Content uploaded by Gail E Stratton
Author content
All content in this area was uploaded by Gail E Stratton on Aug 28, 2015
Content may be subject to copyright.
Behav Ecol Sociobiol (1996) 38 : 17–23 © Springer-Verlag 1996
Sonja J. Scheffer · George W. Uetz · Gail E. Stratton
Sexual selection, male morphology, and the efficacy of courtship
signalling in two wolf spiders (Araneae: Lycosidae)
Received: 30 September 1994 / Accepted after revision: 4 August 1995
Abstract Males of the brush-legged wolf spider,
Schizocosa ocreata (Araneae: Lycosidae), possess a
conspicuous male secondary sexual character: dark pig-
mentation and tufts of bristles on the tibiae of their
forelegs. We tested several hypotheses relating to the
role of this conspicuous trait in sexual selection. Triad
mating experiments suggest that the tufts do not play
an obvious role in the operation of sexual selection by
either male competition or female choice, as there were
no significant differences in the mating success of intact
and experimentally shaved males. However, females
mated more often with males that initiated courtship
first, suggesting that capture of a female’s attention by
male signalling may play a critical role. In behavioral
experiments that paired a single male with a female in
arenas that allowed both visual and vibratory signal
transmission during courtship, female receptivity did
not vary significantly with the presence or absence of
tufts. However, experiments that isolated the visual
component of communication (by eliminating vibra-
tory communication) revealed a significant effect of the
presence of tufts: females showed receptivity less often
to males with tufts removed. Female response to visual
signals was much greater in S. ocreata than in its sib-
ling congener, Schizocosa rovneri, which lacks male
tufts. We hypothesize that the tufts serve to increase
the efficacy of visual displays of S. ocreata, as vibra-
tory communication is constrained by the complex leaf
litter habitat of some populations. Such environmental
constraints may make visual signalling over distance a
critical factor for effective courtship communication,
which may in turn strongly influence male fitness.
Key words Spiders · Schizocosa · Sexual
selection · Courtship communication · Secondary
sexual characteristics
Introduction
Within evolutionary biology, much current debate is
focused on the origin and function of male secondary
sexual characteristics and the evolution of female pre-
ferences for such traits (Andersson 1994). Empirical
investigations of the function of a number of male sec-
ondary sexual characteristics have demonstrated that
these traits can influence male mating success through
male competition, female choice or both (see
Andersson 1995 for review). In many taxa, male sec-
ondary sexual traits appear to increase the conspicu-
ousness of the bearer. In courtship signalling, increased
conspicuousness may increase the efficacy of commu-
nication between signaller and recipient (Guilford and
Dawkins 1991; Hasson 1991). Traits that enable the
receiver to perceive and/or recognize a signal against
a background of visual or auditory “noise” will have
a selective advantage (Guilford and Dawkins 1991).
Since most signals must travel through the environ-
ment, the signalling environment plays an important
role in determining the form of signals used in com-
munication (Endler 1992). In general, it seems that
animal communication systems have evolved to use
most heavily those sensory modalities best suited for
effective signal transmission within particular habitats
(Krebs and Davies 1993). In addition to influencing
which of the various sensory modalities are used, the
signalling environment may also influence the form of
the signals within each modality (Morton 1975; Hunter
and Krebs 1979; Endler 1980, 1992; Fleishman 1992).
S.J. Scheffer1· G.W. Uetz (*)
Department of Biological Sciences, University of Cincinnati,
Cincinnati, OH 45221-0006, USA
G.E. Stratton2
Department of Biology, Albion College, Albion, MI 49224, USA
Present addresses:
1Department of Ecology and Evolution, State University of
New York at Stony Brook, Stony Brook, NY 11794-5245, USA
2Department of Biology, Rhodes College, Memphis, TN 38112-
1690, USA
Comparative studies of closely related species are a
powerful tool for investigating how ecological factors
influence the evolution of courtship (Ryan and Rand
1990, 1993; Basolo 1990; Proctor 1991, 1992). In this
study, we investigate the role played by a conspicuous
male secondary sexual characteristic in the courtship
of the wolf spider Schizocosa ocreata (Hentz) (Araneae:
Lycosidae). The male trait, a conspicuous tuft of bris-
tles on the first pair of legs, is the only known mor-
phological feature that distinguishes S. ocreata from its
sibling species Schizocosa rovneri (Uetz and Dondale
1979; Uetz and Denterlein 1979; Stratton and Uetz
1986). We first investigated the current function of the
tufts in S. ocreata, and then compared courtship
between the two species, with particular emphasis on
the sensory modalities used and the efficacy of signal
transmission in the habitat of each species.
