2011 SOUTHEASTERN NATURALIST 10(4):579–590 The Ecological Significance of Leg Autotomy for Climbing Temperate Species of Harvestmen (Arachnida, Opiliones, Sclerosomatidae) Jennifer E. Houghton1, Victor R. Townsend, Jr.1,*, and Daniel N. Proud2 Abstract - In encounters with predators, sclerosomatid harvestmen may employ a variety of defensive tactics including the voluntary detachment of legs (autotomy). The long-term costs of this evasive defense are not fully understood, but prior studies have documented negative consequences for terrestrial locomotion and foraging. In this study, we investigated the impact of leg loss upon locomotion in adult harvestmen (Leiobunum spp.). In southeastern Virginia, these harvestmen regularly climb vegetation and occupy perches on tree trunks, branches, and leaves that are often 1–2 m or more above the ground. In our study, we measured walking and climbing speeds for individuals with 5, 6, 7, and 8 legs. The results of our field surveys conducted over three seasons revealed relatively high frequencies (36–63%) of leg loss. We also found that individuals with six legs occupied perches that were significantly lower in the understory than those with eight legs. In the lab, we observed significantly slower walking speeds for individuals missing one or more legs. We also found that individuals with five legs climb significantly slower than individuals with eight legs. On the bases of the observed frequencies of leg loss in the field, we infer that leg autotomy is a common (and effective) evasive tactic used by harvestmen. However, the reduction in walking and climbing speeds resulting from leg loss may also affect habitat selection, (e.g., perch height) and may ultimately reduce the survivorship of individuals in future encounters with predators. Introduction Common predators of adult harvestmen (Arachnida, Opiliones) include ants, spiders, anurans, lizards, passerine birds, and insectivorous mammals (reviewed by Cokendolpher and Mitov 2007). To elude or escape from these predators, harvestmen use a variety of primary and secondary defenses (reviewed by Gnaspini and Hara 2007). Primary defenses such as crypsis and anachoresis (i.e., living in a burrow or hole) enable individuals to avoid detection (Pomini et al. 2010). Secondary defenses are used during actual encounters with predators and include morphological features (e.g., hard exoskeleton, large spines) as well as aggressive and evasive behavioral responses (Cokendolpher 1987, Machado et al. 2005). Aggressive behaviors include pinching with legs and chelicerae (Hara and 1Department of Biology, Virginia Wesleyan College, 1584 Wesleyan Drive, Norfolk, VA 23509. 2Department of Biology, University of Louisiana at Lafayette, Box 42451, Lafayette, LA 70504-2451. *Corresponding author - vtownsend@vwc.edu. 580 Southeastern Naturalist Vol. 10, No. 4 Gnaspini 2003, Machado 2002), and the release of chemical secretions produced by exocrine glands (Clawson 1988, Duffield et al. 1981, Hara et al. 2005, Jones et al. 1977; Juberthie 1976, Machado et al. 2005, Roth and Eisner 1962). Common evasive responses include thanatosis (Chelini et al. 2009, Hara and Gnaspini 2003), fleeing (Edgar 1971, Machado et al. 2000), deimatic behavior (Pomini et al. 2010), bobbing (Edgar 1971) and leg autotomy (Edgar 1971, Guffey 1998a). In general, harvestmen employ multiple secondary defenses and may use different tactics depending upon the type of predator encountered (Eisner et al. 2004, Gnaspini and Cavalheiro 1998, Machado and Pomini 2008, Pomini et al. 2010, Willemart and Gnaspini 2004). Leg autotomy or “self-amputation” is a defensive mechanism that is not unique to harvestmen. Leg detachment and its functional and ecological significance has been investigated in a variety of arthropods including crustaceans (Juanes and Smith 1995, Wasson et al. 2002), insects (Bateman and Fleming 2006, Fleming and Bateman 2007, Maginnis 2006), amblypgyids (Weygoldt 2000), and spiders (Amaya et al. 2001, Apontes and Brown 2005, Brueseke et al. 2001, Johnson and Jakob 1999, Punzo 1997, Taylor et al. 2006). While the primary benefit of leg autotomy (i.e., survival) is immediate and significant, the