is known
about their distribution, habitat preferences, biology, and ethology.
In some cases, even the original descriptions are wanting. This article
sheds light on one such poorly known species.

Theridion italiense Wunderlich 1995 was described from a single
male specimen collected in Abruzzo National Park in Italy
(Wunderlich 1995), and other specimens have not been recorded
since. This dearth of observations is probably due to its habitat
preferences: low vegetation (5 to 10 cm above ground level) in rocky
areas, which makes the webs difficult to see and the spiders nearly
inaccessible to collecting with an insect net. However, I have collected
30 individuals from 25 m' of proper habitat in about 2 h,
demonstrating that the species may not be as rare as it has originally
appeared, given its near absence from the European arachnological
literature.

The first specimens from Romania were gathered by hand and
sweeping with an insect net in 2007. The spiders collected in 2008 were
all captured by hand. All the material was preserved in 70% ethanol.
Color of the specimens is described for both live and alcohol-
preserved specimens. All measurements are in millimeters. The
drawings were made with a drawing tube using an 1.0. R. ML-4
microscope and an 1.0. R. stereomicroscope.

The specimens examined in this study are deposited in the following
collections: the National Museum of Natural History “Grigore Antipa”
in Bucharest (MGAB), the Zoological collection of the Faculty of
Chemistry-Biology-Geography in Timisoara (CBGT), the author’s
personal collection (CID), and the Thaler-Knoflach collection (CBK).

TAXONOMY

Family Theridiidae Sundevall 1833
Genus Theridion Walckenaer 1805
Theridion italiense Wunderlich 1995:691-695, figs. 8, 9
(Figs. 1, 2A, B, 3 A, 4A, B, C, D)

Type material. — Holotype male: ITALY: Abruzzo National Park
(Id) collected in July 1994 (leg. Jorg Wunderlich) (in the collection of
Jorg Wunderlich - CJW) - not examined.

Material examined. — ROMANIA: 19 Caras-Severin: Chiacotu
Mic: Berzasca River Basin (44°43'52"N, 22°06'01"E) hand collecting,
1 May 2007, loan Duma leg. (MGAB); 19 Caras-Severin: Baiie
Herculane (44°52'00"N, 22°26'05"E) hand collecting, 10 May 2007,
loan Duma leg. (MGAB); 4 99 Alba: Rimetea (46°27'12"N, 23°
34'46"E) sweep net, 27 May 2007, loan Duma (CID); 25 99, 5 dd 14

May 2008, hand collecting, same location and collector (Id, 399 in
MGAB; Id, 19 in CBK; Id, 19 in IDC; 2dd, 2099in CBGT).

Diagnosis. — Based upon the morphology of the palp and epigy-
num, Theridion italiense closely resembles Theridion petraeum L.
Koch 1872 and Theridion uhligi Martin 1974. Both male and female
T. italiense have dimensions similar to T. uhligi, but are clearly smaller
than T. petraeum. Males’ palps differ in the shape of the median
apophysis (Fig. 3A-C). In females, the copulatory ducts are separate
over their entire length. Using this characteristic feature, females of T.
italiense can be easily differentiated from T. uhligi, whose copulatory
ducts unite before opening in the center of the epigynum. The
difference between female T. italiense and T. petraeum is that in the
latter species the copulatory ducts open in the corners of the epigyne,
not in its center.

The species also differ in their habitat preferences. Theridion italiense
was found at low altitudes (350-1000 m) in limestone mountains.
Theridion petraeum appears to be a species of higher altitudes and T.
uhligi an inhabitant of grassy vegetation in the low plains.

Description. — Male: Dorsal carapace yellowish-brown with nar-
row black median band. On lateral sides of thoracic part, narrow
black lines present as in Theridion uhligi Martin 1974, which largely
disappear in some preserved specimens and thus may be hard to see.
Sternum yellowish with grayish-black lateral margins. Labium
brownish-red in contrast to T. uhligi, on which it is yellowish.
Clypeus yellowish. Abdomen: dorsal part pinkish in live specimens,
with median whitish-pink area bordered by a narrow brown line. In
alcohol, pink disappears quickly and turns to whitish-yellow. Lateral
parts of the abdomen pinkish with small white spots. Epigastric
region large and brown, in contrast to that of T. uhligi, which is
yellowish. Occasional long hairs present on abdomen. Legs: yellowish
with brown annulations on femora, tibia, and tarsus. Annulations of
leg segments may not be visible in some specimens. Leg lengths: leg I:
femur 1.85-2, patella 0.5-0.55, tibia 1.7-1. 8, metatarsus 1.65-1.8,
tarsus 0.7-0.75, total leg length 6.4-6.9; leg II: femur 1.3-1. 4, patella
0.5-0.55, tibia 0.85-0.92, metatarsus 1.0-1.12, tarsus 0.55-0.6, total
leg length 4.2^.6; leg III: femur 0.85-0.95, patella 0.35-0.4, tibia 0.5-
0.55, metatarsus 0.7-0.75, tarsus 0.45-0.5, total leg length 2.85-3.15;
leg IV: femur 1.3-1. 4, patella 0.4-0.45, tibia 0.9-0.95, metatarsus 1.1-
1.17, tarsus 0.5-0.57, total leg length 4.2M.54. Tibial spines: 2:2: 1:2.
Chelicerae yellowish-brown without visible teeth. Basal hump on the
frontal side of the chelicerae visible in lateral view (Fig. 2A). Palp
(Fig. 1): Palp of T. italiense very similar to that of T. uhligi and T.
petraeum, but rounder in ventral view and smaller than in closely
related species. Shape of median apophysis (Fig. 2A) very different
from that of Theridion petraeum, which has stickle shape (Fig. 2C).
Longer than that of Theridion uhligi, with differently shaped base.
Also, on inner face of median apophysis in T. italiense, characteristic
protuberance present visible from slightly prolateral angle as a

383

384

THE JOURNAL OF ARACHNOLOGY

Figure I. — Ventral view of the right palp of TherkUon italieme
male. Scale = 0.1 mm.

second, smaller apex (Fig 1). Somatic features {n = 5); total length
2.9-3. 1 mm. Prosoma 1.0-1.15 mm long and 0.9-0.97 mm wide.

