Behavioral Ecology Vol. 12 No. 3: 360-366
© 2001 International Society for Behavioral Ecology
Pick-up lines: cues used by male crab spiders to find reproductive females
Box G-W, Department of Ecology and Evolutionary Biology, Brown University, Providence, RI 02912, USA
Address correspondence to D.H. Morse. E-mail: d_morse{at}brown.edu .
Received 10 January 2000; revised 15 September 2000; accepted 8 October 2000.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The low population density of the crab spider Misumena vatia and the high percentage of gravid adult females begs the question, how do adult males find reproductive females? We explored one of the potential mate-locating tactics of adult males, their line-following behavior. M. vatia do not build webs; however, they do lay down silken lines during their movements. In both the field and the laboratory, adult males (but not juvenile males) regularly followed lines of other individuals, as well as their own, sometimes multiple times. However, they did not distinguish the direction in which lines were laid, and they even followed lines of other species. Males responded mechanically to these lines, but not chemically. They followed lines of penultimate and adult virgin females, as well as new and old lines, with similar high frequency. Neither washing the lines with water nor washing them with acetone (to remove possible water-soluble or organically soluble pheromones) affected their choice of lines. Due to the low activity of females, their lines may be several days old, over which time any possible chemical information would likely dissipate. Therefore, pheromones seem unlikely to aid males searching for unmated adult females. Nevertheless, even searching males that follow lines largely indiscriminately should derive a benefit because concentrations of lines could denote the presence of females and thereby maximize numbers of females encountered.
Key words: crab spider, mate choice, mate finding, Misumena vatia, pheromones, random searching, silk lines.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Darwin (1871
) was the first
to recognize that complex interactions between the sexes drove selection on
individuals in a way distinct from natural selection. Attention initially
centered on mate choice, including courtship behavior and aggression, in
elucidating issues of reproductive control, parental investment, and fitness.
However, an equally critical issue for species with low population densities
or strongly biased operational sex ratios is the search for a mate. Although
random search (Koopman, 1956;
Stone, 1975) would apply in
the simplest instance, more effective modes of mate finding should enjoy a
strong selective advantage. For instance, in the leafhopper Graminella
nigrifrons, reproductively active adult males participate in time and
energy-intensive "call-fly" search patterns, whereby they jump or
fly from plant to plant and call in order to locate virgin adult females
(Hunt and Nault, 1991). When
individuals rarely encounter potential partners, selection may shift to favor
the early, the swift, or the experienced
(Ghiselin, 1974;
LeGrand and Morse, 2000).
Although mate choice may still occur, choosiness is projected to decrease
along with intrasexual interactions in models presented by Parker
(1978,
1979,
1983) and Real
(1990). And, if a member of
the choosy sex faces a strong probability of encountering only a single
prospective mate, or if a considerable period is projected to pass before
meeting the first potential mate, the first searcher to find a member of the
choosy sex may routinely be accepted. Parker and Real's models were novel in
that they incorporated mate search in the costs of mate choice.
Low-density populations of wandering spiders make excellent subjects for
investigating the relationship between mate choice and encounter rate. Upon
molting into the adult stage, the primary goal of male spiders shifts from
hunting for food (as in previous instars) to searching for females. Male crab
spiders (Misumena vatia: Thomisidae) fit these criteria well. During
their adult stage they spend the majority of their time searching for mates,
with the most-fit males likely to be those that find the most unfertilized
adult females (LeGrand and Morse,
2000). Female M. vatia do not appear to search actively
for males, as suggested by their residence at satisfactory hunting sites for
many days at a time (Morse and Fritz,
1982), nor do they advertise for males
(Holdsworth and Morse, 2000).
This system is somewhat analogous to the behavior of the leafhoppers, in that
males must wander in search of sedentary females, and the males who locate
more females will sire more offspring
(Hunt and Nault, 1991).
One way to ensure paternity is to guard pre-reproductive females
(Parker, 1974;
Ridley, 1983). Yet male M.
vatia guard only a minority of the females before the females molt into
the adult stage. Many females thus do not become mated immediately after
reaching sexual maturity (Holdsworth and
Morse, 2000); nevertheless, adult males are proficient at locating
females, as evinced by the observation that most adult females are eventually
mated (LeGrand and Morse,
2000).
Adult male M. vatia might find unmated adult females in several
ways: (1) "Brownian motion" (i.e., males randomly bump into
females), (2) visual cues supplied by the physical presence of a female, (3)
airborne pheromones released by females and detected by males
(Blanke, 1975;
Watson, 1991), (4) following
female lines when encountered and potentially detecting pheromones deposited
on these lines (Tietjen and Rovner,
1982), (5) systematic searching, in which the males thoroughly
explore substrates (such as walking up and down a stem and circling around
flowers or inflorescences), and (6) concentrating activity and time on flowers
that are preferred by adult females (Chien
and Morse, 1998; LeGrand and
Morse, 2000). These options are not mutually exclusive; indeed,
some or all of them might constitute a comprehensive mate-finding strategy.