Schizocosa ocreata provides a good system in which
to investigate the possible functional significance of a
male secondary sexual characteristic. Because female
discrimination of courtship cues is important at the
heterospecific level within the genus, similar mecha-
nisms might also allow discrimination among
conspecifics. Males and females associate only for the
duration of courtship and copulation. Females receive
no material benefits from males, so female preference,
if it occurs, must be based solely on male morphology
and behavior.
Naural history, morphology, and behavior
Schizocosa ocreata and S. rovneri are ground-dwelling
wolf spiders found in the leaf litter of deciduous forests
throughout the eastern United States. In the northern
portion of its range, S. ocreata is most commonly found
in upland forests with complex leaf litter, while in the
same range S. rovneri is usually found on the com-
pacted leaf litter of floodplain forests (Uetz and
Denterlein 1979). Morphologically, these species can
be distinguished only by a male secondary sexual char-
acteristic, a conspicuous tuft of bristles and dark pig-
mentation on the tibia and patella of the first pair of
legs of mature male S. ocreata, which is lacking in
S. rovneri (as well as in the females and juveniles of
both species; Uetz and Dondale 1979).
The courtship behavior of male lycosid spiders often
includes both a vibratory and a visual component
(Rovner 1967, 1975; Uetz and Stratton 1982). The re-
lative importance of these two components of courtship
can vary among species (Stratton 1982). Conspicuous
tufts on the front legs of male S. ocreata suggest that
visual as well as vibratory communication may be
important in this species. The courtship of male
S. ocreata includes in-unison tapping of the decorated
forelegs, as well as numerous leg waves and leg arches
(Fig. 1). At the same time, substratum-coupled vibra-
tions are produced by stridulatory organs located in
the male pedipalps (Uetz and Stratton 1982). The
courtship of male S. rovneri appears to be much less
visually complex, involving fewer leg movements and
consisting mainly of stationary body pulses. The sub-
stratum-coupled vibrations produced by S. rovneri are
greater in amplitude and differ temporally from those
produced by S. ocreata (see Uetz and Denterlein 1979;
Stratton and Uetz 1981, 1983, 1986 for a complete
description of courtship in these species).
Female Schizocosa can distinguish heterospecific
from conspecific males on the basis of courtship cues
(Stratton and Uetz 1981, 1983). The males of both
species will court heterospecific as well as conspecific
females with almost equal frequency, but female spi-
ders are receptive only to conspecific males (Uetz and
Denterlein 1979; Stratton and Uetz 1981, 1983, 1986;
Uetz and Stratton 1982). Female Schizocosa are slightly
larger than males, and have the ability to reject court-
ing males. Under laboratory conditions, females that
are not sexually receptive may lunge toward a court-
ing male with open fangs, and sometimes even kill and
eat the male. While pre- or post-mating cannibalism is
rare (< 5% of 30–50 matings observed in the labora-
tory, Uetz, unpublished), a male confined with an un-
receptive female will in most cases quickly cease
courtship and remain fairly stationary, suggesting
avoidance of predation. Rarely will copulation occur
until stereotypic receptivity behavior has been dis-
played by the female (Montgomery 1903; Uetz and
Denterlein 1979; Scheffer, unpublished data). These
behaviors include a slow 180°[360° turn and settling
into a position with the cephalothorax lowered and the
abdomen raised.
The role of the conspicuous tufts on the forelegs of
male S. ocreata in species recognition and/ or sexual
selection has not been previously investigated. The tufts
may be important in inter-sexual selection via female
mate choice, and/ or intrasexual selection via male-male
interactions. Males confined together in small spaces
at unnaturally high densities appear to establish a lin-
ear dominance hierarchy (Aspey 1977a, b). Many leg-
waving movements were found to be involved in the
dominance interactions, suggesting a possible role for
the tufts in intrasexual communication.