Female (Fig. 3C).- Carapace yellowish with black median band
extending to posterior row of eyes. Median band narrower in thoracic
part of prosoma than in cephalic part, narrower in younger females
and widening with age. This pattern also occurs on the lateral bands.
Clypeus yellowish with a black triangular spot in front of chelicerae.
Labium brownish. Chelicerae yellowish. Sternum yellowish with
black bands on the sides. Width of bands varies with age of female
(narrower in younger specimens and wider in older ones). Abdomen:
dorsally whitish medially, bordered by narrow, sinuous pinkish-
brown line. Lateral parts pinkish with small white spots. Pink color
disappears in specimens preserved in alcohol and becomes whitish-
cream. On ventral part of the opisthosoma, directly in front of
spinnerets, triangular or square shape black mark bordered by short
white lines on each side. Legs: yellowish with annulations on all
segments except coxae. Annulations not as well defined in young
females as in older ones. Leg lengths: leg I: femur 1.55-1.9, patella
().4-0.5, tibia 1.2- 1.5, metatarsus 1.0- 1.3, tarsus 0.65-0.8, total leg
length 4. 8-6.0; leg II: femur 1. 15-1. 4, patella 0. 3-0.4, tibia 0.65-0.8,
metatarsus 0.9-1. 1, tarsus 0.55-0.7, total leg length 3.55-4.4; leg III;
femur 0. 7-1.0, patella 0.25-0.35, tibia 0.45-0.6, metatarsus 0.55-0.7,
tarsus 0.4-0. 5, total leg length 2.35-3.15; leg IV; femur 1.2-1. 5, patella
0.4-0. 5, tibia 0.9- 1.1, metatarsus 1.0- 1.2, tarsus 0.45-0.6, total leg

A

Figure 2. — Chelicerae of TherkUon italiense. A. Lateral view of the
male left chelicera; B. Frontal view of the female chelicerae. Scale =
0.1 mm.

DUMA— NEW DATA ON THERIDION ITALIENSE

385

Figure 3. — Median apophyses. A. Theridion italiense (inner side, Figure 4. — Female Theridion itaUense from Romania. A. Vulva

ectal view); B. Theridion uhligi (external side, mesal view); C. (ventral view); B. Vulva (dorsal view); C. Habitus (dorsal view); D.
Theridion petraeum (external side, mesal view). Epigynum (Scale = 0. 1 mm for A, B, D; 1 mm for C).

386

THE JOURNAL OF ARACHNOLOGY

Figures 5. — Web of Theridion italien.se. A. Frontal view of a 1-day-
old web; B. Lateral view of a web; C. Frontal view of a 1 -month-old web
constructed in captivity in a plastic jar with a simple retreat; D. Ventral
view of a simple retreat placed under a stone with the adult male and
female; E. Ventral view of a retreat constructed under the branch and
leaves of a small plant with the female prepared to lay its cocoon.

length 3.95-4.9. Tibial spines: 2:2: 1:2. Chelicerae (Fig. 2B): two small
teeth with common base on prolateral margin. Basal hump also
present, but less prominent than in males. Epigynum: oval with
sclerotized margins, especially the posterior one (Fig. 3D). Openings
of copulatory ducts approximately in its center very close to each
other. Copulatory ducts completely separate over entire length, not
merging in common duct as in T. uhligi. Spermathecae round with
thin walls. Somatic features (n = 31): total length 2.4-3.0 mm.
Prosoma 0.9-1. 2 mm long and 0.7-1.025 mm wide.

Ecological notes. — All specimens from Romania were gathered in
May and were already mature. The majority of females were found
with plugged epigyna, suggesting that first matings take place early in
the spring. However, since the holotype male was collected in July in
Italy, I infer that the species may reproduce throughout the warmer
months of the year. The results of the collecting trip of 14 May 2008
suggests a female-biased adult sex ratio, in which case males mate
with more than one female in their lifetimes. However the behavior of
three adult males in the presence of unmated and mated females
suggests that mated females do not mate with other males, similarly to
other species of the Theridion varians group (Knoflach 1998). My
preliminary observations indicate that males avoid the webs of
females with plugged epigyna.

Theridion italiense appears to prefer rocky, sunny places with
adequate moisture. It feeds on small insects; small ants were the most
frequently observed prey.

Habitat notes. — Specimens of Theridion italiense from Romania
were collected from low altitudes in limestone mountains (350-
1000 m) at the edges of Fagus sylvatica forests in open, sunny areas.
This type of habitat is classified as dry calcareous grassland and
steppe (code 34) according to the CORINE land cover project, or as
alpine and subalpine calcareous grasslands (code 6170) according to
the NATURA 2000 project (Donita et al. 2005). In this habitat
Theridion italiense can be found in crevices of rocky walls, in small
bushes, or in grassy vegetation, always close to the ground (about
10 cm above it), with the threads of the web attached to stony walls
on one side, to grass on the other side, and to the ground at its base.

The web of this species is constructed very close to the ground, being
well camouflaged under the branches and leaves of various plants. It
reaches a height of about 5-15 cm. The spider always weaves its web
under the lower branches of small woody or herbaceous plants, close to
the vertical surface of a rock. In the upper part of the web, the threads
are fixed to the underside of the leaves and on nearby stones. The web is
typical of the cobweb weavers, being three-dimensional, irregular, and
formed by densely tangled sticky and non-sticky lines. A series of
gumfoot threads are attached to the ground in order to catch small
terrestrial insects such as ants, small coleopterans, and even collembo-
lans. The web looks simple at the begining of construction (Fig. 5 A, B),
but becomes progressively more complex due to the new threads that
spider adds over time (Fig. 5C).

Individuals occupy the upper part of the web close to the stony wall
or under the leaves of various plants. Before the last molt females
construct in the upper part of the web a more or less coneshaped
retreat close to or under rocks, leaves or branches of small plants
(Fig. 5D, E). The retreat is made of nonsticky threads, as in
Echinotheridion otlum Levi 1963, Theridion nigroannulatum Keyserling
1884 or Theridion evexiim Keyserling 1884 (Eberhard et al. 2008)
When shelters are constructed under leaves they are not curled as in
Theridion nigroannulatum (Eberhard et al. 2008). The retreat is
camouflaged with moss, various vegetal debris and small grains of
sand. The female will stay in this shelter with an adult male (Fig. 5D)
until the last molt, after which copulation takes place. Before laying
the cocoon the female will enlarge this retreat so that it will be big
enough to protect her and her cocoon (Fig. 5E). When disturbed, T.
italiense will retreat into the upper part of the shelter rather than
dropping onto the ground. This kind of defence is similar to that
observed in Theridion evexiim (Eberhard et al. 2008).

DUMA— NEW DATA ON THERIDION ITALIENSE

387

Distribution. — Until now, Theridion italiense has been found only in
central Italy and in the southwestern Carpathian Mountains
(Romania). The Eastern Alps and the Dinaric and Rhodope
Mountains also have many karst formations and the same type of
habitats as those where the species was found (Donita & al. 2005) so I
infer that the species may also occur in the mountains of the former
Yugoslavia and Bulgaria as well.

ACKNOWLEDGMENTS

I especially want to thank Dr. Barbara Knoflach for her precious
help in the identification of the species, Dr. Christo Deltshev and Dr.
Urak Istvan with whom I have had important discussions about this
spider, and Dr. Yuri Marusik and Jorg Wunderlich for providing me
with taxonomical papers on some Theridion species.

LITERATURE CITED

Deltshev, C.D. 1992. A critical review of family Theridiidae (Araneae)
in Bulgaria. Acta Zoologica Bulgarica 43:13-22.

