This action might not be possible to undo. Are you sure you want to continue?
(2005) 267, 1–7
2005 The Zoological Society of London
Printed in the United Kingdom
Social behaviour in an Australian velvet worm, Euperipatoides rowelli (Onychophora: Peripatopsidae)
Judith Reinhard1 * and David M. Rowell2
Visual Sciences, Research School of Biological Sciences, The Australian National University, G.P.O. Box 475, Canberra, ACT 2601, Australia Faculty of Science, Division of Botany and Zoology, The Australian National University, Canberra, ACT 0200, Australia
(Accepted 20 January 2005)
Abstract Onychophora lead a cryptic lifestyle as predators in decaying logs and leaf litter. Hence, very little is known about their behaviour. They are generally assumed to possess only a limited behavioural repertoire, even though their surprisingly evolved brain suggests otherwise. Our studies on the Australian species Euperipatoides rowelli show for the ﬁrst time that onychophorans are indeed capable of complex behaviour. Euperipatoides rowelli individuals form aggregations of up to 15 females, males and young. These aggregations are not random assemblages, but close social groups organized in a hierarchy based on female dominance. Food is hunted collectively, but the dominant female of a group feeds alone, before other females, males and young. Hierarchy within a group is established by aggressive-dominant and passive-subordinate behaviours, the latter leading to tolerance of body contact and aggregation. Euperipatoides rowelli from foreign groups, i.e. from different logs, are met with intense aggression, and individuals rarely aggregate. The reasons for this aggression are not clear, but we suggest that its origins lie in kin recognition. The evolution of social behaviour within the Onychophora is discussed with respect to their ecology, population genetics and phylogeny. Key words: Onychophora, social behaviour, group living, aggression, aggregation, Euperipatoides rowelli
Onychophorans are soft-bodied terrestrial invertebrates commonly called peripatus or velvet worms. They are widely distributed in southern hemisphere temperate regions and in the tropics. Onychophorans are vulnerable to desiccation and hence largely conﬁned to moist, humid microhabitats, including leaf litter and the interior of decomposing logs, where they are predators on small invertebrates (Tait, 1998). Owing to their anatomy, onychophoran phylogeny has received much attention over the years. While traditionally thought to be the ‘missing link’ between annelids and arthropods (Ghiselin, 1984), other studies suggest that the Onychophora forms a clade embedded within the arthropods (Ballard et al., 1992), or that they are a sister group to the arthropods within the Ecdysozoa and to the exclusion of the annelids (Aguinaldo et al., 1997). There have been numerous other scenarios offered, which are beyond the scope of this paper (Hackmann & Goldberg, 1975; Grosberg, 1990; Wheeler, Cartright & Hayashi, 1993; Telford & Thomas, 1995).
*All correspondence to: J. Reinhard. E-mail: email@example.com
Onychophorans have been used as models for a range of studies in both population and evolutionary biology. These encompass biogeography, speciation, population genetics, ecology and reproductive biology (Havel, Wilson & Hebert, 1989; Hebert et al., 1991; Scott & Rowell, 1991; Briscoe & Tait, 1995; Tait & Briscoe, 1995; Tait, Briscoe & Rowell, 1995; Reid, 1996; Gleeson et al., 1998; Trewick, 1998; Walker & Campiglia, 1998; Curach & Sunnucks, 1999; Barclay, Ash & Rowell, 2000; Barclay, Rowell & Ash, 2000; Eriksson & Budd, 2000; Sunnucks et al., 2000; Rockman, Rowell & Tait, 2001; Rowell, Rockman & Tait, 2002; Eriksson, Tait & Budd, 2003; Walker & Tait, 2004). There is some evidence of variation in behaviour amongst different onychophoran genera (Hamer, Samways & Ruhberg, 1997), but ethological studies of onychophorans are rare, perhaps because of their cryptic life style. Their biology restricts them to moist microhabitats, and in drier regions such as temperate Australia, populations are largely conﬁned to the inside of rotting logs (Scott & Rowell, 1991; Barclay, Ash & Rowell, 2000). There are no detailed behavioural studies on encounters between individual onychophorans, but neuroanatomy studies have revealed that the onychophoran brain contains the same neural structures (mushroom bodies) that are
J. REINHARD AND D. M. ROWELL
Fig. 1. Euperipatoides rowelli aggregating behaviour: (a) Petri dish with an aggregation and two walking individuals (female below, male above); (b) magniﬁcation of the aggregation.