18
Fig. 1 Male of Schizocosa ocreata during courtship display, drawn
from a single frame of film
Materials and methods
Organisms
All S. ocreata used in this study were collected as immatures at the
Cincinnati Nature Center (Clermont Co., Ohio) during April and
May of 1987 and 1988. Spiders were individually housed in visu-
ally shielded containers in a laboratory with controlled tempera-
ture (23°C), light regime (similar to field conditions, 13 h light: 11 h
dark) and humidity (RH > 60%).
Several experiments required males whose tufts had been
removed by shaving. This involved anesthetization of males by
exposure to carbon dioxide for 45–75 s. Each spider was placed in
a shallow depression on the stage of a dissecting microscope and
held with a ball of cotton, leaving only the forelegs exposed. The
bristles of the tufts were quickly removed with a microscalpel.
Anesthetization and shaving were usually completed in less than
2 min per spider, and a second dose of carbon dioxide was required
only rarely. The spiders recovered quickly from the shaving process;
most were awake by the time both legs had been shaved and they
immediately began to groom upon release from the shaving depres-
sion. Normal exploratory behaviors were usually exhibited shortly
thereafter. A preliminary study showed that the willingness to court
of anesthetized-and-shaved males, anesthetized-only males, and
control males was not significantly different (as measured by either
the latency of courtship behaviors after exposure to a female or the
presence/absence of courtship after exposure to a female (Kruskal-
Wallis H= 2.105, P> 0.1; v2= 0.3, df = 2, P> 0.1 respectively,
n= 15 spiders per treatment)). As a precaution, unshaved males
used in the experiments were also anesthetized for 45–75 s.
Experiment 1a, b. Influence of the tufts on male mating
success in S. ocreata
If the male tufts are important in either male interactions or female
choice, then there should be a difference in the mating success of
shaved and tufted males. Two triadic (two males, one female) mat-
ing experiments testing this hypothesis were conducted:
a. In the first triadic mating experiment, 20 pairs of S. ocreata
males, each consisting of a shaved male and an intact male, were
used. Males were matched for size; both had cephalothorax
widths within 0.1 mm of each other (x= 3.02 mm, SD = 0.41,
n= 24). A female and her paper cage liner (bearing dragline silk
and silk-bound sex pheromone) were introduced into a circular
culture dish (diameter 15 cm). The female was covered with an
opaque specimen cup (120 cc) and given 5 min to acclimate
before a pair of males was introduced. When both males had
been observed performing at least one “jerky tap” behavior, indi-
cating the onset of courtship, the specimen cup was lifted from
the female. The presence and removal of the specimen cup did
not appear to alter the behavior of the spiders, and it ensured
that the female did not have access to the males until both males
courted. For each triad, the identity of the first male to court
and the identity of the male that successfully copulated were
recorded.
b. The second experiment tested for the possibility that males
require an extended period of time together to establish a dom-
inance hierarchy based on the presence of tufts. This experiment
was conducted in much the same way as the first experiment (1a)
except that the paired males were placed in the circular culture
dishes, with crickets and a water vial, approximately 44 h before
the start of the trials. It was not possible to get both males to
initiate courtship before exposing the female to them; the addi-
tion of a female, a cage liner, and a specimen cup to their cul-
ture dish resulted in escape behaviors, and removing them to
another arena risked destroying any site-specific dominance rela-
tions that might have been established. Therefore, females were
carefully placed in the male arenas at an equal distance from
both males. Again, the data recorded included the identity of
the first male to court in each triadic mating trial as well as the
identity of the male that attained the copulation.
Experiment 2. Relative importance of visual vs vibratory
communication during courtship
Specialized arenas were used to isolate experimentally the vibratory
and visual components of courtship. A male and a female spider
were placed in adjacent rectangular arenas (12 ×6×6cm) that were
separated by ca. 1 cm. Each arena consisted of four sides made
from 0.08-mm thick clear acetate; the bottom was open so that the
spiders were in contact with a cardboard substratum. The dimen-
sions of the cardboard varied with treatment. In the complete-
courtship trials, visual signals were clearly visible through the clear
plastic sides of the arenas while vibratory signals traveled through
the single piece of cardboard that held both the male and the female
arenas. In the vibratory-only trials the insertion of a piece of opaque
paper between the male and female arenas prevented perception of
visual signalling, while vibratory communication remained
unaffected. Finally, in the visual-only trials, the arena for each
sex was located on a separate piece of foam-insulated cardboard
thereby preventing the transmission of vibratory signals while still
permitting visual signalling. Only intact males were used in this
experiment.