long-term costs of leg autotomy are more difficult to assess, but may have a considerable impact on survival. In general, leg autotomy has been observed to affect reproductive behavior (Taylor et al. 2006), foraging efficiency (Brueseke et al. 2001, Guffey 1999), agonistic interactions (Macías-Ordóñez 1997), developmental time (Johnson and Jakob 1999), and locomotion (Amaya et al. 2001, Apontes and Brown 2005, Bateman and Fleming 2006, Fleming and Bateman 2007, Guffey 1998b, Maginnis 2006). Significant reductions in terrestrial locomotion resulting from leg autotomy may also increase the susceptibility of individuals to predation (Bateman and Fleming 2006, Fleming and Bateman 2007). In sclerosomatid harvestmen, autotomy occurs through the voluntarily detachment of the leg at the trochanter-femur joint with little or no bleeding (Miller 1977, Roth and Roth 1984). The detached leg may twitch regularly for 1 min or longer (Miller 1977, Roth and Roth 1984). In populations of Leiobunum nigripes Weed and L. vittatum (Say) in Louisiana, nearly 50% of adults examined in the field were missing at least one leg (Guffey 1998a). In sclerosomatid harvestmen, the second pair of legs is the longest and is believed to be the most important with respect to the sensory ecology of the animal (reviewed by Guffey 1998b). In Louisiana, harvestmen lost legs from this second pair at a significantly greater frequency than legs from other positions (Guffey 1998a). In addition, the loss of multiple legs significantly reduced walking speeds (Guffey 1999), which was hypothesized to negatively influence the ability of individuals to flee effectively from predators. Harvestmen with missing legs were also not as efficient with respect to foraging as conspecifics that had eight legs (Guffey 1999). In intrasexual contests between adult male L. vittatum, winners generally had the same number or more legs than losers (Macías-Ordóñez 1997). 2011 J.E. Houghton, V.R. Townsend, Jr., and D.N. Proud 581 In this study, we examine the functional significance of leg autotomy in Leiobunum formosum Wood and L. politum Weed, two species of sclerosomatid harvestmen that occur in mixed hardwood forests on the coastal plain of Virginia. These species are common inhabitants of the leaf-litter community after dark. During the day, however, individuals frequently occupy perches in the understory and on tree trunks. Through field surveys, we assessed the frequencies of leg autotomy by adults. For L. politum, we also examined the relationship between perch height and leg number. In the laboratory, we measured the impact of leg loss upon walking and climbing speeds. We also investigated the impact of the loss of a leg from a specific anatomical position (1st, 2nd, 3rd, or 4th pair) upon both climbing and walking speeds in L. formosum. Study Area The field study was conducted in the pine and mixed hardwood forest on the campus of Virginia Wesleyan College (VWC) in southeastern Virginia. The pine habitat was comprised of a monotypic stand of Pinus taeda x palustris (hybrid Loblolly Pine trees) with an herbaceous layer dominated by a mixture of Hedera helix L. (English Ivy), Rhus toxicodendron (L.) (Poison Ivy), and Parthenocissus quinquefolia (L.) (Virginia Creeper). The external surfaces of most trees were thickly covered by ivy, and the canopy height was 15–20 m (the age of the stand was approximately 35–40 years). The mixed hardwood area is similar in composition to that reported for deciduous forests throughout the coastal plain of Virginia (Mitchell 1994). Our study site supports populations of four species of harvestmen including Vonones sayi (Simon), Leiobunum formosum, L. politum, and L. ventricosum (Wood). In May and June, adult L. ventricosum and L. politum are common, and occupy perches in the understory and on trees during the day. Nymphs of L. formosum are also relatively abundant, but adults of this species are rare. In August– November, adult L. formosum are the most common harvestmen in the forest. During the day, individuals of this species may be found on the vegetation, but are most readily observed wandering in the leaf litter at night. Adult L. politum and L. ventricosum