specimens were collected from 5 logs located within an area of 10 km2 in the Harolds Cross biogeographic region of Tallaganda (as described by Garrick et al., 2004). Twenty to 50 individuals per log were collected, and animals from 1 log were considered a ‘group’, as they were either found in close proximity to one another or at least potentially in contact through a continuous medium. Animals from different groups were kept separate in plastic containers in the laboratory. The containers were ﬁlled with moist sphagnum moss, closed for humidity maintenance, and stored in the dark at room temperature. Specimens were fed every 2 weeks with crickets Teleogryllus commodus (1–2 cm).
believed to play a role in controlling complex behaviours in arthropods, especially in social insects (Strausfeld et al., 1998). The existence of these neural structures in Onychophora indicates the potential for more complex behaviour than has previously been assumed or demonstrated. The Australian species Euperipatoides rowelli Reid 1996 (Peripatopsidae) (previously included in E. leuckartii) occurs in humid, temperate forest habitats in the south-east of the continent (Scott & Rowell, 1991; Reid, 1996; Barclay, Ash & Rowell, 2000; Barclay, Rowell & Ash, 2000). Here, they live in the crevices of decaying logs, where they prey on small invertebrates such as termites (Scott & Rowell, 1991; Barclay, Ash & Rowell, 2000). Euperipatoides rowelli is ovoviviparous and males reach maturity in their ﬁrst year with the larger females maturing in their second or third year (Sunnucks et al., 2000). Females may contain sperm from multiple males, and there is evidence for female post-copulatory inﬂuence over fertilizations (Curach & Sunnucks, 1999). In contrast to other species, which are usually sparsely distributed, E. rowelli females, males and young are often found in aggregations (Fig. 1). These aggregations can contain between two and 15 individuals, lying within a few cm of one another or in direct physical contact. Only rarely is a single animal found in a log. Are these aggregations merely random assemblies of interchangeable individuals exploiting favourable microhabitats, or are they social groups, in which the members recognize one another? Is there any organization within an aggregation? To investigate these questions, we studied interactions within and between E. rowelli groups.
MATERIALS AND METHODS Collection of specimens
Specimens were collected from Tallaganda State Forest, New South Wales, Australia. This forest occupies a region of 110 × 10 km, and population densities of E. rowelli exceed 1000 individuals/ha on south-eastern slopes (Barclay, Ash & Rowell, 2000). Because genetic studies indicate that there is regional genetic differentiation (Sunnucks & Wilson, 1999; Sunnucks & Tait, 2001),
All experiments were carried out under dim red light (25 W incandescent red light bulb, broad spectrum) to prevent negative phototactic behaviour. Immediately before each experiment, adult E. rowelli were marked individually with fast-drying enamel paint (Tamiya Color XF-7) on the back. For marking, 1 animal at a time was put into a Petri dish and small dots of paint were applied with the tip of a soft, ﬁne brush. Once the paint had dried (c. 10–20 s) the animal was returned to its container. There was no apparent response from the animals to the paint once it had dried, neither to their own nor to other animals’ markings. To investigate group behaviour during prey hunting and feeding and to establish an ethogram for interactions among individuals, 6 females, 4 males, and 5 young (gender ratio corresponding to natural average: Scott & Rowell, 1991; Sunnucks et al., 2000) were placed into a Petri dish (internal diameter 8.5 cm) with moist ﬁlter paper on the bottom. After c. 30 min, when the specimens had adjusted to the Petri dish, a live cricket was introduced as food, and behaviour was recorded for up to 6 h (Sony digital handycam DCR-TRV120E). All behaviour before, during, and after feeding was registered. Once feeding had started, the number of individual females, males and young feeding on the cricket was registered every 15 s for 3 h. From these data, the mean number of feeding