For each trial, a female was introduced into the female com-
partment and given 5–10 min to acclimate. A piece of female cage
liner (a rectangular card bearing female draglines and silk-bound
pheromone) was placed on top of the cardboard in the male arena
to ensure vigorous courtship. When the male was introduced, video
recording equipment was turned on, and the interaction was video-
taped. The sequence of the trials was determined by a randomized
block design. Each individual was used only once.
The stereotypic receptivity turn of females (Montgomery 1903;
Uetz and Denterlein 1979; Stratton and Uetz 1981) was taken as
the sole indication of female receptivity. Previous studies (Uetz and
Denterlein 1979; Stratton and Uetz 1981, 1983; and triadic mating
trials reported here) have shown that when females are receptive,
this display is usually seen (and copulation proceeds) within 4 min
of male-female interaction. We analyzed the first 5 min of each
taped interaction for female receptivity to male signals.
Experiment 3. Relative importance of tufts within courtship
components
In this experiment, shaved and intact males courted within the
visual-only and complete-courtship arenas described above. Since
the tufts are a visual characteristic, the vibratory-only arena was
not used. Experimental protocol and data analyses were the same
as that described above, and the two experiments were conducted
together as part of the same randomized block design.
Experiment 4. Signal transmission in the field
Sound attenuation was measured as a function of distance from
point of origin in both the upland leaf litter habitat of S. ocreata
(Cincinnati Nature Center, Clermont County, Ohio) and in the
floodplain leaf litter habitat of S. rovneri (along the Ohio River
near Taylorsport, Boone County, Ky.). A Bruel and Kjaer
accelerometer (type 4399) and Sound Level meter were used to mea-
sure the sound level (dB) produced by dropping a standard object
(pencil) from a consistent height (5 cm) at 5-cm intervals along tran-
sects in four directions (N, S, E, W) from the point of impact in
both environments (n= 3 recordings at each distance).
The attenuation of the courtship signals produced by live male
spiders of each species was measured as above in both the floodplain
19
habitat of S. rovneri and the upland habitat of S. ocreata. Sound
transmission was measured at 5-cm intervals along transects in four
directions (N, S, E, W) from a courting male (n= 5 to 10 record-
ings at each point).
Results
Influence of the tufts on male mating success
(experiment 1)
In triadic mating experiments, when pairs of shaved
and intact males were introduced to females, the two
types of males were equally likely to attain copulation
(Table 1). This was true regardless of whether or not
the males had been confined together previously
(G= 0.106, df = 1, P> 0.5). In experiment 1b, in which
the males had prior contact and the females were intro-
duced before the males were courting, 79% (11 of 14)
of females mated with the first male to court (v2= 4.57,
df = 1, P< 0.05). Since shaved and tufted males were
equally likely to be the first to court (53% of first males
to court were shaved while 47% were tufted), the female
bias toward the first male does not alter the results of
Table 1.
Relative importance of visual vs vibratory
communication during courtship (experiment 2)
In the arena allowing the transmission of both vibra-
tory and visual signals, 14 of 15 (93.3%) of the females
gave receptivity responses to the male. In the vibratory-
alone arena, 9 of 13 (69.2%) of the females responded
positively, while in the visual-alone arena 9 of 14
(64.3%) of the females gave receptive responses
(Table 2a). Although female receptivity was lower in
the arenas restricting visual or vibrational signalling
than in the complete courtship arena, these differences
were not significant (Table 2a, G= 4.46, df =2,
P> 0.1).