are seldom seen in the fall. The cosmetid harvestman V. sayi is generally uncommon throughout the summer and rarely observed in the litter. This species is most commonly found underneath logs where the forest litter is relatively moist. Methods Adult Leiobunum formosum were captured by hand from the understory, trunks of large trees, and the leaf litter during the afternoon (1300–1600 hrs) and early evening (1700–2100 hrs) in 2003 (9 October–23 November) and 2004 (1 September–17 November). The number of intact legs for each specimen was determined prior to preservation in 70% ethanol. Adult L. politum were collected during the day (1200–1600 hrs) from 26 May–22 June 2010. Prior to collection, 582 Southeastern Naturalist Vol. 10, No. 4 perch heights were measured using meter sticks and the perch type (species of plant) was recorded. The number of legs, reproductive status (nymph or adult), and sex for each individual was also determined. Following capture, each harvestman with missing legs was preserved in 70% ethanol. Individuals with eight legs were placed into a common mesh cylinder (14 x 12 x 22 cm) and returned to the lab for use in the behavioral experiments. From 1 September–18 November 2010, we only captured adult L. formosum that had eight legs (1300–1700 hrs). These individuals were placed into a common mesh cylinder and returned to the lab for use in the behavioral experiments. In the laboratory, the harvestmen were housed in a large aquarium (41 x 21 x 17 cm) lined with a thin layer (3 cm) of mulch and covered by an aluminum mesh. Egg crates were placed in the aquarium to serve as diurnal refugia. The harvestmen were kept under low-light conditions (10 ± 3 lux) at room temperature (23 ± 2 ºC) and provided baby food, wingless fruitflies (Drosophila melanogaster Meigen), and water ad libitum. Individuals were communally housed for 6–12 days. Approximately 24 hrs before each trial, harvestmen were induced to autotomize one or more legs. The number and position of the legs to be lost were determined with the aid of a random numbers table. To induce autotomy, we followed the protocol of Guffey (1999). Each leg was grasped mid-femur by a pair of forceps and the harvestman was allowed to autotomize the leg. This response generally occurred within a few seconds of seizure. Control individuals (eight legs) were handled, but not grabbed with forceps, and returned to the aquarium. For the first laboratory experiment (4 June–1 July 2010), we examined the general impact of leg loss upon walking and climbing speeds for adult L. politum. Individuals of this species were divided into four treatment groups based upon the number of legs (8, 7, 6, or 5). For the five-leg treatment group, leg autotomy was controlled so that each harvestman would have at least two functional legs on each side of the body. Thus, for this treatment, each individual had two legs on one side and three legs on another; no individual had four legs on one side and only one leg on the other. The sequence of the trials was randomized with respect to type (climbing or walking) and treatment group (8, 7, 6 or 5 legs). To assess walking speeds, we created a linear trackway (100 x 7 x 4 cm) using six meter sticks on a flat table top. On each side of the track, the first two meter sticks were placed flat, and a third was positioned on its side on top of the first two, enabling us to easily read cm increments while providing enough space so that harvestmen could easily walk down the trackway without impairment. To prevent escapes, the top was covered with a sheet of glass that also enabled us to observe the harvestmen as they moved on the track. At the beginning of each trial, each harvestman was grasped by the posterior-most pair of legs and the anterior margin of the body was held at the 0 cm mark on the test track until it ceased moving (following the protocol of Guffey 1999). The 2011 J.E. Houghton, V.R. Townsend, Jr., and D.N. Proud 583 individual was released, and we measured the time that it took to reach the 100 cm mark. On occasion, a harvestman would pause or cease moving and then would resume walking. In these trials, we only measured the actual time in which the