Donita, N., M. Pauca-Comanescu, A. Popescu, S. Mihailescu & LA.
Biri§. 2005. Habitatele din Romania. Editura Tehnica Silvica,
Bucure§ti.

Duma, 1. 2008. Theridion uhligi Martin, 1974 (Araneae: Theridiidae)
new to Romania. Entomologica Romanica 13:297-299.

Eberhard, W.G., 1. Agnarsson & H.W. Levi. 2008. Web forms and
phylogeny of theridiid spiders (Araneae: Theridiidae): chaos from
order. Systematics and Biodiversity 6:1-61.

Jocque, R. 1977. Contribution a la connaissance des araignees de
Belgique. Description de Theridion huhlei n. sp. Revue Arachno-
logique 1:59-63.

Knoflach, B. 1998. Mating in Theridion varians Hahn and related species
(Araneae: Theridiidae). Journal of Natural History 32:545-604.

Platnick, N.I. 2009. The World Spider Catalog, Version 10.0. American
Museum of Natural History, New York. Online at: http://research.
anmh.org/entomology/spiders/catalog/index.html. (Accessed 7 July
2009).

Weiss, 1. & A. Petri§or. 1999. List of the spiders (Arachnida: Araneae)
from Romania. Travaux du Museum National d’Histoire Natur-
elle, “Grigore Antipa.” 41:79-107.

Wunderlich, J. 2008. On extant and fossil (Eocene) European comb-
footed spiders (Araneae: Theridiidae), with notes on their
subfamilies, and with descriptions of new taxa. Beitriige zur
Araneologie 5:140-469.

Wunderlich, J. 1995. Beschreibung von drei bisher unbekannten west-
palaarktischen Arten den Gattung Theridion Walckenaer 1805
(Arachnida: Araneae: Theridiidae). Beitrage zur Araneologie
4:691-695.

Manuscript received 24 August 2008, revised 5 August 2009.

2009. The Journal of Arachnology 37:388-391

SHORT COMMUNICATION

Capture efficiency of an ant-eating spider, Zodmiellum asiaticum (Araneae: Zodariidae),

from Kazakhstan

Stano Pekar: Department of Botany and Zoology, Faculty of Sciences, Masaryk University, Kotlafska 2, 611 37 Brno,
Czech Republic. E-mail: pekar@sci.muni.cz

Abstract. Zodariellum asiaticum (Tyschchenko 1970) is an ant-eating spider from Central Asia. Using five syntopicaily
occurring ant species, namely Cataglyphis aenescens, Formica cunicularia (both Formicinae), Messor aralocaspiiis,
Tetramoriiim caespitiim (both Myrmicinae), and Tapinoma erraticum (Dolichoderinae) in a laboratory study of prey-
capture behavior, I evaluated capture frequency, attack latency, number of attacks, and paralysis latency. Although spiders
captured all five ant species, capture efficiency varied when spiders were tested with the different ant species, being highest
when the spiders were tested with F. cunicularia. I concluded that small juvenile Z. asiaticum probably adapt to feed
primarily on species of small dolichoderine and myrmicine ants and that large juvenile and the adult Z. asiaticum adapt to
feed primarily on large formicine ants.

Keywords: Stenophagy, specialization, myrmecophagy, prey-capture behavior

Spiders are well known for being euryphagous predators, (i.e.,
consuming a wide variety of prey), but many have an aversion to ants.
This makes any group of spiders that routinely eat ants (“myrme-
cophagy”) especially interesting. Myrmecophagic spiders are of
further significance because they often seem to adopt ant-specific
prey-capture behavior, and they may actively choose ants in
preference to other prey (e.g., Huseynov et al. 2008). There is
particular interest in spider species that might be exclusively
myrmecophagic.

Although myrmecophagy may be a rare phenomenon in spiders as
a whole, examples are found in an assortment of spider families,
including the Gnaphosidae, Oecobiidae, Salticidae, Theridiidae,
Thomisidae, and Zodariidae (Glatz 1967; Heller 1976; Porter &
Eastmond 1982; Jocque 1991; Castanho & Oliveira 1997; Jackson
et al. 1998). Species that appear to be exclusively myrmecophagic
include a theridiid Dipoena and a thomisid Aphantochihis (Umeda
et al. 1996; Castanho & Oliveira 1997). However, regardless of
whether a species is exclusively or partially myrmecophagic,
arachnologists have tended to envision myrmecophagic spiders as
preying on ants in general rather than as having become adapted to
particular kinds of ants. Yet there is evidence that at least some of
the myrmecophagic spiders show a preference for certain genera or
even species of ants within the family Formicidae as a whole.
Researchers need to gather more information about myrmecopha-
gic spiders so that we can determine how important the targeting of
particular ant taxa is for these predators. “Targeting” includes a
variety of adaptations by which a spider might specialize on
particular kinds of ants, including adaptation related to morphol-
ogy, physiology, and behavior.

The myrmecophagic spiders I investigated are from the family
Zodariidae, one of the most diversified families of spiders (Platnick
2009). The family Zodariidae is known for including a number of
myrmecophagic species (Jocque 1991), but the natural history of the
great majority of these species is still poorly known. Available
evidence suggests the species in four of the genera in the subfamily
Zodariinae, namely Diores, Trygetus, Zodariellum, and Zodarion, are
exclusively myrmecophagous (Marikovsky & Tyschchenko 1970;
Pekar et al. 2005; Haddad & Dippenaar-Schoeman 2006). However,
evidence that the predator distinguishes prey below the level of family
(Formicidae) has come from only one of these genera, namely
Zodarion (Pekar 2005; Pekar et al. 2008).

Here I consider a species from the genus Zodariellum. The species in
this genus are morphologically uniform (Marusik & Koponen 2001),
meaning that distinctive interspecific differences are evident only in
the details of sexual organ shape, not in structures such as the spider’s
chelicerae that function directly in predation. This suggests that, for
finding evidence of adaptation to specific types of ants, we should
investigate behavioral and predatory-related physiological traits.

In this study, I focus on some traits related primarily to behavior.
Ant-eating spiders typically capture ants using a ‘bite-and-release’
tactic (e.g., Jackson & van Olphen 1992; Cushing & Santangelo 2002).
One probable advantage of this mode of attack is that it enables the
predator to avoid being injured or killed when ants counter-attack.
Using this mode of attack, species from the genus Zodarion can
subdue a number of different kinds of ants, but apparently the
spider’s efficiency in capturing different kinds of ants varies
considerably. Evidence that efficiency varies comes from data on
paralysis latency (i.e., the time elapsing between when the ant is
attacked and when it becomes immobile) and the frequency of attacks
before the prey is eaten (e.g., Pekar 2005). These are the parts of the
predatory sequence to which I paid particular attention in this study
of Zodariellum.