females, males, and young was calculated for every 5 min interval, and the means were plotted against time. The experiment was carried out 5 times with 5 different E. rowelli groups, one each from of the 5 collection sites. Behavioural patterns displayed in interactions between individuals were deﬁned as ‘biting’, ‘chasing’, ‘evading’, ‘running’, ‘climbing’ and ‘tolerating climber’ (see Results). Based on these behaviours, a bioassay investigating inter-individual aggression and its effect on aggregating behaviour was carried out. Two E. rowelli individuals were placed in a Petri dish with moist ﬁlter paper. The frequency with which they displayed the above behaviours was registered continuously for 15 min. It was also noted if, and when, they formed an aggregation within the 15 min test duration. The bioassay was carried out pairing 2 females, or 2 males, or 1 female and 1 male
Social behaviour in Onychophora
No. of feeding Euperipatoides rowelli
from the same group, and repeated 45 times each. Animals were randomly chosen from the collections, and some may have been used in several different trials, but care was taken that any combination of 2 individuals (recognizable by their individual paint marks) was tested only once. When pairing individuals of the same gender, animals of comparable size were used (size-matched by eye). The same bioassay was carried out pairing individuals from different groups, drawn randomly, and repeated 20 times. Means and standard deviations of behavioural frequencies were calculated and the paired t-test was used to determine if the 2 individuals of a test showed signiﬁcant differences in behaviour. Percentages of aggregations formed and the mean time required for formation was also calculated. The t-test for independent samples was used to determine whether individuals from different combinations (female– female vs male–male vs female–male) and from the 2 experimental set-ups (individuals from same group vs individuals from different groups) showed signiﬁcant differences in behavioural frequencies.
6 4 2 0 6 4 2 0 6 4 2 0 6 4 2 0 6 4 2 0 Females Males Young
RESULTS General observations on aggregation behaviour
When stored in plastic containers with sphagnum moss, animals could be maintained at relatively high densities, i.e. up to 50 adult individuals. In these circumstances the animals were commonly found in groups or close vicinity, although there was no obvious tendency for tight physical contact. When placed in a Petri dish without a sphagnum substrate, the formation of tight clumps of two or more animals in intimate physical contact was common. The reason for this difference was not immediately apparent, but the Petri dish set-up clearly presented the opportunity to easily observe and quantify interactions between individuals.
Group hunting and feeding
90 Time (min)
Behavioural interactions within a group during prey hunting and feeding were consistent for all ﬁve E. rowelli groups tested. When a cricket was introduced into the Petri dish with an E. rowelli group, it was hunted collectively by adult females and males, with attacks being triggered by random encounters of the prey; young E. rowelli were not involved in the hunt. Depending on the size of the cricket, its immobilization with the sticky secretion from the onychophorans’ slime glands took from 5 min to 1 h, after which feeding started (for prey capture in onychophorans, see: Read & Hughes, 1987). We detected a strict femaledominated feeding hierarchy: the ﬁrst individual to feed was always a female, and she remained the sole feeder for 45–60 min (Fig. 2). This female fed in bouts of up to 10 min, interrupted by short intervals during which she moved around the Petri dish. During her solitary feeding time, males and young of the group remained distant,
Fig. 2. Feeding hierarchy in Euperipatoides rowelli groups. The ﬁve diagrams represent ﬁve different groups of E. rowelli, each composed of six females, four males and ﬁve young in a Petri dish, observed for 3 h while feeding on a cricket. Each diagram shows the mean number of females (solid line), males (dashed line), and young (dotted line) feeding simultaneously, sampled every 15 s and averaged for every 5 min interval.