A comparison of data showing the relative impor-
tance of visual and vibratory communication in S. ocre-
ata from this study with published results of a similar
study for S. rovneri (Uetz and Stratton 1982) reveals a
pronounced difference in the importance of visual sig-
nals (Table 2b). A three-way log-linear contingency
analysis shows significant non-independence overall
(G= 16.87, df = 7, P< 0.05), and a significant inter-
action between species and receptivity based on stim-
ulus (G= 16.57, df = 5, P< 0.001). While there is no
significant difference in the response of female S. ocre-
ata to conspecific males courting in the different are-
nas (G= 4.46, df = 2, P> 0.1), the response of female
S. rovneri to conspecific males courting in the visual
arena is significantly less than the responses given to
males in the vibratory or the complete courtship are-
nas (G= 16.64, df = 2, P< 0.001).
Relative importance of tufts within courtship
components (experiment 3)
A three-way log-linear contingency analysis of female
responses to shaved vs intact males shows that the
response depends on the stimulus components avail-
able (overall G= 25.32, df = 4, P< 0.001). When vibra-
tory and visual signalling were permitted, the female
response to shaved males did not differ significantly
from the response to tufted males (Table 3, G= 0.46,
df = 1, P> 0.3). However, when only visual signalling
was possible, the female response to shaved males was
significantly lower than the response to tufted males
(Table 3, G= 7.79, df = 1, P< 0.01).
Signal transmission in the field (experiment 4)
Although the upland (complex) leaf litter microhabi-
tat of S. ocreata and the floodplain (compressed) leaf
litter microhabitat of S. rovneri often occur within the
same deciduous forests, they show a considerable
difference in the transmission of substratum-conducted
vibration (Fig. 2). Sound levels (measured as decibels)
20
Table 1 The number of S. ocreata females that mated (+) with
either the tufted or the shaved male when given a choice between
paired males. In the two experiments reported here, the male pairs
had either (a) no previous contact, or (b) were confined together
for 44 h prior to introduction of the female
Previous contact
between males Male
Tufted Shaved
(a) None (+) 8 11
% receptive 42.1 57.9
(b) 44 h (+) 9 10
% receptive 47.4 52.6
G= 0.106, df = 1, P> 0.5
Table 2 The number of S. ocreata and S. rovneri females that
responded with a receptivity display (+) or did not display recep-
tivity ([) to the various components of conspecific courtship
Courtship components
Female Vibratory Visual Complete
Species receptivity only only
(a) S. ocreata (+) 9 9 14
([)451
% receptive 69.2 64.3 93.3
(b) S. rovneri (+) 15 6 16
([) 4 10 2
% receptive 78.9 37.5 88.9
(a) S. ocreata, G= 4.46, df = 2, P> 0.1; (b) S. rovneri data from
Uetz and Stratton (1982), G= 16.64, df = 2, P< 0.005
from impact vibration produced by a standard test
weight (pencil drop) are lower at the point of origin
(impact absorption) and attenuate much more rapidly
in the upland leaf litter microhabitat of S. ocreata than
in the floodplain microhabitat of S. rovneri.
Courtship vibrations in the upland deciduous litter
microhabitat of S. ocreata were so rapidly attenuated
as to be undetectable beyond the leaf holding the court-
ing spider (< 10 cm), and therefore we could only use
data for the floodplain microhabitat. Attenuation of
the courtship signals of live male spiders from each
species, measured in this same microhabitat type (com-
pressed floodplain leaf litter) indicate that sound level
produced by substratum-conducted courtship vibra-
tions of the two species are also very different (Stratton
and Uetz 1983, 1986). The vibratory signals of S. ocre-
ata are not as loud, and travel a shorter distance than
do signals of S. rovneri. We estimated the efficacy of
signal transmission for each species in its own habitat,
using the information on vibration attenuation from
each microhabitat as a scaling factor applied to the
data from signal transmission of both species in the
floodplain microhabitat (Fig. 3).
Discussion
In S. ocreata the presence or absence of a conspicuous
male secondary sexual characteristic, foreleg tufts, does
not appear to be involved in an obvious way with the
current operation of either female mate choice or male
mating competition. When given a choice, females were
most likely to mate with the first male that captured
their attention, regardless of the condition of his tufts.