Worldwide, there are 22 species in the genus Zodariellum, about 10
of which appear to be endemic to Central Asia (Platnick 2009). There
are published anecdotal prey records for two species, Z. asiaticum
Tyschchenko 1970 and Z. saharieme Denis 1959, feeding on ants
(Pierre 1959; Marikovsky & Tyschchenko 1970). Marikovsky
reported that the ant on which Z. asiaticum preys is primarily
Formica cunicularia Latreille, but predation was also observed on
Tetramoriiim caespitiim (Linnaeus), Messor aralocaspius Ruzsky, and
Cataglyphis aenescens (Nylander) (Marikovsky & Tyschchenko 1970;
Marikovsky 1979).

Zodariellum asiaticum, the species I investigated, occurs in
southeastern Kazakhstan. The specimens I used (eight female and
seven subadult individuals; body lengths 3.5M-.5 mm) were collected
in April on the sandy slopes of a semi-desert habitat along the Illi
River, near the city of Kapchagay (43°56'93.4N, 77°03'56.7E).
Spiders were identified using Marikovsky & Tyschchenko (1970)
and Marusik & Koponen (2001) and kept in glass tubes (diameter
10 mm, length 60 mm) with moistened substrate (plaster of Paris). Al!
spider specimens are deposited in the collection of arachnids of the
Department of Botany and Zoology, Masaryk University, Brno.

388

PEKAR— CAPTURE EFFICIENCY OF AN ANT-EATING SPIDER

389

Figure 1. — Comparison of four predatory traits of Z. asiaticum for five ant species (from three ant subfamilies). A. Mean capture frequency. B.
Mean latency to the first attack. C. Mean number of attacks. D. Mean latency to complete paralysis. Whiskers are 95% confidence intervals for means.

The ant fauna occurring syntopically with Z. asiaticum was
surveyed at the same sites (ants identified using Marikovsky 1979).
Five ant species were used in the experiments: Cataglvphis aenescens
(body length 4.5-8 mm) and Formica cimicularia (4—6 mm) (both
Formicinae), Messor aralocaspius (4.5-8 mm) and Tetramorium
caespitwn (3-3.5 mm) (both Myrmicinae), and Tapinoma erraticiim
(Latreille) (3.5^ mm) (Dolichoderinae). We collected ants used as
prey in the field a few hours before we used them in the experiment.

I chose one of the ant species (Messor) to serve as the standard for
initial feeding (i.e., after being collected, the spiders were fed with a
single individual of Messor the next day). Three days later, each spider
was offered successively, in random order, a single ant from each of the
five species. I tested each spider with each ant species only once.

There was a 2-day interval between successive trials. A single trial
consisted of releasing an ant into a dish occupied by a spider
(diameter 40 mm; filter paper glued to the bottom; thin layer of fluon
on the sides). Each spider had been in the Petri dish for one day
before the trial began. Spiders usually attacked within 30 s. If the
spider did not attack the ant within 10 min, I terminated the trial.
These aborted trials were classified as rejection of prey. For each trial,
I recorded attack latency (i.e., the time between when the spider
oriented itself toward the ant and the first attack), number of
successive attacks, and paralysis latency (i.e., the time between the
first attack and the prey becoming completely immobilized).

Data were analyzed using Linear Mixed-Effects Models (LME)
from the NLME-package within the R-environment (R Development

Core Team 2007). I chose this method because observations were not
independent (i.e., there was repeated use of the spider individuals) and
LME is designed for taking into account repeated measurements
(Pinheiro & Bates 2000). Both latencies appeared to come from
asymmetrical (skewed to the right) distributions. Therefore, I applied
logarithmic transformation, after which the data distribution
approximated the normal distribution. I expected that the size of
ant prey might affect the number of attacks and the paralysis latency.
Therefore, I used prey size as a covariate when analyzing the data.
Frequency of capture was compared using the Cochran Q test.

Predatory sequences in encounters between Z. asiaticum and ants
were similar to predatory sequences in encounters between species of
Zodarion and ants (e.g., Pekar 2004). Z. asiaticum approached ants
quickly from behind and attacked, usually by contacting the ant’s
dorsal thorax or leg, releasing the ant, and then continuing to attack
several more times or waiting until the ant became paralyzed.

Spiders attacked all five ant species, the most frequently attacked
species being F. cimicularia and the least frequently attacked being T.
erraticum (Fig. lA). However, differences in frequency of attacking
ant species were not statistically significant (Cochran Q test, = 4.8,
P = 0.31). Differences in latency of making the first attack were also
non-significant across ant species (LME, F442 = 1.9, P = 0.13,
Fig. IB) and for number of attacks (LME, ^4,42 = 0.7, P = 0.6,
Fig. 1C). When number of attacks were considered in relation to ant
size, differences in the data were not significantly different (i.e., number
of attacks is independent of ant size) (LME, Fi_42 = 0.07, P = 0.79).

390

THE JOURNAL OF ARACHNOLOGY

25

c

1

>.

S)

• C. aenescens
O F. cunicularia

20 -

15

m 10 -

5 -

5 6 7 8

Ant size [mm]

Figure 2. — Relationship between paralysis latency and the size of
ants for two formicine (A) and two myrmicine (B) ant species. Linear
model (^v = -0.78 + 3.33 x) is shown for myrmicine.

However, there were significant differences between ant species for
paralysis latency (LME, ^4,42 = 10.8, P < 0.0001, Fig. ID), latency for
Messor being about three times as long as latencies for Cataglyphis,
Formica and Tapinoma. Latency for Tetramoriiim was similar to
latency for Messor (Fig. ID). For formicine ants {Cataglyphis and
Formica), there were no significant differences in paralysis latency when
considered whether latency depended on ant size (LME, Fi 23 = 0.14, P
= 0.72, Fig. 2A). For myrmicine ants {Messor and Tetramoriiim), on
the other hand, latency for large ants was significantly longer than for
small ants (LME, F|,24 = 13.4, P = 0.001, Fig. 2B).

I found that Z. asiaticiim captured each of the five ant species used
in this study, suggesting that this spider may have some general
adaptations that enable it to be effective at capturing ants in general.
Yet there are some critical differences indicating that this predator
distinguishes ants below the level of family and has acquired
adaptations by which it can be particularly effective as a predator
of certain kinds of ants. Preference is a motivational trait that drives a
predator’s prey-choice behavior (Huseynov et al. 2008), and it seems
likely that my data on capture frequency and attack latency are
especially closely linked to the preferences of Z. asiaticiim. I might
predict that a predator will be more likely to attack and quicker to
attack prey it prefers. However, my hypothesis is that paralysis
latency and the numbers of attacks made by Z. asiaticiim on different
ant species are determined to a large extent by biochemical specificity

of the spider’s venom. If these predictions and hypotheses are correct,
then I have no evidence from this study of preference for particular
kinds of ants, nor does the number of attacks made before the ant is
immobilized provide evidence of venom specificity. However,
significant differences among ants were found for paralysis latency.
The data for paralysis latency do not simply corroborate the venom-
specificity hypothesis, but they add strength to this hypothesis and
imply that further investigation would be of interest.