often in physical contact with one another. The other females assumed a ‘waiting’ position close to the prey, circled it, and occasionally tried to feed. If noticed by the ﬁrst female, they were attacked by her, bitten, kicked and pushed away. After 45–60 min, the ﬁrst female tolerated other individuals at the food. Now, the remaining females started feeding in groups, soon followed by the males and
J. REINHARD AND D. M. ROWELL
two individuals become motionless, forming an aggregation. Juvenile E. rowelli showed no aggressive behaviour nor were they ever attacked. They followed and climbed adults whenever possible. The adults usually tolerated juvenile climbers, sometimes carrying them ‘piggy-back’ while moving around and when aggregating. Using the behaviours biting, chasing, evading, running and climbing as measurement for aggression and hierarchical status, behavioural interactions were investigated between two adult E. rowelli individuals, collected either from the same log (= same original group) or from different logs (= different groups). It was also noted if, and when, they formed an aggregation within the test interval. When pairing two females from the same log, a hierarchy was evident within seconds; the dominant individual was more aggressive, showing signiﬁcantly more biting and chasing behaviour, while the subordinate showed more evading, running, and also more climbing behaviour (Fig. 4a, top). Following repeated climbing attempts by the subordinate individual, aggregations between the two females were formed in 80% of the cases taking an average of 9.8 ± 3.1 min. The results were the same when pairing two males from the same log, with behavioural frequencies only differing for the behaviour ‘climbing’ (Fig. 4a, centre), which was displayed signiﬁcantly (P < 0.008) more often in the male–male (mean frequencies: 2.4 ± 2.4 and 6.1 ± 3.9) than in the female–female confrontation (mean frequencies: 0.8 ± 0.8 and 4.2 ± 2.6). Aggregations between males from the same group were formed less frequently (55.6%) and signiﬁcantly more slowly (11.3 ± 2.6 min, P < 0.043) compared to the female–female encounter. When pairing a female and a male from the same log, there was little aggression between individuals. The much smaller male was clearly subordinate, and the only behaviour displayed at high frequency was climbing by the male (Fig. 4a, bottom). Aggregations between a female and a male were formed more often (84.4%) and signiﬁcantly faster (6.4 ± 2.8 min, P < 0.001) than in female–female and male–male pairings. When pairing two E. rowelli females that originated from different logs, i.e. different groups, there was signiﬁcantly more aggression than between females from the same log (P < 0.001, t-test for independent samples). Both females displayed biting and evading at high frequency, and a dominant and subordinate individual could not be determined (Fig. 4b, top). Climbing was hardly ever observed, and consequently aggregations were not formed. When pairing two males from different logs, the results were the same; there was no hierarchy, both males were aggressive (Fig. 4b, centre). However, they were less aggressive than the two females, i.e. males displayed less biting and evading, and males also showed climbing behaviour. Consequently some aggregations were formed (10% in 11.5 ± 1.3 min). When pairing a female and a male from different logs, there was intense and equal aggression, at levels comparable to male–male and female– female encounters (Fig. 4b, bottom). Aggregations were
Fig. 3. Behavioural patterns of Euperipatoides rowelli displayed by two individuals: left, dominant behaviours: right, subordinate behaviours. Dominant-aggressive behaviour is followed by subordinate-passive behaviour, except for the behaviour ‘climbing’, which is initiated by the subordinate in order to form an aggregation. Once the dominant individual ‘tolerates the climber’, the two individuals remain motionless in aggregation.
young (Fig. 2). The number of feeding males reached its maximum once the females started leaving the food. All individuals eventually fed, but rarely all simultaneously. After 3 h, feeding was over, and the group reaggregated once the individuals had cleaned their mouthparts and antennae.