individual was moving. Thus, we were able to determine a walking speed (cm/s) for each individual. Following each trial, the harvestman was preserved in 70% ethanol. To determine climbing speeds, we used a large piece of roughened corkboard (43 x 29 x 2 cm), which we placed into a plastic shoebox (23 x 19 x 11.5 cm). We placed three bricks around the base of the board to provide stability and hold the corkboard in a vertical position. We flooded the box with 3–5 cm of water to discourage harvestmen from escaping. At the beginning of each trial, we grasped the harvestman by the posterior-most pair of legs and held it at the base of the board until movement ceased (similar to Guffey 1999). The specimen was released, and we measured the time required for the individual to climb to the 31 cm mark on the board. As with the walking trials, we only measured the actual time in which an individual was moving. Thus, we were able to determine a climbing speed (cm/s) for each individual. Following each trial, the harvestman was preserved in 70% ethanol. For the second laboratory experiment (17 September–5 November 2010), we investigated the impact of the loss of one specific leg upon walking and climbing speeds for adult L. formosum. Therefore, we divided the adult L. formosum into five treatment groups based upon the position of the leg lost. The four experimental treatment groups featured individuals missing one leg from the 1st, 2nd, 3rd or 4th position. The control treatment group featured individuals with eight legs. The sequence of trials was randomized with respect to type (climbing or walking) and treatment group. We assessed walking and climbing speeds using the same protocols described for the first experiment. Following each trial, harvestmen were preserved in 70% ethanol. Although the field data for perch height deviated slightly from normality, residuals approximated a normal distribution based on normal quantile plots, and variances were homogeneous. A one-way ANOVA was utilized to evaluate differences in the perch heights as a function of the number of lost legs (0, 1, or 2). The post-hoc Tukey honest significant differences (Tukey HSD) method was used to determine which groups differed. Means and standard errors of measured perch height (cm) are reported. The laboratory-based measurements of walking and climbing speeds deviated only slightly from a normal distribution. The residuals of walking and climbing velocities approximated a normal distribution based on normal quantile plots. Therefore, we employed a two-way ANOVA to evaluate differences in velocity as a function of leg number (5, 6, 7, or 8) and trial type (walking or climbing). Tukey’s HSD tests were used to determine which pairs of means were statistically different (α = 0.05). All statistical analyses were performed in R (R Development Core Team 2010). 584 Southeastern Naturalist Vol. 10, No. 4 Results Rates of leg autotomy observed in the forest ranged from 36–64% (Table 1). For L. formosum, more than 50% of adults in the population were missing one or two legs. For L. politum, approximately one third of adults were missing one or two legs. Relatively few individuals (2–9%) in any of our samples were missing three or more legs (Table 1). We did not observe any pattern with respect to the loss of legs from specific anatomical positions (e.g., leg I vs. leg II). Our statistical analysis of the field data for perch heights revealed that individuals with eight legs occupied the highest perches (mean = 115.3 cm; n = 139; SE = 4.2), followed by individuals with seven (mean = 104.4 cm; n = 52; SE = 6.5) and six legs (mean = 79.8 cm; n = 22; SE = 9.0). The number of lost legs had a significant effect on the height at which individuals were observed (F2,210 = 5.4, P < 0.01). A pairwise post-hoc Tukey HSD test indicated that mean perch heights were significantly higher for individuals with eight legs compared to those with six legs (P < 0.01); all other comparisons were not statistically significant. With respect to the laboratory experiments, individuals with eight legs exhibited the fastest mean walking speeds (9.33 cm/s ± 0.61 SE), followed by individuals with seven (6.04 cm/s ± 0.54 SE), six (3.17 cm/s ± 0.38 SE), and five (2.08 cm/s ± 0.28 SE) legs. Climbing velocities showed a similar pattern. Harvestmen with all eight legs were fastest (3.89 cm/s ± 0.30 SE), followed by individuals with seven (2.75 cm/s ± 0.29 SE), six (2.69 cm/s ± 0.19 SE), and five (1.71 cm/s ± 0.15 SE) legs. The analysis of variance revealed that the main effects of leg number (F3,232 = 59.96, P < 0.001) and trial type (F1,232 = 82.3, P < 0.001) were significant in the model, and their interaction effect also had a significant effect on velocity (F3,232 = 21.3, P < 0.001). Pairwise comparisons using the Tukey HSD test revealed that mean walking speeds decreased significantly as leg number decreased from eight to seven and seven to six legs (all P < 0.001). The walking velocities of individuals with five legs were significantly slower than individuals with eight or seven legs (both P < 0.001), but did not differ from those with six legs (P = 0.36; Fig. 1A). Pairwise comparisons revealed that mean climbing speeds were only weakly affected by leg number. Mean climbing velocities were significantly slower for individuals with five legs compared to those with eight legs (P = 0.002), Table 1. Comparison of leg numbers observed for adult Leiobunum formosum and L. politum in populations at the study site over three different sampling periods. Leiobunum formosum Leiobunum politum Number of legs present 2003 (n = 292) 2004 (n = 393) 2010 (n = 222) 8 44% 37% 64% 7 34% 36% 24% 6 19% 18% 10% 5 or fewer 3% 9% 2% 2011 J.E. Houghton, V.R. Townsend, Jr., and D.N. Proud 585 but no differences were detected in climbing speeds between other pairwise comparisons (Fig. 1B). Based on the Tukey HSD multiple comparisons, the mean walking velocity of individuals with eight or seven legs was significantly faster than mean climbing velocity (both P < 0.001). No significant differences were detected between mean walking and climbing speeds for individuals with six (P = 0.98) or five (P = 0.99) legs. The statistical analysis of the results for the second lab experiment (Table 2) revealed no significant differences between treatments (position of leg loss) for walking (F4,95 = 0.39, P = 0.82) or climbing speeds (F4,95 = 0.62, P = 0.65). Discussion Among arthropods (i.e., crickets, decapod crustaceans, wolf spiders, amblypygids, and sclerosomatid harvestmen), leg autotomy is a common and Figure 1. Comparison of mean (A) walking and (B) climbing speeds (cm/s) for individuals of Leiobunum politum with different numbers of legs. Means are based on time trials of 30 individuals for each combination of trial type and leg number (n = 240). Letters above bars indicate groups that statistically differ from one another based on two-way ANOVA followed by a Tukey honest significant difference test. Error bars represent ± 1 standard error. Table 2. Comparison of walking and climbing speeds (cm/s) measured in the laboratory for adult Leiobunum formosum that were missing one single leg from the 1st, 2nd, 3rd, or 4th position (n = 20 for each treatment group). In the control group, harvestmen had eight legs (n = 20). There were no significant differences among the treatment groups. Walking speed Climbing speed Treatment Mean SE Mean SE 1st position missing 6.31 0.77 6.83 0.81 2nd position missing 6.46 1.23 8.42 1.08 3rd position missing 6.64 0.94 7.59 0.91 4th position missing 5.32 0.62 6.70 0.71 Control 6.79 0.71 7.17 1.08 586 Southeastern Naturalist Vol. 10, No. 4 effective secondary defense mechanism (Brueseke et al. 2001, Fleming and Bateman 2007, Gnaspini and Hara 2007, Johnson and Jakob 1999, Juanes and Smith 1995, Roth and Roth 1984, Weygoldt 2000). For harvestmen, there are relatively few field studies that report frequencies of leg autotomy (Guffey 1998), and only two lab-based investigations that have examined the significance of leg loss (Guffey 1999, Macías-Ordóñez 1997). Our field survey in southeastern Virginia revealed that 36–63% of adults in the populations of adult Leiobunum spp. were missing at least one leg, although individuals missing more than 2 legs were relatively uncommon (Table 1). These results mirror those reported by Guffey (1998) for populations of