This result is similar to that seen for salticid spiders
(Clark and Uetz 1992), and reinforces the argument
that if a bias toward the first male to court is a com-
mon feature of female behavior, caution should be used
when analyzing the results of simultaneous choice
experiments (Clark and Uetz 1992). However, respon-
ses of females to male movement may reflect sexual
selection based on sensory biases (Guilford and
Dawkins 1991; Endler 1992; Fleishman 1992; Ryan and
Rand 1993).
The results of the triadic mating experiments need
not mean that the tufts are unimportant in courtship
of S. ocreata, nor does it mean that they play no role
in sexual selection. In the absence of vibratory sig-
nalling, removal of the tufts from males resulted in
significantly fewer receptive responses from females.
When vibratory signals are transmitted to the female,
as in the triadic mating experiments, they are sufficient
21
Table 3 The number of S. ocreata females that displayed receptiv-
ity (+) or did not display receptivity ([) to shaved or tufted males
Male treatment
Courtship Female
components receptivity Shaved Tufted
(a) Visual only (+) 2 9
([)125
% receptive 14.3 64.3
(b) Complete (+) 12 14
([)21
% receptive 85.7 93.3
Courtship took place (a) within the visual courtship arena
(G= 7.79, df = 1, P< 0.01); or (b) within the complete courtship
arena (G= 0.46, df = 1, P> 0.3).
Fig. 2 Field transmission of vibrations of a standard test weight
(pencil) dropped along a transect in two deciduous forest habitats.
Each point represents the mean of 12 recordings (3 each from 4
directions from the point of origin) at each distance (± 2 SE)
Fig. 3 Estimation of the efficacy of transmission of vibratory
courtship signals of S. ocreata and S. rovneri in their respective
habitats, from data taken from live male spiders (n= 5–10 record-
ings per spider at each point) in the same microhabitat (compressed
floodplain litter) and scaled by the% attenuation of a test weight
in the particular microhabitat (Fig. 2). Female visual range esti-
mate (arrow) based on predatory responses toward visual cues from
prey at varying distances (Uetz, unpublished data)
to elicit a female response and any effect of the tufts
may be overridden. We did not conduct an explicit
analysis of the function of the tufts in terms of a dom-
inance hierarchy among males (see Aspey 1977a, b),
but such a hierarchy would only be important in intra-
sexual selection if it translated into differential mating
success. The experiment giving males 44 h to interact
prior to the introduction of a female was designed to
test for this effect, and revealed no difference.
Intersexual selection seems a more likely explanation
for the presence of tufts, given the behavioral and mor-
phological differences between species.
The conspicuous tufts present in male S. ocreata
appear to be important for successful courtship under
circumstances in which vibratory signalling is less
effective or not possible. The high frequency of
responses of female S. ocreata to the visual component
of male courtship may be partially explained by the
presence of the male tufts, or alternatively, may be the
selective agent favoring their evolution (Ryan and Rand
1990, 1993; Basolo 1990; Proctor 1991, 1992; Endler
1992; Fleishman 1992). Evidence from experiments iso-
lating the visual component of courtship suggests that
the tufts influence the efficacy or detectability of male
visual signalling (see Guilford and Dawkins 1991;
Hasson 1991) under environmental conditions that
constrain vibratory communication. Our estimates of
male signalling efficacy (Fig. 3) suggest that for male
S. rovneri, vibrational signalling via substratum-cou-
pled stridulation (Rovner 1975) is an effective means
of communicating, even over what must be consider-
able distances to a 1-cm spider. In contrast, the vibra-
tional signal of S. ocreata does not travel very far at
all in the upland habitat. (Our estimate of 10 cm, based
on transmission in compressed floodplain litter, is prob-
ably an overestimate, since in the upland leaf litter habi-
tat we were unable to detect spider vibrations beyond
the leaf with which the spider was in contact.)
Female Schizocosa are slightly larger than males and
feed on a variety of arthropods including conspecifics.
Therefore, it may be essential for males to communi-
cate their identity effectively. Females of both
Schizocosa species react to moving prey stimuli with
typical predatory responses (orient, stalk, lunge) at dis-
tances up to 30 cm and beyond (Rennecker, Valerius,
Cook and Uetz, unpublished), similar to other lycosid
species (Land 1985). Because the vibratory signals of
male S. rovneri in the floodplain habitat are transmit-
ted approximately this distance (Fig. 3), there is no
range within which a courting male is visible to a female
but not heard. In contrast, however, a courting S. ocre-
ata male may come into the visual range of a female
15 to 20 cm before she can detect his vibratory signals.