Aggression, hierarchy and aggregation formation
How do the individuals of a group establish this hierarchy, and determine their ranks within a group? A number of behavioural interactions were observed between mature individuals, which could be classiﬁed as dominant and subordinate behaviours (Fig. 3): biting: an individual bites another, mostly during frontal encounters; evading: an individual avoids being bitten by turning away; chasing: an individual runs after another, pushing it in the rear; running: an individual tries to escape by running away at high speed; climbing: an individual climbs on top of another from behind. The other individual usually reacts by shaking the climber off, also trying to bite it; tolerating climber: the dominant individual tolerates the body contact with the climber and the
Social behaviour in Onychophora
Female (a) 12 8 * 4 0 12 Frequency (n) 8 4 0 12 8 4 * NS 0 Climbing Climbing Running Running Biting Chasing Biting Chasing Evading Evading NS NS * * 84.4% Aggregation (in 6.4 + 2.8 min) * NS * NS 0% Aggregation * * * NS NS 55.6% Aggregation (in 11.3 + 2.6 min) * * * * NS 10% Aggregation (in 11.5 + 1.3 min) * * * * * * NS 80% Aggregation (in 9.8 + 3.1 min) (b) 0% Aggregation * * Male
Fig. 4. Hierarchy and aggregation formation in Euperipatoides rowelli. (a) Two females (top), two males (centre), or one female and one male (bottom) from the same group were paired in a Petri dish for 15 min. Shown are the mean frequency ± SD (n = 45) of the behaviours biting, chasing, evading, running, and climbing, as displayed by the two individuals during the test. The percentage of aggregations formed and the mean time needed is also shown. (b) The two individuals of a test originated from different groups (n = 20). Paired t-test: *, signiﬁcant difference at least P < 0.05; NS, no signiﬁcant difference.
not formed, in spite of repeated climbing behaviour by the male.
This is the ﬁrst study of its kind showing that onychophorans are capable of surprisingly complex interactive behaviour. We have demonstrated that E. rowelli aggregations are more than random assemblages of individuals; they are social groupings organized in a female-dominated hierarchy. Female dominance is probably because of their larger body size (Sunnucks et al., 2000), which might be the general cue for triggering dominant and subordinate behaviour. Indeed, during pair encounters of E. rowelli individuals, the specimens were frequently observed running their antennae over the entire length of the other individual’s back when walking past. Considering the high number of sensory receptors on the antennae of onychophorans (Storch & Ruhberg, 1993), this might be a means to estimate the other individual’s size. Moreover, in preliminary experiments, where the size of females was not matched, smaller females were
consistently subordinate. Such size-based dominance hierarchies are not unusual and have been described for other invertebrates (Edwards, Issa & Herberholz, 2003). Communal hunting and feeding as a way to overwhelm a meal is typical of predators with relatively large prey. This has been shown for many animals, and the behaviour of E. rowelli is similar to that described for some social spiders (Nentwig, 1985; Rowell & Aviles, 1995). While few detailed studies of prey speciﬁcity have been carried out for onychophorans (Read & Hughes, 1987), there is an enormous size range of potential prey items in their natural habitat, and under laboratory conditions they will attack crickets many times their own size. Simple size-based dominance displays when competing for food have been observed in other invertebrates such as crayﬁsh (Edwards et al., 2003). However, the strict feeding hierarchy within E. rowelli groups, with one individual monopolizing the meal for an extended period has, to our knowledge, never been described for an invertebrate. It is unclear whether the behaviour exhibited by E. rowelli is a characteristic of onychophorans in general or peculiar to this species. As yet no other species have
J. REINHARD AND D. M. ROWELL
been studied in comparable detail. It may be relevant that E. rowelli is found in enormous densities compared to most other onychophoran species investigated (over 1000 individuals/ha on south-east facing slopes) and that in many regions of the forest virtually every habitable log contains specimens of E. rowelli (Barclay, Ash & Rowell, 2000). While this may be a human-induced effect (previous logging has left large volumes of dead wood on the forest ﬂoor), it may be associated with their social behaviour. Perhaps the ability of E. rowelli to tolerate conspeciﬁcs and live in groups, permits this species to exist at higher densities than other species of Onychophora in Tallaganda State Forest, which are sparse in comparison. Tolerance of conspeciﬁcs is essential for evolution of group living (Choe & Crespi, 1997), and in E. rowelli this might be achieved via the dominance hierarchies which control inter-individual aggression levels. The onesided aggression observed during encounters between individuals from the same group, and the fact that subordinate climbing behaviour always led