More specifically, paralysis data support a hypothesis that Z.
asiaticiim has adapted in special ways as a predator that targets
formicine ants and F. cunicularia in particular. Stated more precisely,
Z. asiaticum specializes on the formicine ants. In this context,
“specialized” refers to having special characteristics (in this instance,
formicine-specific venom) that make a predator especially effective at
preying on a particular kind of prey.

It is of interest that no size-dependent relationship was evident for
formicine ants. This is what we might expect if the venom of Z.
asiaticum is specific to these ants. In an earlier study (Pekar et al.
2008), we demonstrated similar results for Zodarion germanicum (C.L.
Koch 1837), as this species performed best, in terms of growth,
development and survival, on a diet consisting of only formicine ants.
For this species as for Z. asiaticum, there was no significant variation
in paralysis efficiency dependent on ant size when the ants were
formicines. It is also noteworthy that Marikovsky reported a paralysis
latency of about 4 min when Z. asiaticum attacked F. cunicularia in
the field (Marikovsky & Tyschchenko 1970), as this corresponds
closely to the latencies I found in the laboratory.

Without further experiments, the connection between paralysis
latency and venom specificity remains only a hypothesis. Paralysis
latency may be influenced by variables that could not be controlled in
this study. For example, the volume of venom injected may influence
paralysis latency, but venom volume was not determined in this study.
That venom volume needs to be considered is illustrated by recent
findings from a study on an unrelated spider of the genus Cupiennius.
This spider was shown to inject larger volumes of venom when it
detected that the prey was especially dangerous (Wigger et al. 2002).

For the genus Zodarion, two distinct groups appear to be
identifiable in the context of prey specificity. First, there are species
that may normally exploit a single ant species. More specifically, there
appear to be species that exploit solely ants from the genus Messor.
Being polymorphic, ants of this genus may be feasible prey for
exploitation by Zodarion throughout the spider’s life cycle, as there
would always be available, regardless of the developmental stage to
which an individual of Zodarion might belong, a suitable size morph
of this ant species (Pekar, unpublished). The other group consists of
Zodarion species whose adults tend to prey most often on
monomorphic ant species. I hypothesize that these Zodarion species,
specifically the smaller juveniles, prey primarily on ant species that are
smaller than the species on which the adults prey. This implies that
the spiders adapt to the ants they target as they progress through their
life cycles. Often the stage-specific switch may be from smaller to
larger ant species belonging to the same ant subfamily (Pekar et al.
2008).

Field data suggest that Z. asiaticum frequently exploits ant species
from different subfamilies. For Z. asiaticum, there may be a
developmental adaptation in the ants primarily targeted, with small
juveniles feeding on small dolichoderine ants such as Tapinoma and
on small myrmicine ants such as Tetramorium, and with large
juveniles and adults feeding primarily on large formicines such as F.
cunicularia. On one occasion, I found a juvenile of Z. asiaticum
feeding on Tapinoma in the field, which helps to support this
hypothesis. If this hypothesis is corroborated, then it will be of
interest to investigate the venom specificity of the early juvenile stages
of Z. asiaticum to determine whether small juveniles have venom that
is specific not to formicines, but instead to dolichoderines or
myrmicines or both.

PEKAR— CAPTURE EEEICIENCY OF AN ANT-EATING SPIDER

391

ACKNOWLEDGMENTS

I would like to thank M. Martisova for help with the spider
collection in the field and A. Gromov for help with the logistics in
Kazakhstan. Further, I am very grateful to R. Jackson for useful
comments and extensive text alterations. The study was supported by
project no. MSM0021622416 provided by the Ministry of Education,
Youth and Sports of the Czech Republic.

LITERATURE CITED

Castanho, L.M. & P.S. Oliveira. 1997. Biology and behaviour of the
neotropical ant-mimicking spider Aphcmtochilus rogersi (Araneae:
Aphantochilidae): nesting, maternal care and ontogeny of ant-
hunting techniques. Journal of Zoology 242:643-650.

Cushing, P.E. & R.G. Santangelo. 2002. Notes on the natural history
and hunting behavior of an ant-eating zodariid spider (Arachnida,
Araneae) in Colorado. Journal of Arachnology 30:618-621.

Glatz, L. 1967. Zur Biologie und Morphoiogie von Oecohius annulipes
Lucas (Araneae, Oecobiidae). Zeitschrift fiir Morphoiogie der
Tiere 61:185-214.

Haddad, C.R. & A.S. Dippenaar-Schoeman. 2006. Epigeic spiders
(Araneae) in pistachio orchards in South Africa. African Plant
Protection 12:12-22.

Heller, G. 1976. Zum Beutefangverhalten der ameisenfressenden
Spinne Callilepis noctunui (Arachnida: Araneae: Drassodidae).
Entomologica Germanica 3:100-103.

Huseynov, E.F., F.R. Cross & R.R. Jackson. 2005. Natural diet and
prey-choice behaviour of Aeliirilliis muganicus (Araneae: Sal-
ticidae), a myrmecophagic jumping spider from Azerbaijan.
Journal of Zoology 267:159-165.

Huseynov, E.F., R.R. Jackson & F.R. Cross. 2008. The meaning of
predatory specialization as illustrated by Aelurillus m-nigrum, an
ant-eating jumping spider (Araneae: Salticidae) from Azerbaijan.
Behavioral Processes 77:389-399.

Jackson, R.R., D. Li, A. Barrion & G.B. Edwards. 1998. Prey-capture
techniques and prey preferences of nine species of ant-eating
jumping spiders (Araneae: Salticidae) from the Philippines. New
Zealand Journal of Zoology 25:249-272.

Jackson, R.R. & A. van Olphen. 1992. Prey-capture techniques and
prey preferences of Chrysitia, Natta and Siler, ant-eating jumping
spiders (Araneae, Salticidae) from Kenya and Sri Lanka. Journal
of Zoology 227:163-170.

Jocque, R. 1991. A generic revision of the spider family Zodariidae
(Araneae). Bulletin of the American Museum of Natural History
201:1-160.

Marikovsky, P.I. & V.P. Tyschchenko. 1970. Pauk-myrmekofil
(Zodarion asiaticum Tysch. sp. nova) i nekotorye tcherty jevo
biologii [Myrmecophilous spider (Zodarion asiaticum Tysch. sp.
nova) and some aspects of its biology]. Trudy Alma-Atinskogo
Gosudarstvenogo Zapovednika 9:196-201.

Marikovsky, P.I. 1979. Muravy pustynh Semiretchia [Ants of
Semiretchie Deserts]. Nauka, Akademia nauk Kazachskoj SSR,
Alma-Ata.