However, some of our previous research has eliminated the possibility that an
airborne molting or sex pheromone (option 3) operates in this context
(LeGrand and Morse, 2000).
Additionally, Holdsworth and Morse
(2000) have shown that males
seldom respond to the presence of females more than approximately 5 cm from
them, roughly the distance from the opposite sides of large flowers,
eliminating option 2. Beyond this range males do not clearly detect the
presence of females. Therefore, possible visual or olfactory cues to the
location of a female probably can only operate when the males and females are
in close proximity. In this regard, they approximate Brownian motion (option
1). Chien and Morse (1998) have
shown, however, that adult males exhibit flower choices similar to those of
adult females, which frequently position them in sites favored by females
quite independently of the females' choices (option 6). Options 1 and 5 are
both concerned with instances in which male spiders do not use cues emitted by
females or cues otherwise associated by the presence of females.
We have focused on the putative trail-following behavior of adult males
(option 4) here; however, our results provide further insights for options 1-3
and 5 and 6. The majority of spider species leave silken lines behind them
during each movement (Foelix,
1996), and M. vatia is no exception. M. vatia
also produce lines that extend between plants, greatly increasing their rate
of movement through the top of the vegetation.
Many spider species communicate via their lines. In some species, wanderers
and web weavers alike, females emit chemical pheromones on their lines
(Tietjen and Rovner, 1982),
which males follow. Such line-following behavior could potentially provide
male M. vatia with information about the reproductive status of the
female that produced the line and/or the direction in which the female was
moving. At the very least, the male would be likely to increase his
probability of finding a female by following the line, rather than moving
randomly. In this study we investigated male responses to both mechanical and
chemical properties of lines, which age classes of males follow lines, and the
line-following behaviors of males in the field. Finally, we evaluated the
mate-finding abilities of adult male M. vatia as they relate to the
points noted above.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Study species
Misumena vatia is a semelparous, sit-and-wait predator with a Holarctic distribution (Gertsch, 1939
; Morse and Stephens,
1996); this species does not weave webs. M. vatia is
protandrous; males molt into the adult stage in May and early June, with
numbers of adult males peaking from 5 June to 15 July, whereas females molt
into the adult stage in mid- to late June, peaking around 25 June
(LeGrand and Morse, 2000).
During the early summer, adult females hunt in the upper stratum of the field
vegetation, primarily on the flowers of ox-eye daisy (Chrysanthemum
leucanthemum), red clover (Trifolium pratense) and buttercups
(Ranunculus acris). Concurrently, the adult males search out these
females. The population density of these spiders is relatively low; in 2
successive years Holdsworth and Morse
(2000) recorded 138 females
and 54 males/ha, and 206 females and 40 males/ha. Additionally,M.
vatia's operational sex ratio (OSR) shifts over the season from male
biased to female biased and back to male biased
(LeGrand and Morse, 2000).
After molting into the adult stage, males do not increase their body mass,
which averages about 4 mg for both of the final instars
(Chien and Morse, 1998).
However, their body morphology does change after the molt from the penultimate
to adult stage; the forelimbs become longer and the abdomen smaller
(LeGrand and Morse, 2000).
Although adult males feed, their primary activity is finding females. Female
M. vatia reach a maximum length of 9-12 mm and a mass of 200-400 mg
(Fritz and Morse, 1985;
LeGrand and Morse, 2000).
Care of the spiders
We kept the spiders in 7-dram vials (5 cm high, 3 cm diam), feeding them a
diet of flies, mosquitoes, moths, and other small insects every other night.
During the day, when they were not being tested, we kept them outside in the
shade so they would not dehydrate.
Site description
We conducted the field studies at the Darling Marine Center of the
University of Maine, South Bristol, Lincoln County, Maine, USA, in a field of
3.5 ha surrounded by mixed coniferous—deciduous forest
(Morse, 1999). The field
contains several grasses (Gramineae), with a wide variety of forbs scattered
throughout. During late spring and early summer, ox-eye daisy, common
buttercup, and red clover are among the commonest and most widespread large
flowering forbs in this field. The field is mowed yearly in October and thus
kept in an early successional stage.
Field experiments and observations
Trials with female M. vatia: persistence of female lines
To assess the integrity of female lines, we induced 10 adult females to
produce lines between the stems of ox-eye daisies. Because M. vatia
almost always lay down lines when moving, we could easily induce them to
produce lines by placing them on a substrate, then slowly picking them up and
moving them toward the desired opposite end of the line. By looping this line
around the substrate at the desired end of the line it was possible to produce
lines of up to 0.5-1.0 m with regularity, although we worked primarily with
lines of 5-20 cm. Females produced one to four usable lines (mean = 2.2
± 1.0 SD) in this way. We removed the females after they produced two
to four lines (mean = 2.5 ± 0.7 SD) connecting different daisies. We
created these lines at approximately 1500 h, recorded the lengths of each
line, and monitored them at 1000 h and 1620 h of the next day to determine the
number still present.