L. nigripes and L. vittatum in Louisiana. Similarly, our measurements for walking speeds are similar to those reported by Guffey (1999). In the current study, we found that individuals with five or six legs exhibited walking speeds that were significantly slower than individuals with seven or eight legs (Fig. 1). We also found a significant difference for walking speeds between individuals with seven and eight legs for L. politum. Guffey (1999) observed that individuals missing three legs walked significantly slower than individuals with seven or eight legs, but did not find any significant difference for average walking speeds between individuals with seven or eight legs. Previous studies of the costs associated with leg autotomy have revealed that individuals with missing legs are not as efficient with respect to foraging as conspecifics with eight legs (Guffey 1999). In intrasexual contests, winners also generally had greater or equal numbers of legs as losers (Macías-Ordóñez 1997). Our study represents the first examination of the impact of leg autotomy upon climbing speeds. In general, we found that individuals climbed at significantly slower speeds than they walked. In addition, we observed that individuals with five legs climbed significantly slower than individuals with eight legs. The results of our field study revealed that individuals with six legs occupy perches on the vegetation that are significantly lower in the understory than individuals with eight legs. Thus, in sclerosomatid harvestmen, the costs associated with leg autotomy include not only reductions in foraging efficiency (Guffey 1999) and competitive ability (Macías-Ordóñez 1997), but also changes in microhabitat selection. In addition to measuring walking and climbing speeds, we also examined the impact of the loss of legs from specific positions upon locomotion through the removal of one leg from the 1st, 2nd, 3rd, or 4th position for adult L. formosum. Under laboratory conditions, Guffey (1999) observed that the time between first contact with a prey item and its consumption was significantly less for individuals that first contacted the prey with a leg from the 1st position rather than a leg from the 2nd position. For L. politum, we found a significant difference between walking, but not climbing, speeds for individuals with seven and eight legs. However, we did not find any significant differences for walking or climbing speeds between individuals of L. formosum that were missing one leg from any position (or our control). Our results, thus, appear to support Guffey’s (1999) “spare-leg 2011 J.E. Houghton, V.R. Townsend, Jr., and D.N. Proud 587 hypothesis”. According to this hypothesis, the loss of one or even two legs does not have a significant effect upon an individual’s survival. The results of our field study of L. politum provide additional support for this hypothesis in that we did not detect any significant differences for perch height between individuals with seven or eight legs. Despite their high abundance in forested habitats in the southeastern US, relatively little is known about the natural history of most species of sclerosomatid harvestmen in this region. Most of what is generally known about the natural history of these arachnids in the eastern US is based upon the ecological studies by Edgar (1971) in Michigan and Guffey (1998b) in Louisiana. Sclerosomatid harvestmen are known to employ a diverse array of defensive behaviors, including bobbing, fleeing, chemical defense, and leg autotomy, when encountering potential predators (Gnaspini and Hara 2007). The efficacy of these defenses against syntopic vertebrate and invertebrate predators remains largely unexamined. Most evidence of predation is based upon examinations of gut contents, fecal samples, remains found associated with webs, or direct feedings of captive animals (Cokendolpher and Mitov 2007). Additional ecological studies of sclerosomatid harvestmen are needed to evaluate the efficacy of leg autotomy in encounters with different types of invertebrate and vertebrate predators as well as the impact of leg loss upon survival and reproduction. At our study site, we have observed adult