Marusik, Y.M. & S. Koponen. 2001. Spiders of the family Zodariidae
from Mongolia (Arachnida: Araneae). Reichenbachia 34(4):39-48.

Norgaard, E. 1956. Environment and behaviour of Theridion scixatile.
Oikos 7:159-192.

Pekar, S. 2004. Predatory behavior of two European ant-eating
spiders (Araneae, Zodariidae). Journal of Arachnology 32:31-41.

Pekar, S. 2005. Predatory characteristics of ant-eating Zodarion
spiders (Araneae: Zodariidae): potential biological control agents.
Biological Control 34:196-203.

Pekar, S., J. Krai & Y.D. Lubin. 2005. Natural history and karyotype
of some ant-eating zodariid spiders (Araneae, Zodariidae) from
Israel. Journal of Arachnology 33:50-62.

Pekar, S., S. Toft, M. Hruskova & D. Mayntz. 2008. Dietary and
prey-capture adaptations by which Zodarion gernumiciun, an ant-
eating spider (Araneae: Zodariidae), specialises on the Formicinae.
Naturwissenschaften 95:233-239.

Pierre, F. 1959. Le mimetisme chez les araignees myrmecomorphes.
Annee Biologie 35:191-201.

Pinheiro, J.C. & D.M. Bates. 2000. Mixed-Effects Models in S and S-
PLUS. Springer, New York.

Platnick, N.I. 2009. The World Spider Catalog, Version 9.5. American
Museum of Natural History, New York. Online at http://research.
amnh.org/entomology/spiders/catalog/INTRO 1 .html.

Porter, S.D. & D.A. Eastmond. 1982. Euryopis coki (Theridiidae), a
spider that preys on Pogonornvnnex ants. Journal of Arachnology
10:275-277.

R Development Core Team. 2007. R: A Language and Environment
for Statistical Computing. R Foundation for Statistical Comput-
ing, Vienna. Online at http://www.R-project.org.

Umeda, Y., A. Shinkai & T. Miyashita. 1996. Prey composition of
three Dipoena species (Araneae: Theridiidae) specializing on ants.
Acta Arachnologica 45:95-99.

Wigger, E., L. Kuhn-Nentwig & W. Nentwig. 2002. The venom
optimisation hypothesis: a spider injects large venom quantities
only into difficult prey types. Toxicon 40:749-752.

Manuscript received 2 February 2009, revised 7 August 2009.

2009. The Journal of Arachnology 37:392-393

INSTRUCTIONS TO AUTHORS

(instructions last revised September 2009)

General: The Journal of Arachnology publishes scientific
articles reporting novel and significant observations and data
regarding any aspect of the biology of arachnid groups.
Feature articles and short communications must be scientif-
ically rigorous and report substantially new information.
Submissions that are overly narrow in focus (e.g., local faunal
lists), have poorly substantiated observational data, or that
present no new information will not be considered.

Manuscripts must be in English and should be prepared in
general accordance with the current edition of the Council of
Biological Editors Style Manual unless instructed otherwise
below. Use the active voice throughout. Authors should
consult a recent issue of the Journal of Arachnology for
additional points of style. Manuscripts longer than three
printed journal pages (12 or more double-spaced manuscript
pages) should be prepared as Feature Articles, shorter papers
as Short Communications. One invited Review Article will be
published in the first issue of each year at the discretion of the
editors. Suggestions for review articles may be sent to the
Managing Editor. Unsolicited review articles are also wel-
comed. All review articles will be subject to the same review
process as other submissions.

Submission: Submissions must be sent electronically in
Microsoft Word (or Word compatible) format. Large plates
and figures may be sent as pdf or jpg files. Send submissions to
the Managing Editor of the Journal of Arachnology: Douglass
H. Morse, Managing Editor, Hermon Carey Bumpus Professor
of Biology Emeritus, Department of Ecology & Evolutionary
Biology, Box G-W, Brown University, Providence, RI 02912
USA [Telephone; 401-863-3152; Pax: 401-863-2166; E-mail:
d_morse@brown.edu]

The Managing Editor will acknowledge receipt of the
manuscript, assign it a manuscript number and forward it to
an Associate Editor for the review process. Correspondence
relating to manuscripts should be directed to the Associate
Editor and should include the manuscript number. If the
manuscript is accepted, the author will be asked to submit the
final copy electronically to the Associate Editor. Submission
of final illustrations is detailed below. Authors are expected to
return revisions promptly. Revised manuscripts that are not
returned in a reasonable time period (no longer than six
months for minor revisions and one year for major revisions)
will be considered new submissions.

Voucher Specimens; Specimens of species used in your
research should be deposited in a recognized scientific
institution. All type material must be deposited in a recognized
collection/institution.

FEATURE ARTICLES

Title page. — The title page includes the complete name,
address, and telephone number of the corresponding author; a
FAX number and electronic mail address if available; the title
in sentence case, with no more than 65 characters and spaces

per line in the title; each author’s name and address; and the
running head.

Running head. — The author’s surname(s) and an abbreviat-
ed title should be typed in all capital letters and must not
exceed 60 characters and spaces. The running head should be
placed near the top of the title page.

Abstract. — The heading should be placed at the begin-
ning of the first paragraph set off by a period.

Keywords. — Give 3-5 appropriate keywords or phrases
following the abstract. Keywords should not duplicate words
in the title.

Text. — Double-space text, tables, legends, etc. throughout.
Three levels of heads are used.

• The first level (METHODS, RESULTS, etc.) is typed in
capitals and centered on a separate line.

• The second level head begins a paragraph with an in-
dent and is separated from the text by a period and a
dash.

• The third level may or may not begin a paragraph but is
italicized and separated from the text by a colon.

Use only the metric system unless quoting text or
referencing collection data. If English measurements are used
when referencing collection data, then metric equivalents
should also be included parenthetically. All decimal fractions
are indicated by a period (e.g., -0.123). Include geographic
coordinates for collecting locales if possible.

Citation of references in the text: Cite only papers already
published or in press. Include within parentheses the surname
of the author followed by the date of publication. A comma
separates multiple citations by the same author(s) and a
semicolon separates citations by different authors, e.g., (Smith
1970), (Jones 1988; Smith 1993), (Smith & Jones 1986, 1987;
Jones et al. 1989). Include a letter of permission from any
person who is cited as providing unpublished data in the form
of a personal communication.

Citation of taxa in the text: Include the complete taxonomic
citation for each arachnid taxon when it first appears in the
manuscript. For Araneae, this information can be found online
at http://research.amnh.org/entomology/spiders/catalog/INTR02.
html. For example, Araneus diadematus Clerck 1757. Citations
for scorpions can be found in the Catalog of the Scorpions of the
World (1758-1998) by V. Fet, W.D. Sissom, G. Lowe & M.E.
Braunwalder. Citations for pseudoscorpions can be found at
http://www.museum.wa.gov.au/arachnids/pseudoscorpions. Cita-
tions for some species of Opiliones can be found in the Annotated
Catalogue of the Laniatores of the New World ( Arachnida,
Opiliones) by A.B. Kury. Citations for other arachnid orders can
be found in Catalogue of the Smaller Arachnid Orders of the World
by M.S. Harvey.