We also tested lines in the field during windy periods, a time when they were subject to rapid damage. Windy weather usually occurred on the first days of high pressure fronts, and in the study area routinely resulted in winds of 5-25 cm/s, speeds seldom otherwise equaled [occasionally by thunder-showers accompanied by high winds or by major coastal storms (northeasters)]. On average, such winds occurred once every several days.
Using the same technique described above, we strung lines ranging from 5 to 20 cm between daisy stems. We also compared these artificially generated lines with naturally-produced lines of the same individuals to establish whether this treatment affected the tendency of these lines to break.
Adult male activity on grasses
We measured the rates at which adult males traveled in the field when the
weather was sunny with negligible wind. We released these males on grasses
away from flowers because they move more rapidly on this substrate than on
flowering forbs, upon which they may spend considerable time hunting (Anderson
JT, personal observation).
During these trials, we followed the movements of a male for 30 min or until we lost sight of him. We recorded the distance the males moved on substrates and on their own lines, as well as the number of times they assumed hunting position, released lines, or performed other conspicuous behaviors. The grass in this open-field environment ranged between 0.5 and 1.5 m tall, and many of the grass stems bore terminal inflorescences.
Adult male behavior near female lines
We collected adult females (either gravid or virgin) to spin lines between
blades of grass. Subsequently, we released a male 2 cm below the line on the
blade of grass where the female's line commenced. We tested each male with
lines from different females to control for any individual variation. We
recorded whether a male followed a line the female had produced or whether
that line was one of his. We then compared the activity levels of adult males
who traveled without the aid of female lines with that of adult males who
followed female lines.
We induced adult females to produce lines between fresh, healthy ox-eye daisy inflorescences and senescing ox-eye daisy inflorescences that attracted few insects and spiders. The senescent daisies still retained a few partly shriveled ray flowers. We allowed the female to remain on the senesced plant, while we introduced a male to the stem of the healthy plant, immediately below the flower head. This experiment allowed us to test whether the males responded primarily to a female line (option 4) or to a high-quality flower (option 5) at this distance. We recorded the time of each male movement, whether he followed the female line, the number of times he followed the lines, and details of any interactions between the males and the females. We ended the trial after the males and females interacted or after the males left the original plant substrate.
Penultimate male activity
As penultimate males appeared in August, we tested whether they would
follow female lines in the field. We induced handreared, virgin adult females
to produce a line between two grass stems. We then released a penultimate male
2 cm below her line, ending the trial when the male followed the female line
or the male left the grass stem. We recorded the time and distance of any
movement, as well as whether the male followed a line he had produced or one
the female had produced. Different females provided the lines for each of the
different trials.
We tested penultimate male activity levels in the absence of female lines
in the same way that we tested those of adult males. Juveniles of both sexes
spend the majority of their time hunting on goldenrod (Solidago spp.)
in August, probably a consequence of the high density of prey on these plants
(Morse, 1981). Therefore, it
was essential to conduct this experiment 1 m or more away from goldenrod to
ensure that the penultimate males did not simply move toward the
goldenrod.
Laboratory studies with adult males
We studied the behavior of adult male spiders in a large room with several
windows that provided ample light and cross-currents. We placed five daisy
stems equidistant on a 30 x 20 cm surface
(Figure 1). We ran lines among
only three of the stems (here denoted as A, B, and C); however, we also
provided the spiders with stems D and E so that the males would have the
opportunity to travel to plant substrates without using lines made by other
individuals. Males readily hunt on ox-eye daisies
(Chien and Morse, 1998);
therefore, we removed the flower heads from these stems to increase the
activity levels of the males by preventing them from assuming hunting
positions on the flower heads. We oriented each of these surfaces randomly to
eliminate any possible effects of location or direction. We either wiped the
stems clean or changed stems after each trial.
|
We released a female on either stem A or stem C (see Figure 1). We then placed stem B adjacent to the female, and after she crawled onto it, we moved this stem back to its original position. We repeated this protocol with the last stem and then removed the female. In this manner, the females produced a line of silk that stretched from A to B to C, or vice versa; we recorded the direction in which the silk line was stretched. We then released a male 2 cm below where the female silk connected to stem B. We recorded his behavior, including time to first move, whether he contacted the female line (and, if so, time until contact), whether he followed the female line, his direction of travel, and other information specific to the particular test. We ended a trial either 5 min after the male moved from stem B to any other location (in order to record information on subsequent moves) or after 20 min, if he remained sedentary.
Responses to modified lines
We tested male spiders' responses to (1) directionality (did the male move
in the same direction as the female?) and (2) age class of the line producer
(did the male follow lines of both penultimate and adult female spiders?). The
design of this experiment differed from the general description above only in
that we tested one set of male spiders on penultimate female lines and a
second set on virgin, adult female lines.