lycosid spiders actively consuming nymphs and adults of L. formosum, as well as adult harvestmen of both of our study species trapped in the webs of araneid spiders. Additional studies that include investigations of microhabitat selection, diet, and activity may provide further insights into the short- and long-term consequences of leg autotomy for harvestmen. Acknowledgments We thank Bryanna Cunningham for assistance with collecting harvestmen. This research was supported by a VWC Summer Faculty Development Grant (V.R. Townsend, Jr.), a Batten Endowed Professorship (V.R. Townsend, Jr.), and the VWC Science Undergraduate Research Fund (J.E. Houghton). Voucher specimens will be deposited in the collection of the AMNH. Literature Cited Amaya, C.C., P.D. Klawinski, and D.R. Formanowicz. 2001. The effects of leg autotomy on running speed and foraging ability in two species of wolf spider (Lycosidae). American Midland Naturalist 145:201–205. Apontes, P., and C.A. Brown. 2005. Between-sex variation in running speed and a potential cost of leg autotomy in the wolf spider Pirata sedentarius. American Midland Naturalist 154:115–125. Bateman, P.W., and P.A. Fleming. 2006. Increased susceptibility to predation for autotomized House Crickets (Acheta domesticus). Ethology 112:670–677. Brueseke, M.A., A.L. Rypstra, S.E. Walker, and M.H. Persons. 2001. Leg autotomy in the wolf spider Pardosa milvina: A common phenomenon with few apparent costs. American Midland Naturalist 146:153–160. 588 Southeastern Naturalist Vol. 10, No. 4 Chelini, M., R.H. Willemart, and E.A. Hebets. 2009. Costs and benefits of freezing behavior in the harvestman Eumesosoma roeweri (Arachnida, Opiliones). Behavioral Processes 82:153–159. Clawson, R.C. 1988. Morphology of defense glands of the opilionids (daddy longlegs) Leiobunum vittatum and L. flavum (Arachnida: Opiliones: Palpatores: Phalangiidae). Journal of Morphology 198:363–381. Cokendolpher, J.C. 1987. Observations on the defensive behaviors of a Neotropical Gonyleptidae (Arachnida, Opiliones). Revue Arachnologique 7:59–63. Cokendolpher, J.C., and P.G. Mitov. 2007. Natural enemies. Pp. 339–373, In R. Pinto-da- Rocha, G. Machado, and G. Giribet (Eds.). Harvestmen: The Biology of Opiliones. Harvard University Press, Cambridge, MA. Duffield, R.M., O. Olubajo, J.W. Wheeler, and W.H. Shear. 1981. Alkyo-phenols in the defensive secretion of the Nearctic opilionid, Stygnomma spinifera (Arachnida: Opiliones). Journal of Chemical Ecology 7:445–452. Edgar, A.L. 1971. Studies on the biology and ecology of Michigan Phalangida (Opiliones). Miscellaneous Publications of the Museum of Zoology, University of Michigan 144:1–64. Eisner, T., C. Rossini, A. González, and M. Eisner. 2004. Chemical defense of an opilionid (Acanthopachylus aculeatus). Journal of Experimental Zoology 207:1313–1321. Fleming, P.A., and P.W. Bateman. 2007. Just drop it and run: The effect of limb autotomy on running distance and locomotion energetic of field crickets (Gryllus bimaculatus). Journal of Experimental Biology 210:1446–1454. Gnaspini, P., and A.J. Cavalheiro. 1998. Chemical and behavioral defenses of a Neotropical cavernicolous harvestman: Goniosoma spelaeum (Opiliones, Laniatores, Gonyleptidae). Journal of Arachnology 26:81–90. Gnaspini, P., and M.R. Hara. 2007. Defense mechanisms. Pp. 374–399, In R. Pinto-da- Rocha, G. Machado, and G. Giribet (Eds.). Harvestmen: The Biology of Opiliones. Harvard University Press, Cambridge, MA. Guffey, C.A. 1998a. Leg autotomy and its potential fitness costs for two species of harvestmen (Arachnida, Opiliones). Journal of Arachnology 26:296–302. Guffey, C.A. 1998b. The behavioral ecology of two species of harvestmen (Arachnida: Opiliones): The effects of leg autotomy, parasitism by mites, and aggregation (Leiobunum nigripes, Leiobunum vittatum). Ph.D. Dissertation. University of Louisiana at Lafayette, Lafayette, LA. Guffey, C.A. 1999. Costs associated with leg autotomy in the harvestmen Leiobunum nigripes and Leiobunum vittatum (Arachnida, Opiliones). Canadian Journal of Zoology 77:824–830. Hara, M.R., and P. Gnaspini. 