to tolerance of physical contact and aggregation, support this hypothesis. Interestingly, the tolerance displayed by members from the same group was in strong contrast to the highly aggressive encounters between individuals from different groups. Tolerance of conspeciﬁcs in the former situation and the lack thereof in the latter, suggest the existence of some form of group or kin recognition in E. rowelli. The basis of this apparent kin recognition has yet to be determined, but considering the habitat E. rowelli live in, it is likely that chemical cues such as pheromones, cuticular hydrocarbons, or scents from common diets might be involved. The potential role of the crural gland excretion of E. rowelli, which has been shown to attract conspeciﬁcs (Barclay, Rowell & Ash, 2000), is currently under investigation. The intense aggression between females and the somewhat reduced aggression between males from different groups could explain the fact that genetic isolation-by-distance is more marked in E. rowelli females than in males (Sunnucks et al., 2000; Sunnucks & Tait, 2001; Sunnucks pers. comm.). Females usually remain in the same log or do not move very far, and thus their groups become genetically distant even if only a few hundred metres apart, while males disperse in search for new habitats and potential mating partners (Barclay, Rowell & Ash, 2000). However, mating does not usually occur. The somewhat surprising aggression between females and males from different groups could explain the fact that males have little post-dispersal reproductive success. Preliminary results from a more detailed aggression study pairing E. rowelli specimens collected from all over Tallaganda State Forest, indicate that aggression levels depend not only on gender and group, but also on the area/type of collection site as well as geographic distance between sites. The effect of this apparent recognition ability and subsequent aggression on mate choice in E. rowelli, its possible function as isolating mechanism and hence its role in speciation are under investigation.
The evolution of group-living and sociality in animals has been intensively studied resulting in a variety of models (Choe & Crespi, 1997; Dugatkin, 1997). Euperipatoides rowelli sociality might have originated from aggregation of possibly unrelated individuals, because of favourable ecological conditions or limited habitable sites. Factors improving ﬁtness, such as cooperative predation of (and defence against) larger invertebrates, easier mate ﬁnding, better protection and care of juveniles, and control of microhabitat conditions, could have further sustained the evolution of group-living in E. rowelli. Onychophorans are susceptible to desiccation and the close aggregations may help to minimize water loss. This might also explain why aggregations in Petri dishes are much closer than in containers of moist sphagnum moss. Alternatively, the evolution of sociality in E. rowelli might have originated from extended parental care or prolonged association of siblings, and may have been inﬂuenced by kin selection. To explore the potential role of kin selection in the evolution of social behaviour in Onychophora, however, detailed information is required on the degree of relatedness within and between E. rowelli groups. Although the above aspects, especially the genetic relationships, remain to be investigated, the discovery of social behaviour in E. rowelli is of great importance in itself shedding further light on the mystery of onychophoran phylogeny. Social behaviour from sub-, semi- to eusocial is well documented for arthropods, including aphids, thrips, ambrosia beetles, spiders, shrimps and of course termites, wasps, ants and bees (Choe & Crespi, 1997). While the case of E. rowelli is fundamentally different from the complex colonies of eusocial insects, it is very similar to some of the behaviour reported for social spiders. In particular, the colonial spider Delena cancerides shows similar poorly coordinated group participation in prey capture and a marked difference in social interactions between individuals depending on whether they are derived from the same or different colonies, most notably for adult females (Rowell & Aviles, 1995). In D. cancerides, intolerance of outsiders is even more extreme than in E. rowelli, with encounters usually ending with the death of one animal, usually the smaller. The close resemblance to spider social behaviour is of interest given the numerous contradictory placements of the Onychophora in relation to the arthropods. Ongoing studies of the social interactions and behaviour of both groups may shed further light on this controversy.
We thank J. Zeil for technical support, M. Jennions for supplying crickets, and P. Sunnucks’ research group for help with animal collecting. Collection was carried out under NPWS Scientiﬁc Investigation Licence No. S11071 and State Forests of NSW Special Purposes Permit No. 15416. N. Tait and P. Sunnucks provided valuable comments on an earlier draft of the manuscript. This study was partly supported by an Australian Research Council Grant to P. Sunnucks and DMR (DP0211156).