Responses to pheromones
The second experiment tested whether males responded to water-soluble or
organically soluble pheromones on female lines. During the evening before this
experiment we constructed setups with both penultimate female lines and with
adult female lines. Then, to remove any water-soluble chemicals, we used a
dropper to dispense water along the length of each line. On the following
morning we ran the males on these lines in the same way as previously
described. We then repeated this protocol substituting acetone for water to
determine whether males responded to organically soluble pheromones on female
lines (see Pollard et al.,
1987, for information about washing lines with polar [such as
water] and nonpolar [organic] solvents in reference to studying sexual
pheromones).
Choice of lines
We constructed additional setups with penultimate female lines and adult
female lines to establish whether male spiders would distinguish between
unmodified and washed lines. Lines to be washed, constructed on the evenings
before tests, only ran from stem B to stem A or from stem B to stem C; that
is, they did not span the length of the surface. We then washed each line with
water. On the following morning, immediately before any trial with males, we
induced the same females to produce lines between stem B and the vacant stem.
Therefore, attached to stem B was one washed line and one unwashed line. As
soon as a female had constructed a unwashed line, we removed her from the
setup and released a male 2 cm below the lines on stem B. We recorded whether
the male contacted a line and/or followed it, as well as which line he
contacted first. Additionally, we noted the time until the male followed the
first line and whether he returned and followed the other line after he moved
off stem B.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Field experiments and observations
Trials with female M. vatia: persistence of female lines
After 19 h and a night of light rain (1.0 cm), one or more lines of 9 of the 10 females were still intact. Overall, 16 of the 25 lines (64%) we had induced females to produce the previous afternoon remained, and all of these lines remained intact 6 h later (midday).
However, during brisk winds these lines remained intact for only short periods. Twenty-three lines ranging from 5-20 cm and strung between stems or leaves, each produced by a different adult, virgin, female spider, remained for 1.4-120 min (mean ± SD = 22.4 ± 31.4 min). These lines broke as a result of the movement of two points of attachment on the vegetation against each other, although in dense vegetation other stems or leaves would be blown across them as well. The highly variable breakage times are the consequence of occasional gusts of wind (20+ m/s), which seemed responsible for all the breaks.
Brisk winds are not the norm in the study area, typically occurring every few days, usually associated with large high-pressure cells of clear and windy weather. No correlation existed between the length of a line and its survival time (linear regression, r2 =.08, df = 21, p >.1), likely the consequence of the primary importance of the gusts. These hand-drawn lines did not differ significantly from naturally-laid lines of the same individuals in their propensity to break under the conditions in which both sets of lines were tested (Wilcoxon matched-pairs, signed-ranks test, two-tailed, z = 1.306, n = 15, p >.1).
Adult male activity on grasses and behavior on female lines
We tested the response of 15 adult males to female lines; however, only 10
(67%) of the males followed female lines (followers). Followers remained in
the area the female had earlier occupied for 20.5 ± 8.5 min (mean
± SD), and four individuals remained in the area of the female lines
after 30 min. We combined the data for the five males that ignored these lines
with data for males not presented with female lines (nonfollowers). Thus, the
terms "follower" and "nonfollower" refer to the males'
behavior in the presence or absence of female lines only. Members of both of
these groups of males used their own lines. Because we sometimes lost the male
before 30 min, we express the results in rates (variable/h) to control for
slight time differences.
Nonfollower males (n = 12) released more lines (10.5 ± 7.2) than they eventually used (7.6 ± 7.8; Wilcoxon matched-pairs, signed-ranks test, two-tailed, z = 3.000, p <.01), suggesting that they ignored some of the lines they released. However, follower males (n = 10) used all but one of the female lines (lines present = 2.2 ± 1.0; lines followed = 2.1 ± 1.0; same test, z = 0.102, p >.9). Follower males thus used a higher proportion of female lines than nonfollower males used their own lines (Mann-Whitney U test, two-tailed, z = 3.561, p <.001).
For both of these groups of males, we calculated the difference between the number of times a male followed a line and the number of lines he followed. Although males generally released more lines than they followed (see above), we scored only "established" lines—either a female line, produced before the male arrived, or a line released by the male and followed at least once. Values for this parameter are (1) > 1: male followed the lines in his vicinity (either lines a female produced, or lines the male produced) multiple times; (2) = 1: male followed each line an average of one time; or (3) < 1: male did not follow all of the lines at least once. Follower males used female lines more often than nonfollower males used their own lines (follower males = 4.0 ± 4.3, nonfollower males = 0.8 ± 1.4; Mann-Whitney U test, two-tailed, z = 2.253, p <.05).
Rate of movement of adult males
The rate of movement (lines + substrate) of males on female lines
(followers) and males only on their own lines (nonfollowers) did not differ
(7.2 ± 7.8 m/h vs. 6.6 ± 5.4 m/h; Mann-Whitney U test,
two-tailed, n = 10, 12, z = 0.396, p >.7).