2003. Comparative study of the defensive behavior and morphology of the gland opening area among harvestmen (Arachnida, Opiliones, Gonyleptidae) under a phylogenetic perspective. Arthopod Structure and Development 32:257–275. Hara, M.R., A.J. Cavalheiro, P. Gnaspini, and D.Y.A.C. Santos. 2005. A comparative analysis of the chemical nature of defensive secretions of Gonyleptidae (Arachnida: Opiliones: Laniatores). Biochemical Systematics and Ecology 33:1210–1225. Johnson, S.A., and E.M. Jakob. 1999. Leg autotomy in a spider has minimal costs in competitive ability and development. Animal Behaviour 57:957–965. 2011 J.E. Houghton, V.R. Townsend, Jr., and D.N. Proud 589 Jones, T.H., J. Meinwald, K. Hicks, and T. Einser. 1977. Characterization and synthesis of volatile compounds from the defensive secretions of some “daddy longlegs” (Arachnida: Opiliones: Leiobunum spp.). Proceedings of the National Academy of Science USA 74:419–422. Juanes, F., and L.D. Smith. 1995. The ecological consequences of limb damage and loss in decapod crustaceans: A review and prospectus. Journal of Experimental Marine Biology and Ecology 102:47–53. Juberthie, C. 1976. Chemical defence in soil Opiliones. Revue d'Écologie et de Biologie du Sol 13:155–160. Machado, G. 2002. Maternal care, defensive behavior, and sociality in Neotropical Goniosoma harvestmen (Arachnida, Opiliones). Insectes Sociaux 49:1–6. Machado, G., and A.M. Pomini. 2008. Chemical and behavioral defenses of the Neotropical harvestman Camarana flavipalpi (Arachnida, Opiliones). Biochemical Systematics and Ecology 36:369–376. Machado, G., R.L.G. Raimundo, and P.S. Oliveira. 2000. Daily activity schedule, gregariousness, and defensive behavior in the Neotropical harvestmen Goniosoma longipes (Opiliones: Gonyleptidae). Journal of Natural History 34:587–596. Machado, G., P.C. Carrera, A.M. Pomini, and A.J. Marsaioli. 2005. Chemical defense in harvestmen (Arachnida, Opiliones): Do benzoquinone secretions deter invertebrate and vertebrate predators? Journal of Chemical Ecology 31:2519–2539. Macías-Ordóñez, R. 1997. The mating system of Leiobunum vittatum Say 1821 (Arachnida: Opiliones: Palpatores): Resource defense polygyny in the striped harvestman. Ph.D. Dissertation, Lehigh University, Bethlehem, PA. Maginnis, T.L. 2006. Leg regeneration stunts wing growth and hinders flight performance in a stick insect (Sipyloidea sipylus). Proceedings of the Royal Society B 273:1811–1814. Miller, P.L. 1977. Neurogenic pacemakers in the legs of Opiliones. Physiological Entomology 2:213–224. Mitchell, J.C. 1994. Reptiles of Virginia. Smithsonian Institution Press, Washington, DC. Pomini, A.M., G. Machado, R. Pinto-da-Rocha, R. Macías-Ordóñez, and A.J. Marsaioli. 2010. Lines of defense in the harvestman Hoplobunus mexicanus (Arachndia: Opiliones): Aposematism, stridulation, thanatosis, and irritant chemicals. Biochemical Systematics and Ecology 38:300–308. Punzo, F. 1997. Leg autotomy and avoidance behavior in response to a predator in the wolf spider Schizocosa avida (Araneae, Lycosidae). Journal of Arachnology 25:202–205. R Development Core Team. 2010. R: A language and environment for statistical computing. R Foundation for Statistical Computing. Vienna, Austria. Available online at http://www.R-project.org. Roth, L.M., and T. Eisner. 1962. Chemical defenses of arthropods. Annual Review of Entomology 7:107–136. Roth, V.D., and B.M. Roth. 1984. A review of appendotomy in spiders and other arachnids. Bulletin of the British Arachnological Society 6:137–146. Taylor, P.W., J.A. Roberts, and G.W. Uetz. 2006. Compensation for injury? Modified multi-modal courtship of wolf spiders following autotomy of signaling appendages. Ethology, Ecology, and Evolution 18:79–89. 590 Southeastern Naturalist Vol. 10, No. 4 Wasson, K., B.E. Lyon, and M. Knope. 2002. Hair-trigger autotomy in porcelain crabs is a highly effective escape strategy. Behavioral Ecology 13:481–486. Weygoldt, P. 2000. Whip Spiders (Chelicerata: Amblypgyi): Their Biology, Morphology, and Systematics. Apollo Books, Stenstrup, Denmark. Willemart, R.H., and P. Gnaspini. 2004. Spatial distribution, mobility, gregariousness, and defensive behaviour in a Brazilian cave harvestman Goniosoma albiscriptum (Arachnida, Opiliones, Gonyleptidae). Animal Biology 54:221–235.