Social behaviour in Onychophora REFERENCES
Aguinaldo, A. M. A., Turbeville, J. M., Linford, L. S., Rivera, M. C., Garey, J. R. & Raff, R. A. (1997). Evidence for a clade of nematodes, arthropods and other moulting animals. Nature (Lond.) 387: 489–493. Ballard, J. W. O., Olsen, G. J., Faith, D. P., Odgers, W. A., Rowell, D. M. & Atkinson, P. W. (1992). Evidence from 12S ribosomal RNA sequences that onychophorans are modiﬁed arthropods. Science 258: 1345–1348. Barclay, S. D., Ash, J. E. & Rowell, D. M. (2000). Environmental factors inﬂuencing the presence and abundance of a log-dwelling invertebrate, Euperipatoides rowelli (Onychophora). J. Zool. (Lond.) 250: 425–436. Barclay, S. D., Rowell, D. M. & Ash, J. E. (2000). Pheromonally mediated colonization patterns in the velvet worm Euperipatoides rowelli (Onychophora). J. Zool. (Lond.) 250: 437–446. Briscoe, D. A. & Tait, N. N. (1995). Allozyme evidence for extensive and ancient radiations in Australian Onychophora. Zool. J. Linn. Soc. 114: 91–102. Choe, J. C. & Crespi, B. J. (1997). Social behavior in insects and arachnids. New York: Cambridge University Press. Curach, N. & Sunnucks, P. (1999). Molecular anatomy of an onychophoran: compartmentalized sperm storage and heterogeneous paternity. Mol. Ecol. 8: 1375–1385. Dugatkin, L. A. (1997). Cooperation among animals: an evolutionary perspective. Oxford: Oxford University Press. Edwards, D. H., Issa, F. A. & Herberholz, J. (2003). The neural basis of dominance hierarchy formation in crayﬁsh. Microsc. Res. Tech. 60: 369–376. Eriksson, B. J. & Budd, G. E. (2000). Onychophoran cephalic nerves and their bearing on our understanding of head segmentation and stem-group evolution of Arthropoda. Arthropod Struct. Dev. 29: 197–209. Eriksson, B. J., Tait, N. N. & Budd, G. E. (2003). Head development in the onychophoran Euperipatoides kanangrensis with particular reference to the central nervous system. J. Morphol. 255: 1–23. Garrick, R. C., Sands, C. J., Rowell, D. M., Tait, N. N., Greenslade, P. & Sunnucks, P. (2004). Phylogeography recapitulates topography: very ﬁne-scale local endemism of a saproxylic ‘giant’ springtail at Tallaganda in the Great Dividing Range of south-east Australia. Mol. Ecol. 13: 3329–3344. Ghiselin, M. T. (1984). Peripatus as a living fossil. In Living fossils: 214–217. Eldredge, N. & Stanley, S. M. (Eds). New York: Springer-Verlag. Gleeson, D. M., Rowell, D. M., Tait, N. N., Briscoe, D. A. & Higgins, A. V. (1998). Phylogenetic relationships among Onychophora from Australasia inferred from the mitochondrial cytochrome oxidase subunit I gene. Mol. Phylogenet. Evol. 10: 237–248. Grosberg, R. K. (1990). Out on a limb: arthropod origins. Science 250: 632–633. Hackmann, R. H. & Goldberg, M. (1975). Peripatus: its afﬁnites and its cuticle. Science 190: 582–583. Hamer, M. L., Samways, M. J. & Ruhberg, H. (1997). A review of the Onychophora of South Africa, with discussion of their conservation. Ann. Natal Mus. 38: 283–312. Havel, J. E., Wilson, C. C. & Hebert, P. D. N. (1989). Parental investment and sex allocation in a viviparous onychophoran. Oikos 56: 224–232. Hebert, P. D. N., Billington, N., Finston, T. L., Boileau, M. G., Beaton, M. J. & Barrette, R. J. (1991). Genetic variation in the