However, because most of the follower males eventually left the area of the
female lines, the rates of follower males before and after leaving the female
areas would provide a more sensitive comparison. We can compare six of the
individuals in this way. These males traveled faster in the vicinity of
females than they did away from females (13.2 ± 9.0 m/h vs. 7.2
± 6.6 m/h, Wilcoxon matched-pairs, signed-ranks test, two-tailed,
z = 2.240, p <.05).
Follower males spent less of their traveling time on their own lines than did the nonfollower males (19.5 ± 19.8% vs. 59.3 ± 40.0%; Mann-Whitney U test, two-tailed, on the original data, z = 2.394, p <.02). However, we could not discern whether males traveling on their own lines ended farther from the site of release than males who began their journey on female lines. The percentage of total distance traveled by follower males was higher on female lines than on their own lines (71.8 ± 23.1% vs. 19.5 ± 19.8%; Wilcoxon matched-pairs, signed-ranks test, two-tailed, on the original data, z = 2.599, p <.01).
Adult male behavior near adult female lines in flowers
Eight of nine males followed female lines off the original healthy plant to
the senesced plant (binomial test, two-tailed, n = 9, p
=.04). We have more detailed information for six of these individuals, all of
which followed the female lines. In three of these trials, the females
produced two lines and the respective three males followed both lines; in the
other three trials, the females produced only one line and the respective
three males each followed that line. Four of the six males encountered the
female; three pairs mated, but the female of the fourth pair refused to mate
and dropped on a line away from the male. In the other two instances, the
females had moved away from the senescing daisy before the males' arrival, and
we collected the males immediately after they followed the female lines;
therefore, the males had no opportunity to follow subsequent lines that the
females may have produced. In this experiment, the males only followed the
lines one time due to the presence of the female at the other side. In none of
the six runs did any male follow a line of his own to the senesced plant,
although five of the six males followed lines they had produced before moving
to the senesced plant.
Activity of penultimate males
Only 15% of the penultimate males followed female lines in the field,
whereas 67% of the adult males did so (see above); penultimate and adult male
behaviors differed significantly (G test on the original data,
G = 7.98, df = 1, n = 13, 15, p <.01). Only one
penultimate male followed an entire female line, although one other
penultimate male followed part way, stopped, and returned to the original
plant. Since 46% of the penultimates and 73% of adults made some sort of
contact with the female line, the majority of the penultimates seemed to be
eschewing female lines, either by not contacting them in the first place, or
not following them if they did come into contact. Adult males did not contact
foreign lines more frequently than penultimate males (same test, G =
1.05, df = 1, p >.2).
Conversely, 92% of the penultimate males followed a line they had produced, whereas only 67% of the adult males eventually followed a similarly self-produced line; however, this difference was not significant (same test, G = 2.95, df = 1, p >.05). Penultimate males and adults did not differ in the time required to leave the first plant substrate (9.1 ± 14.9 min vs. 6.5 ± 8.3 min, Mann-Whitney U test, two-tailed, z = 0.276, p >.7).
Penultimate and adult males did not differ significantly in rates of movement on grass stems in the open field (3.3 ± 2.9 m/h, n = 22 vs. 4.3 ± 4.1 m/h, n = 26, Mann-Whitney U test, two-tailed, z = 0.476, p >.3). Although adults averaged higher rates than penultimates, the high variance accounted for the lack of significant differences. However, most of the rapidly moving individuals were adults.
Laboratory trials with adult males
Responses to unmanipulated lines
All males contacted (touched) penultimate female lines (binomial test,
two-tailed, n = 13, p <.002), and all but one contacted
adult, virgin, female lines (same test, n = 14, p <.01).
The males also usually followed both penultimate (11 of 13, same test,
p <.05) and adult (12 of 14, same test, p <.02) female
lines. Males did not distinguish between penultimate and adult females in
either contacting or following lines (contacts: same test, n = 11 on
penultimate lines, p >.5; n = 12 on adult lines,
p >.7). Males also routinely followed lines of other adult male
M. vatia [all six tested by LeGrand and Morse
(2000), four of five males
that we tested (same test, n = 11, p =.02].
Lines washed with water
All males contacted both water-washed penultimate female lines (n
= 15) and water-washed adult female lines (n = 15; binomial tests,
two-tailed, p <.001, <.001]. All of these males also followed
these penultimate female lines, and all but one followed the adult female
lines (same tests, p <.001, <.001). Additionally, another five
males tested on both penultimate and adult female lines contacted and followed
each line presented to them.
Lines washed with acetone
All but one male contacted acetone-washed penultimate female lines
(n = 15), and all contacted acetone-washed adult female lines
(n = 15; binomial tests, two-tailed, p <.002, <.001].
All but one of the males that contacted the penultimate lines (n =
14) followed them, as did all but one (n = 15) contacting adult
female lines (same tests, p <.01, <.002).