Visual signals may thus be critically important in mate
identification when the male is within the visual range
of the female but not yet within the range of effective
vibratory communication. If tufts increase the chances
that the male will be detected and recognized by the
female, resulting in receptivity rather than predatory
behavior, there is potential for a strong selective advan-
tage – both in terms of sexual selection by increased
mating success because of more efficient courtship com-
munication, and in terms of natural selection by a
reduction in the probability of predation. Further
experiments will investigate whether the presence of the
tufts influences female predatory behavior when the
spiders are vibrationally isolated.
Based on the results of this study, we hypothesize
that the conspicuous tufts of male S. ocreata function
to increase the efficacy of courtship communication (see
Guilford and Dawkins 1991) in an environment that
physically constrains the more common mode of vibra-
tional signalling. The complex leaf litter habitat of
S. ocreata may create an important physical constraint
on the effectiveness of vibrational signalling; success-
ful courtship communication may require an enhanced
contribution from the visual component of male
courtship. This appears to have been effected behav-
iorally – by an increased level of courtship movement
(especially of the forelegs) – as well as morphologically
– by the conspicuous male foreleg pigmentation and
tufts. The distribution of male morphology and behav-
ior within the genus Schizocosa is consistent with this
hypothesis. Although the presence of male foreleg tufts
is uncommon within the genus (Dondale and Redner
1978), leg-waving visual displays are an important com-
ponent of courtship in species with tufts or other dec-
orations; such behaviors appear to be absent in species
that lack tufts (Stratton and Lowrie 1984), a trend that
is seen in other lycosid genera as well as in Schizocosa
(E. Hebets, pers. commun.).
Because a complete phylogeny is not yet available
for Schizocosa, we can only speculate as to the evolu-
tionary origin of the tufts and visual signals. Given the
vibrational signalling biology of most lycosid spiders
and the distribution of the tufts within the genus, it is
very possible that the tufts represent an apomorphic
feature of S. ocreata (McClintock and Uetz 1995). If
this is indeed the case, it is possible that the tufts have
arisen as a means of exploiting some pre-existing visual
bias in female wolf spiders, in the manner suggested by
Ryan et al. (1990), Basolo (1990, 1995), and Proctor
(1991, 1992) for other taxa.
Acknowledgements This research was supported in part by a Sigma
Xi Grant-in-Aid of Research (to S.J.S.), and a grant from the
Hewlett-Mellon Fund of Albion College (to G.E.S.). The Cincinnati
Nature Center kindly allowed us to collect spiders from their
grounds. We thank Karen Ausdenmoore and Keith Karoly for
assistance with the spiders. Doug Futuyma, Leo Shapiro, Kevin
Omland, and students in the Uetz laboratory provided valuable
comments on the manuscript.
References
Andersson M (1995) Sexual selection. Princeton University Press,
Princeton
22
Aspey WP (1977a) Wolf spider sociobiology. I. Agonistic display
and dominance-subordinance relations in adult male Schizocosa
crassipes. Behavior 62:103–141
Aspey WP (1977b) Wolf spider sociobiology. II. Density parame-
ters influencing agonistic behavior in Schizocosa crassipes.
Behavior 62:142–163
Basolo AL (1990) Female preference predates the preference of the
sword in swordtail fish. Science 250:808–810
Basolo AL (1995) Phylogenetic evidence for the role of a pre-exist-
ing bias in sexual selection. Proc R Soc Lond [Biol] 259:307–311
Clark DL, Uetz GW (1992) Morph-independent mate selection in
a dimorphic jumping spider: demonstration of movement bias
in female choice using video-controlled courtship behaviour.