Literature Cited section. — Use the following style and
formatting exactly as illustrated; include the full unabbre-
viated journal title. Personal web pages should not be included

392

INSTRUCTIONS TO AUTHORS

393

in Literature Cited. These can be cited within the text as (John
Doe, pers. website) without the URL. Institutional websites
may be included in Literature Cited.

Carico, J.E. 1993. Trechaleidae: a “new” American spider
family. Pp. 305. In Proceedings of the Ninth International
Congress of Arachnology, Panama 1983. (W.G. Eberhard,
Y.D. Lubin & B.C. Robinson, eds.). Smithsonian Institu-
tion Press, Washington, D.C.

Huber, B.A. & W.G. Eberhard. 1997. Courtship, copulation,
and genital mechanics in Physocyclus globosus (Araneae,
Phoicidae). Canadian Journal of Zoology 74:905-918.
Krafft, B. 1982. The significance and complexity of commu-
nication in spiders. Pp. 15-66. In Spider Communications:
Mechanisms and Ecological Significance. (P.N. Witt & J.S.
Rovner, eds.). Princeton University Press, Princeton, New
Jersey.

Platnick, N.I. 2009. The World Spider Catalog, Version 9.5.
American Museum of Natural History, New York. On-
line at http://research.amnh.org/entomology/spiders/catalog/
INTR01.html

Roewer, C.F. 1954. Katalog der Araneae, Volume 2a. Institut
Royal des Sciences Naturelles de Belgique, Bruxelles. 923 pp.

Footnotes. — Footnotes are permitted only on the first
printed page to indicate current address or other information
concerning the author. All footnotes are placed together on a
separate manuscript page. Tables and figures may not have
footnotes.

Taxonomic articles. — Consult a recent taxonomic article in
the Journal of Arachnology for style or contact the Subject
Editor for Taxonomy and Systematics. Papers containing
original descriptions of focal arachnid taxa should be listed in
the Literature Cited section.

Tables. — Each table, with the legend above, should be
placed on a separate manuscript page. Only horizontal lines
(no more than three) should be included. Tables may not have
footnotes; instead, include all information in the legend.

Illustrations. — Original illustrations should be sent electron-
ically when the manuscript is submitted, preferably in tiff or
jpeg format. Distribution maps should be considered figures
and numbered consecutively with other figures. (Authors
wishing to submit figures as hard copies should contact the
Editor-in-Chief for specifications.) At the submission and
review stages, the resolution standards may be low as long as
editors and reviewers can view figures effectively. Final
illustrations must be submitted to the Editor-in-Chief,
typically by e-mail or on a CD, to ensure that the electronic
versions meet publication standards and that they match the
printed copy. All figures should be at least 4 inches wide, but
no larger than a sheet of letter-size paper with 1-inch margins
all around. The resolution should be at least 300 dpi (or ppi)
for halftone or color figures and 1200 dpi for line drawings. A
guide to the Digital Art Specs for Allen Press is posted online
at: http://www2.aIlenpress.coni/allen_press/apguides/Digital_
Art_Spec.pdf To determine if your electronic figures adhere
to the Allen Press specifications, you can also go to http://
verifig.allenpress.com, type in the password, allenpresscmy.

and follow the instructions. Color plates can be printed, but
the author must assume the full cost paid in advance, currently
about $1100 US per color plate. Alternatively, an author can
opt to have a figure printed in black and white, but for $30/
page have that figure appear in color in the online version of
the journal. Most figures will be reduced to single-column
width (9 cm, 3.5 inches), but large plates can be printed up to
two-columns width (18 cm, 7 inches).

Address all questions concerning illustrations to the Editor-
in-Chief of the Journal of Arachnology: James E. Carrel,
Editor-In-Chief Division of Biological Sciences, 209 Tucker
Hall, University of Missouri-Columbia, Columbia, MO 65211-
7400, USA [Telephone: 573-882-3037; FAX: 573-882-0123; E-
mail: carrelj@missouri.edu]

Legends for illustrations should be placed together on the
same page(s) and separate from the illustrations. Each plate
must have only one legend, as indicated below:

Figures 1M-. A-us x-us, male from Timbuktu. 1, Left leg; 2,
Right chelicera; 3, Dorsal aspect of genitalia; 4, Ventral aspect
of abdomen. Scale = 1.0 mm.

The following alternate Figure numbering is also accept-
able:

Figures la-e. A-us x-us, male from Timbuktu, a. Left leg;
b. Right chelicerae; c. Dorsal aspect of genitalia; d. Ventral
aspect of abdomen. Scale = 1.0 mm.

Assemble manuscript. — The manuscript should appear in
separate sections or pages in the following sequence; title page,
abstract, text, footnotes, tables with legends, figure legends,
figures. If possible, send entire manuscript, including figures,
as one Microsoft Word document. If figures or plates are
large, please separate them from the text and send them as a
pdf or jpg file.

Page charges, proofs and reprints. — Page charges are
voluntary, but non-members of AAS are strongly encouraged
to pay in full or in part for their article ($75 / journal page).
The author will be charged for changes made in the proof
pages. Hard copy or pdf reprints are available only from Allen
Press and should be ordered when the author receives the
proof pages. Allen Press will not accept reprint orders after the
paper is published. The Journal of Arachnology also is
available through www.bioone.org and www.jstor.org. There-
fore, you can download the PDF version of your article from
one of these sites if you or your institution is a member. PDFs
of articles older than one year will be made freely available
from the AAS website.

SHORT COMMUNICATIONS

Short Communications are usually limited to three journal
pages, including tables and figures (1 1 or fewer double-spaced
manuscript pages including Literature Cited; no more than 2
small figures or tables). Internal headings (METHODS,
RESULTS, etc.) are omitted. Short communications must
include an abstract and keywords.

COVER ARTWORK

Authors are encouraged to send quality photographs
(preferably in color) to the editor-in-chief to be considered
for use on the cover. Images should be at least 300 dpi.

I ‘ - ^,*11^ !’• IX

i.;

K|ht>ti uni>m «<i fcittM i tfi) ' 'i *' ^ii■c^/^>^M/.

r''';snv - vJ .. jPic-ir. ).3fl!