Choice of lines
All but one male contacted at least one of the penultimate female lines in
the choice test (binomial test, two-tailed, n = 15, p
<.002), and all but two of these followed at least one line (same test,
p <.02). Similarly, all but one male contacted at least one of the
adult female lines (same test, p <.01), and all of these followed
at least one line (same test, p <.001). Of the 12 males that
followed a penultimate female line, 6 first followed the unwashed line and 6
first followed the washed line (same test, p >.9). Of the 14 males
that followed an adult line, seven first followed the unwashed line and seven
first followed the washed line (same test, p >.9). Nine of the 12
males following at least one penultimate female line eventually followed both
lines, as did 11 of the 14 that followed at least one adult female line. These
proportions did not differ significantly (G test, G = 0.05,
df = 1, p >.9).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Behavior of males
Although adult and penultimate males did not differ significantly in their rates of motion on grasses in the absence of females or flowers, the generally high activity level of adult males should aid in the search for reproductively active females. Wandering adult males of three ctenid spiders similarly capitalize on their high activity levels in finding females (Schmitt et al., 1990
). The
significant difference between adult and penultimate male M. vatia,
however, was that the majority of adult males followed lines of other spiders,
whereas most penultimate males did not. Adult males may accrue a benefit from
following these lines, or penultimate males may avoid a cost by not following
them. Penultimate males might eschew foreign lines because of the possibility
that the spider at the other end would prey upon them. However, because
penultimate and adult males do not differ in mass
(LeGrand and Morse, 2000), one
might expect the predation pressures upon them to be similar. Therefore,
potential rewards for line following should be higher for adult males than for
penultimate males. Although adult males were more inclined to follow foreign lines than were penultimate males, these adults were unable to distinguish between different types of silk lines. Males responded to lines of penultimate and adult females in a similar fashion; additionally, they followed lines that had been washed with both acetone and water, and they did not prefer unwashed lines to lines that had been washed with water. They also routinely followed lines of other adult male M. vatia and even followed the lines of other similarly-sized wandering spider species (Anderson JT, Morse DH, unpublished results). We thus suggest that adult males are simply responding to the mechanical stimulus of the lines, not to pheromones present on the lines.
Nonetheless, it is interesting to note that males that followed female
lines (followers) did so repetitively, whereas males that followed their own
lines (nonfollowers) tended to follow each line only once; this evidence
further emphasizes the likely importance of lines in locating females.
Additionally, followers explored nearly all of the female lines, whereas
nonfollowers ignored some self-produced lines. Follower males thoroughly
searched the area in which the female line was located and presumably
unequivocally determined whether the female was in the vicinity. When given
the choice between a healthy flower, which a female would find an attractive
hunting site, and a line stretching from that flower to a senesced flower,
adult males followed that line. This behavior suggests that a line activates
adult males more strongly than a favorable hunting site does. The majority of
males that followed female lines eventually left the vicinity of these lines,
however, consistent with a giving-up time
(Charnov, 1976), a trait that
operates in other parts of M. vatia's life history as well
(Kareiva et al., 1989).
Other workers (Pollard et al.,
1987; Tietjen,
1979) have reported that adult males of certain species follow
adult (and in certain instances, subadult) female lines, but not male or
heterospecific lines, which they interpreted as evidence of males responding
to pheromones on female lines. In light of accumulating evidence for the
importance of female pheromones in mate finding by male spiders, we ran the
washing experiments to test further our initial conclusion that male M.
vatia do not use olfactory cues in mate searching. Unlike males of other
species (e.g., Tietjen and Rovner,
1982), male M. vatia do not appear to respond to the
spider producing the line before encountering that spider. Male M.
vatia do not determine the direction in which that spider traveled, much
less its reproductive status, cues used by males of other species tested
(Pollard et al., 1987;
Tietjen and Rovner, 1982).
Male M. vatia do not even appear to establish whether the individual
producing this line is a female conspecific. This result is consistent with
the inability of female M. vatia guarding their silk-covered nests to
distinguish among their own silk, that of conspecifics, and that of certain
other species (Morse,
1989).
Relative importance of different cues for mate-seeking male M.
vatia
The cues and resulting strategies male M. vatia use to find
females thus appear to be but a small part of those used by some other species
of spiders (Dodson and Beck,
1993; Tietjen and Rovner,
1982; Watson,
1991). Adult males do follow lines (option 4), but they only use
mechanical cues in the process. The apparent absence of female cues results in
males carrying out a pattern of mate seeking that in several key ways
resembles the conditions for random search
(Koopman, 1956;
Stone, 1975), incorporating
fewer contributions from sensory stimuli than usual in making decisions of
this sort (Pyke, 1978). Put in
the context of the problem faced by the males, these conditions are that (1)
the female is stationary and equally likely to be found at any point within
the area searched, (2) the male's search path is random, and (3) the female
must be detected if she falls within the path searched by the male. Although
the areas searched by M. vatia are somewhat patchy
(Holdsworth and Morse, 2000),
the spiders exhibit little ability to sort out foraging patches from the
surrounding environment at a distance
(Morse, 1993), and they
exhibit little tendency to respond to long-distance cues. However, males are
relatively adept at responding to females within a very limited range
(Holdsworth and Morse,
2000).