onychophoran Plicatoperipatus jamaicensis. Heredity 67: 221– 229. Nentwig, W. (1985). Social spiders catch larger prey: a study of Anelosimus eximius (Araneae: Arachnida). Behav. Ecol. Sociobiol. 17: 79–85. Read, V. M. St. J. & Hughes, R. N. (1987). Feeding behaviour and prey choice in Macroperipatus torquatus (Onychophora). Proc. R. Soc. Lond. B biol. Sci. 230: 483–506. Reid, A. L. (1996). Review of the Peripatopsidae (Onychophora) in Australia, with comment on peripatopsid relationships. Invertebr. Taxon. 10: 663–936. Rockman, M. V., Rowell, D. M. & Tait, N. N. (2001). Phylogenetics of Planipapillus, lawn-headed onychophorans of the Australian Alps, based on nuclear and mitochondrial gene sequences. Mol. Phylogenet. Evol. 21: 103–106. Rowell, D. M. & Aviles, L. (1995). A new form of social behaviour from an Australian huntsman spider. Insectes Soc. 42: 287– 302. Rowell, D. M., Rockman, M. V. & Tait, N. N. (2002). Extensive Robertsonian rearrangement: implications for the radiation and biogeography of Planipapillus Reid (Onychophora: Peripatopsidae). J. Zool. (Lond.) 257: 171–179. Scott, I. A. W. & Rowell, D. M. (1991). Population biology of Euperipatoides leuckartii (Onychophora: Peripatopsidae). Aust. J. Zool. 39: 499–508. Storch, V. & Ruhberg, H. (1993). Onychophora. In Microscopic anatomy of invertebrates: 11–56. Harrison, F. W. (Ed.). New York: Wiley-Liss. Strausfeld, N. J., Hansen, L., Li, Y., Gomez, R. S. & Ito, K. (1998). Evolution, discovery, and interpretations of arthropod mushroom bodies. Learn. Mem. 5: 11–37. Sunnucks, P., Curach, N. C., Young, A., French, J., Cameron, R., Briscoe, D. A. & Tait, N. N. (2000). Reproductive biology of the onychophoran Euperipatoides rowelli. J. Zool. (Lond.) 250: 447–460. Sunnucks, P. & Tait, N. N. (2001). Velvet worms: expect the unexpected. Nature Australia Winter edn 27: 61–69. Sunnucks, P. & Wilson, A. C. C. (1999). Microsatellite markers for the onychophoran Euperipatoides rowelli. Mol. Ecol. 8: 899– 900. Tait, N. N. (1998). The Onychophora and Tardigrada. In Invertebrate Zoology 9: 204–220. Anderson, D. T. (Ed.). Oxford: Oxford University Press. Tait, N. N. & Briscoe, D. A. (1995). Genetic differentiation within New Zealand Onychophora and their relationships to the Australian fauna. Zool. J. Linn. Soc. 114: 103–113. Tait, N. N., Briscoe, D. A. & Rowell, D. M. (1995). Onychophora – ancient and modern radiations. Mem. Assoc. Australas. Palaeontol. 18: 21–30. Telford, M. J. & Thomas, R. H. (1995). Demise of the Atelocerata? Nature (Lond.) 376: 123–124. Trewick, S. A. (1998). Sympatric cryptic species in New Zealand Onychophora. Biol. J. Linn. Soc. 63: 307–329. Walker, M. H. & Campiglia, S. S. (1998). Spermatheca in gravid and virgin female Peripatus (Macroperipatus) acacioi Marcus and Marcus (Onychophora, Peripatidae). J. Morphol. 237: 127– 136. Walker, M. H. & Tait, N. N. (2004). Studies of embryonic development and the reproductive cycle in ovoviviparous Australian Onychophora (Peripatopsidae). J. Zool. (Lond.) 264: 333–354. Wheeler, W. C., Cartwright, P. & Hayashi, C. Y. (1993). Arthropod phylogeny: a combined approach. Cladistics 9: 1–39.
This action might not be possible to undo. Are you sure you want to continue?
We've moved you to where you read on your other device.
Get the full title to continue reading from where you left off, or restart the preview.