^ > c.-./.vKwi

^Snmuikt ,i)he^

i-iiT'ir- 1 ft^-^ f4tf

4^ njrar ’^w- '"‘rrs'v^ ‘■^' -* r»5fli^wtT * ^ ^ i*.^v.'^*-

^ .!is*!0to '■«* ■ ' • ^' ' ' ■ ■ - '> ' ’ 'K - * ■ ' “« »

^fifmff3(jlt^ /.••'lift* ni'iJ'^th^ilrii 7*jJj«'i t© •tyiLlwj Ui^jUn

H^.»t>,nfhQ»'-vj^, . V|UR5 iKnar* .>l/nu^ ‘

' ^ ^ ■ - • - • r - . — ^4 ^ ^ . 4 •*v«Nr'4j ,. W' * ' ** J

Kr../ _ . ,;V^ ..„ : ^ : , -. . _ , , , ,

iJm V ^■mQ,> ■ J^vwrf/

4) - . ,t'i1i^.

’ll 'I t^.

- , Ij'jV Ti'rf-> ■ j \ 'v'U'iiTPTU.'Or’l" -«wwJ|r’j|&®9ep£ln9B(1

sill wsO'.-'i ^ ^ jr‘1t?C'>^'vwp jfc *

' «*KQ&il4l "**^ -*-- - '■* ^ ^ ^ i(rJBnw*.>fc.U?.»jaUi .»« ... t*^»^ii. j^.i. .. .<^/»jAiiUi'irjuit*.i3k.iVv.rt

<fr4ll^.ilMfi|iV.'

i**i|! •. 4,. *V«t'^gE;9l<)1*£ O '■‘■V *rJttp,i h-ff0 tl| ..

fc,:« • • ->:■ ‘-U 'iith*xt$9m' -> iiJi'^rt.%i«-i<irt( :

'-V

I. 'I.

♦rtnitTpMVfej ■Uiv 't i. to • •■•'

ikaw^^Ka^. : ^'j'\<’-~iu'p ni, *;'! ' '■ ,;ji)' 4'i'i5«5r.a:

ttfi'

m

'S' >

l>-v 4- *

-BrH<n

,'S J>.. 4 »-■ ,1

-•" ■* >•

• • .4. 4,4-.-* <i>4»ii'v

"■.aW*- - ->n -i'»;^ -:^ vA- .„';•/ .4^4^

1

..*■1 1

Volume 37

CONTENTS
Journal of Arachnology
Featured Articles

SMITHSONIAN INSTITUTION LIBRARIES

3 9088 01513 4224

Number 3

Frame-web-choice experiments with stingless bees support the prey-attraction hypothesis for silk decorations in Argiope

savignyi by Dumas Galvez 249

Conflict or cooperation in the courtship display of the white widow spider, Latrodectus pallidus by Ally R, Harari,

Merav Ziv & Yael Lubin 254

Redescription and transfer of Geolycosa grandis (Araneae: Lycosidae) to the genus Hogna by Jozef Slowik & Paula E.

Cushing 261

Nephilid spider eunuch phenomenon induced by female or rival male aggressiveness by Matjaz Kuntner, Ingi

Agnarsson & Matjaz Gregoric 266

On the charitonovi species group of the spider genus Coelotes (Araneae: Amaurobiidae) by Xin-Ping Wang &

Ming-Sheng Zhu 272

Predation by Misumenops pallidus (Araneae: Thomisidae) on insect pests of soybean cultures in Buenos Aires Province,

Argentina by Alda Gonzalez, Gerardo Liljesthrom, Elizabet Minervino, Dolores Castro, Sandra Gonzalez &

Andrea Armendano 282

Karyotypes of the Neotropical pseudoscorpions Semeiochernes armiger and Cordylochernes scorpioides

(Pseudoscorpiones: Chernetidae) by Frantisek §fahlavsky, Jeanne A. Zeh, David W. Zeh & Jiri Krai 287

Palpal urticating hairs in the tarantula Ephebopus: fine structure and mechanism of release by Rainer Foelix, Bastian

Rast & Bruno Erb 292

Possible niche differentiation of two desert wandering spiders of the genus Syspira (Araneae: Miturgidae) by Irma

Gisela Nieto-Castaneda & Maria Luisa Jimenez-Jimenez 299

Construction and function of the web of Tidarren sisyphoides (Araneae: Theridiidae) by Ruth Madrigal-Brenes &

Gilbert Barrantes 306

Scorpion taphonomy: criteria for distinguishing fossil scorpion molts and carcasses by Victoria E. McCoy & Danita S.

Brandt 312

Temperature and desiccation effects on the antipredator behavior of Centruroides vittatus (Scorpiones: Buthidae) by B.

Evan Carlson & Matthew P. Rowe 321

Cytogenetics of three species of scorpions of the genus Brachistosternus from Argentina (Scorpiones: Bothriuridae) by

Sergio G. Rodriguez-Gil, Andres A. OJanguren-Affilastro, Leonel M. Barral, Cristina L. Scioscia & Liliana M.

Mola 331

Redescription of Plesiochactas mitchelli (Scorpiones: Euscorpiidae): a rare scorpion from Central America by Kaleb

Zarate-Galvez & Oscar F. Francke 338

New data on the genus Urophonius in Patagonia with a description of a new species of the exochus group (Scorpiones:

Bothriuridae) by Andres Alejandro OJanguren-Affilastro & German Cheli 346

Foraging strategies and diet composition of two orb web spiders in rice ecosystems by Hafiz Muhammad Tahir, Abida

Butt & Sher Muhammad Sherawat 357

Book Review

Dominican Amber Spiders: a Comparative Palaeontological-Neontological Approach to Identification, Faunistics, Ecology
and Biogeography. By David Penney. 2008. Siri Scientific Press, Manchester, UK. 176 pp.

ISBN 978-0-9558636-0-8 by Yuri M. Marusik 363

Short Communications

Habitat selection and potential antiherbivore effects of Peucetia flava (Oxyopidae) on Solanum thomasiifolium
(Solanaceae) by Giuliano B. Jacobucci, Lenice Medeiros, Joao Vasconcellos-Neto & Gustavo Q.

Romero 365

Reducing scorpion fluorescence via prolonged exposure to ultraviolet light by Carl T. Kloock 368

Presence of Vaejovis franckei in epiphytic bromeliads in three temperate forest types by Demetria Mondragon &

Gabriel Isaias Cruz Ruiz 371

Taxonomic notes on the genus Microfilistata (Araneae: Filistatidae), with a description of a new species from

Turkmenistan by Sergei L. Zonstein 373

Evidence for multiple paternity in broods of the green lynx spider Peucetia viridans (Araneae: Oxyopidae) by Martin G.
Ramirez, Elizabeth C. Wight, Victoria A, Chirildan, Evelyn S, Escobedo, Lauren K. Quezada,

Antu Schamberger, Jodi A. Kagihara & Carolyne L. Hoey 375

A new approach to examining scorpion peg sensilla: the mineral oil flood technique by Elizabeth D. Knowlton «&

Douglas D. Gaffin 379

New data on Theridion italiense, with description of the unknown female by loan Duma 383

Capture efficiency of an ant-eating spider, Zodariellum asiaticum (Araneae: Zodariidae), from Kazakhstan by Stano

Pekar 388

USERNAME: hentz09

PASSWORD: sphodros09