Adult male M. vatia appear to exhibit an element of random search
(option 1), which only appears to shift to a more systematic search (option 5)
when lines are discovered. However, little discrimination occurs among lines,
and once deciding to follow a line, the direction taken is random, although if
they fail to find a female, they often retrace that line. Combined with
options 2, 3, and 6 (visual cues, airborne olfactory cues, flower choice), a
general line-following tendency and similar male and female flower choice
(Chien and Morse, 1998) emerge
as the only contributors to mate finding, other than for likely short-range
cues operating among these three options at within-flower distances
(Holdsworth and Morse, 2000).
These lines thus appear to provide more important cues than flower choice.
Possible basis for dearth of cues
In light of the widespread distribution of contact pheromones among spiders
and other terrestrial arthropods, the likely ancestral nature of chemical
signaling (Pollard et al.,
1987; Tietjen and Rovner,
1982), and the important role of pheromones in helping individuals
to locate and assess each other (Shorey,
1976; White et al.,
1995), why doesn't M. vatia use pheromones? Two
nonexclusive possibilities may be noted: the relationship between pheromone
production and micro-habitat (Tietjen and
Rovner, 1982) and suppression of pheromone production by
females.
Even though chemical signaling is widespread among spiders, its importance
varies widely among them. Tietjen
(1977,
1979) suggests that sheltered
microhabitats occupied by many species should favor the evolution of accurate
line-following behavior. For example, male Lycosa punctulata, which
live low in the herbaceous stratum of open fields with many other spiders,
depend more heavily on line-borne pheromones than do L. rabida,
inhabitants of the upper stratum where fewer other spiders venture
(Tietjen and Rovner, 1982).
L. rabida would presumably not benefit significantly from an
elaborate pheromone system, and in fact rely more on the mechanical presence
of a line than the pheromones present on it (Tietjen,
1977,
1979). M. vatia
resemble L. rabida in spending the majority of their time in the
upper stratum of the herbaceous layer. Additionally, female M. vatia
habitually remain at the same site for several days
(Morse and Fritz, 1982) and
thus may often be surrounded by their own lines. Since such lines retain
chemical activity for a day or less
(Pollard et al., 1987), it
would be to the males' advantage to respond mechanically to these lines, but
opportunities to respond chemically (assuming the presence of pheromones)
would be much less frequent.
Minimizing the frequency of precopulatory guarding may decrease the
harassment of females during a time critical for maximizing foraging success.
Although lack of precopulatory guarding would also minimize the opportunity of
females to exercise indirect mate choice
(Wiley and Poston, 1996) by
precipitating male—male combat (see
Watson, 1991), it could
facilitate indirect mate choice through scramble competition
(Andersson, 1994;
LeGrand and Morse, 2000). Such
a relationship should prevail only within a narrow window of opportunity,
where males are common enough to ensure a female of being mated, but scarce
enough that scramble competition could prevail as a nonrandom mode of
selection.
|
ACKNOWLEDGEMENTS |
|---|
We thank S. Bruner, D. Iyer, A. Kopelman, A. Lucky, R. Lutzy, N. Malik, and E. K. Morse for their help in the field and for their many conversations about crab spiders. We also thank J. Kraus, H. Hu, H. Sullivan, J. M. Hooper, and A. Kopelman for additional insights and R. Feldman, A. Kopelman, J. Kraus, and J. K. Waage for comments on the manuscript. K. J. Eckelbarger, T. E. Miller, and other staff members of the Darling Marine Center of the University of Maine facilitated work on the premises. In particular, T. E. Miller helped in many ways to expedite this research. J.T.A. was supported by a Howard Hughes undergraduate summer fellowship. This work was partially supported by National Science Foundation grant IBN98-16692.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Andersson M, 1994. Sexual selection. Princeton, New Jersey: Princeton University Press.
Blanke R, 1975. Untersuchungen zum Sexualverhalten von Cyrtophora cicatrosa (Stoliczka) (Araneae, Araneidae). Z Tierpsychol 37: 62-74.[Medline]
Charnov EL, 1976. Optimal foraging: the marginal value theorem. Theor Popul Biol 9: 129-136.[ISI][Medline]
Chien SA, Morse, DH, 1998. Foraging patterns of male crab spiders Misumena vatia. J Arachnol 26: 238-243.
Darwin C, 1871. The descent of man, and selection in relation to sex. London: Murray.
Dodson GN, Beck MW, 1993. Precopulatory guarding of penultimate females by male crab spiders, Misumenoides formosipes. Anim Behav 46: 951-959.
Foelix R, 1996. The biology of spiders, 2nd ed. Oxford: Oxford University Press.
Fritz RS, Morse DH, 1985. Reproductive success, growth rate and foraging decisions of the crab spiders, Misumena vatia. Oecologia 65: 194-200.