Anim Behav 43:247–254
Dondale CD, Redner JH (1978) Revision of the nearctic wolf
spider genus Schizocosa (Araneida: Lycosidae). Can Ent 110:
143–181
Endler JA (1980) Natural selection on color patterns in Poecilia
reticulata. Evolution 34:76–91
Endler JA (1992) Signals, signal conditions, and the direction of
evolution. Am Nat [Suppl] 139:125–153
Fleishman LJ (1992) The influence of the sensory system and the
environment on motion patterns in the visual displays of ano-
line lizards and other vertebrates. Am Nat [Suppl] 139:S36–S61
Guilford T, Dawkins M (1991) Receiver psychology and the evo-
lution of animal signals. Anim Behav 42:1–14
Hasson O (1991) Sexual displays as amplifiers: practical examples
with an emphasis on feather decorations. Behav Ecol 2:189–197
Hunter ML, Krebs JR (1979) Geographical variation in the song
of the great tit (Parus major) in relation to ecological factors.
J Anim Ecol 48:759–785
Krebs JR, Davies NB (1993) An introduction to behavioral
ecology, 3rd edn. Blackwell Scientific Publications, Oxford
Land MF (1985) The morphology and optics of spider eyes. In:
Bartk FG (ed) Neurobiology of Arachmids. Springer, Berlin
Heidelberg New York, pp 53–78
McClintock WJ, Uetz GW (1996) Female mate choice in two wolf
spiders (Araneae: Lycosidae): pre-existing bias or preference for
a novel trait. Anim Behav (in press)
Montgomery TH Jr (1903) Studies on the habits of spiders, par-
ticularly those of the mating period. Proc Acad Nat Sci Phil
55:59–149
Morton ES (1975) Ecological sources of selection on avian sounds.
Am Nat 109:17–34
Proctor HC (1991) Courtship in the water mite Neumania papilla-
tor: males capitalize on female adaptations for predation. Anim
Behav 42:589–598
Proctor HC (1992) Sensory exploitation and the evolution of male
mating behaviour: a cladistic test using water mites (Acari:
Parasitengona). Anim Behav 44:745–752
Rovner JS (1967) Acoustic communication in a lycosid spider
(Lycosa rabida Walckenaer). Anim Behav 15:273–281
Rovner JS (1975) Sound production by nearctic wolf spiders:
a substratum-coupled stridulatory mechanism. Science 190:
1309–1310
Ryan MJ, Rand AS (1990) The sensory basis of sexual selection
for complex calls in the tungara frog, Physalaemus pustulosus
(sexual selection for sensory exploitation). Evolution
44:305–314
Ryan MJ, Rand AS (1993) Species recognition and sexual selection
as a unitary problem in animal communication. Evolution 47:
647–657
Ryan MJ, Fox JH, Wilczynski W, Rand AS (1990) Sexual selec-
tion for sensory exploitation in the frog Physalaemus pustulo-
sus. Nature 139:66–67
Stratton GE (1982) Reproductive behavior and behavior genetics
in Schizocosa wolf spiders. Ph.D. Dissertation, University of
Cincinnati
Stratton GE, Lowrie DC (1984) Courtship behavior and life cycle
of the wolf spider Schizocosa mccooki (Araneae: Lycosidae).
J Arachnol 12:223–228
Stratton GE, Uetz GW (1981) Acoustic communication and repro-
ductive isolation in two species of wolf spiders. Science
214:575–577
Stratton GE, Uetz GW (1983) Communication via substratum-
coupled stridulation and reproductive isolation in wolf spiders
(Araneae: Lycosidae). Anim Behav 31:164–172
Stratton GE, Uetz GW (1986) The inheritance of courtship behav-
ior and its role as a reproductive isolating mechanism in two
species of Schizocosa wolf spiders (Araneae: Lycosidae).
Evolution 40:129–141
Uetz GW, Denterlein G (1979) Courtship behavior, habitat, and
reproductive isolation in Schizocosa rovneri Uetz and Dondale
(Araneae: Lycosidae). J Arachnol 7:121–128
Uetz GW, Dondale CD (1979) A new wolf spider in the genus
Schizocosa (Araneae: Lycosidae) from Illinois. J Arachnol 7:
86–88
Uetz GW, Stratton GE (1982) Acoustic communication and repro-
ductive isolation in spiders. In: Witt PN, Rovner JS (eds) Spider
communication: mechanisms and ecological significance.
Princeton University Press. Princeton, pp 123–159
Communicated by M. A. Elgar
23