Gertsch WJ, 1939. A revision of the typical crab-spiders (Misumeninae) of America north of Mexico. Bull Am Mus Nat Hist 76: 277-442.
Ghiselin MT, 1974. The economy of nature and the evolution of sex. Berkeley: University of California Press.
Holdsworth AR, Morse DH, 2000. Mate guarding and aggression by the crab spider Misumena vatia in relation to female reproductive status and sex ratio. Am Midl Nat 143: 201-211.
Hunt R, Nault L, 1991. Roles of interplant movement, acoustic communication and phototaxis in mate-location behavior of the leaf-hopper Graminella nigrifrons. Behav Ecol Sociobiol 28: 315-320.
Kareiva P, Morse DH, Eccleston J, 1989. Stochastic prey arrivals and crab spider giving-up times: simulations of spider performance using two simple `rules of thumb.' Oecologia 78: 542-549.
Koopman BO, 1956. The theory of search II: target detection. Operat Res 4: 503-531.
LeGrand RS, Morse DH, 2000. Factors driving extreme sexual size dimorphism under low density. Biol J Linn Soc 71: 643-664.
Morse DH, 1981. Prey capture by the crab spider Misumena vatia (Clerk) (Thomisidae) on three common native flowers. Am Midl Nat 105: 358-367.
Morse DH, 1989. Nest acceptance by the crab spider Misumena vatia (Araneae, Thomisidae). J Arachnol 17: 49-57.
Morse DH, 1993. Choosing hunting sites with little
information: patch choice responses of crab spiders to distant cues.
Behav Ecol 4:
61-65.
Morse DH, 1999. Choice of hunting site as a consequence of experience in late-instar crab spiders. Oecologia 120: 252-257.
Morse DH, Fritz RS, 1982. Experimental and observational studies of patch-choice at different scales by the crab spider Misumena vatia. Ecology 65: 172-182.
Morse DH, Stephens EJ, 1996. The consequence of adult foraging success on the components of lifetime fitness in a semelparous, sit-and-wait predator. Evol Ecol 10: 361-373.
Parker GA, 1974. Courtship persistence and female-guarding as male time investment strategies. Behaviour 48: 157-184.
Parker GA, 1978. Searching for a mate. In: Behavioral ecology: an evolutionary approach (Krebs JR, Davies NB, eds). Sunderland, Massachusetts: Sinauer Associates; 214-244.
Parker GA, 1979. Sexual selection and sexual conflict. In: Sexual selection and reproductive competition in insects (Blum M, Blum N, eds). New York: Academic Press; 123-166.
Parker GA, 1983. Mate quality and mating decisions. In: Mate choice (Bateson P, ed). Cambridge: Cambridge University Press; 141-166.
Pollard SD, MacNab AM, Jackson RR, 1987. Communication with chemicals: pheromones and spiders. In: Ecophysiology of spiders (Nentwig W, ed). Berlin: Springer-Verlag; 133-141.
Pyke GH, 1978. Are animals efficient harvesters? Anim Behav 26: 241-250.
Real LA, 1990. Search theory and mate choice. I. Models of single-sex discrimination. Am Natur 136: 376-405.
Ridley M, 1983. The explanation of organic diversity. Oxford: Clarendon Press.
Schmitt A, Schuster M, Barth FG, 1990. Daily locomotor activity patterns in three species of Cupiennius (Araneae, Ctenidae): the males are the wandering spiders. J Arachnol 18: 249-255.
Shorey HH, 1976. Animal communication by pheromones. New York: Academic Press.
Stone LD, 1975. Theory of optimal search. New York: Academic Press.
Tietjen WJ, 1977. Dragline-following by male lycosid spiders. Psyche 84: 165-178.
Tietjen WJ, 1979. Tests for olfactory communication in four species of wolf spiders (Araneae, Lycosidae). J Arachnol 7: 197-206.
Tietjen WJ, Rovner JS, 1982. Chemical communication in lycosids and other spiders. In: Spider communication: mechanisms and ecological significance (Witt PN, Rovner JS, eds). Princeton, New Jersey: Princeton University Press; 249-279.
Watson PJ, 1991. Multiple paternity and first-mate sperm precedence in the Sierra dome spider Linyphia litigiosa Keyserling (Linyphiidae). Anim Behav 41: 135-148.
White CS, Lambert DM, Foster SP, 1995. Chemical signals and the recognition concept. In: Speciation and the recognition concept: theory and application (Lambert DM, Spencer HG, eds). Baltimore, Maryland: Johns Hopkins University Press; 301-326.
Wiley RH, Poston J, 1996. Indirect mate choice, competition for mates, and coevolution of the sexes. Evolution 50: 1371-1381.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  M. M. Kasumovic, M. J. Bruce, M. E. Herberstein, and M. C.B. Andrade Risky mate search and mate preference in the golden orb-web spider (Nephila plumipes) Behav. Ecol., January 1, 2007; 18(1): 189 - 195. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||