2009 SOUTHEASTERN NATURALIST 8(4):587–598 Time Budgets of Wild Nine-banded Armadillos Kier A. Ancona1 and W. James Loughry1,* Abstract - Dasypus novemcinctus (Nine-banded Armadillo) produces litters of genetically identical quadruplets and, because of this, has long been considered a potential model system for the study of kin selection. However, long-term field studies have failed to reveal any obvious instances of kin-selected altruism in this species. Social interactions, such as altruism, require time and energy. The timeconstraints hypothesis proposes that, because of certain aspects of their biology, Nine-banded Armadillos may have to devote most of their active time to foraging, thus precluding any opportunity for the evolution of kin-selected social behavior. To determine the potential validity of this conjecture, we recorded time budgets of wild Nine-banded Armadillos at a study site in western Mississippi during two summers. Both focal and instantaneous sampling showed armadillos allocated 77–90% of their above-ground active time to foraging. A search of the literature indicated that this is among the highest values reported for any mammal. We interpret these findings as consistent with the time-constraints hypothesis. Introduction Understanding the evolution of social behavior has long been a major area of study for behavioral ecologists. Altruism, the seemingly paradoxical case where an animal decreases its fitness to increase the fitness of another, has been of particular interest. Hamilton (1964) provided a key insight by recognizing that, in many cases, animals should not be indiscriminantly altruistic, but rather target their altruism at genetic relatives in order to gain indirect fitness benefits. Instances of kin-selected altruism have since been documented in a wide array of species (Griffin and West 2002, Queller and Strassman 1998, Sherman 1977). Dasypus novemcinctus L. (Nine-banded Armadillos; hereafter armadillos) exhibit obligate polyembryony, whereby females produce litters of genetically identical quadruplets each time they reproduce (Craig et al. 1997, Enders 2008, Prodöhl et al. 1996). Because siblings are clones of one another, their high coefficient of relatedness has led many authors (e.g., Dawkins 1976) to consider this species an ideal model system for the study of kin selection. However, to date, no positive evidence of kinselected altruism has been reported in armadillos (review in Loughry et al. 2005, McDonough and Loughry 2008). Although many possibilities exist (Loughry et al. 2005), so far no definitive explanation for this surprising result has been identified. 1Department of Biology, Valdosta State University, Valdosta, GA 31698-0015. *Corresponding author - jloughry@valdosta.edu. 588 Southeastern Naturalist Vol. 8, No. 4 Dunbar (1992) was the first to argue that the evolution of social behavior may be dramatically infl uenced by time constraints (see also Lehmann et al. 2007, Pollard and Blumstein 2008, Rasmussen et al. 2008). In essence, he proposed that social interactions require investments of time and energy by the participants. In cases where such time and energy are not available, the evolution of complex social behavior would be precluded. Nine-banded Armadillos may represent a classic example of this phenomenon. Loughry et al. (2005) proposed that several features of armadillos may combine to make this species time-limited and thus constrain opportunities for kin-selected social behavior to evolve. Those features are: (1) armadillos have one of the lowest metabolic rates reported for any placental mammal (McNab 1980); (2) they have a very short active period (4–6 h) and may sleep up to 20 h/day in their burrows (McNab 1980); and (3) they typically feed on various soil invertebrates that are often widely scattered and individually of low-quality (review in McDonough and Loughry 2008). Given this, it seems possible that armadillos do not have the time to be social because they must spend most of their short active period acquiring sufficient food before they retire back to their burrows. The obvious first step in evaluating this time constraints hypothesis is to record the time budgets of individual armadillos. In this study, we did exactly that for a wild population of armadillos in western Mississippi. We then searched the literature for studies of time budgets in other mammals. If armadillos are truly time-limited, then we predicted their proportional allocation of active time to feeding would be substantially higher than that in these other species. Methods Study species Nine-banded Armadillos are medium-sized (≈4 kg adult body weight), burrowing mammals found throughout much of the southern United States (Taulman and Robbins 1996). Adults are usually solitary and mostly active at night (Layne and Glover 1978, 1985; McDonough and Loughry 1997a). Mating occurs in the summer (McDonough 1997), but females delay implantation of the fertilized egg until late fall or early winter (Peppler 2008). Young are born in early spring and typically first come above ground between May–July. Littermates are more social than adults, sharing burrows and foraging together, but litters appear to break up by fall, perhaps due to dispersal or mortality (Loughry and McDonough 2001, McDonough and Loughry 1997b). Study site This study was conducted at the Yazoo National Wildlife Refuge, Hollandale, MS from 14 May–13 July 2007 and 20 May–19 July 2008. 2009 K.A. Ancona and W.J. Loughry 589 General methods Basic protocols for studying the animals followed those developed by McDonough and Loughry (e.g., 2005). Briefl y, animals were caught in long dipnets during nightly censuses. Once caught, animals were sexed, weighed, measured, permanently marked with a passive transponder (PIT) tag, and marked for long-range identification by gluing various colors and shapes of refl ective tape to different regions of the carapace. For adult females, lactational status was also recorded (Loughry and McDonough 1996). Behavioral observations Behavioral data were collected in two complementary ways. First, during the nightly censuses to capture armadillos, instantaneous samples were recorded at first sighting of each animal. These data were collected between 16:00–24:00 h. For animals subsequently captured, information about individual identity (age, sex, lactational status) was also included. For individuals that eluded capture, we still used body size to identify the individual as an adult or juvenile, but other information was lacking. Because some animals did escape capture, it is possible that some individuals were observed more than once on a single night. However, in most cases this was extremely unlikely because we usually did not resample the same areas multiple times and, when we did, animals were widely scattered from one another (generally by several hundred m), making it unlikely the same animal moved a great distance to be resampled (armadillos typically move less than 200 m between successive sightings; see Loughry and Mc- Donough 2001). Focal animal observations constituted the second type of behavioral data. These data were obtained throughout the entire night, from 16:00– 6:00 h (see Ancona 2009 for further details about focal sampling protocols). Focal data were recorded with a handheld PDA (Palm Treo), using custom-designed data-acquisition software that provided the total number of times a behavior was observed as well as the total duration of time (in sec) spent in each behavior. Observations emphasized marked animals, but, as with the instantaneous samples, sometimes this was not possible, in which case we could specify the age of the animal (adult versus juvenile), but nothing else. To avoid biasing the data from over-sampling of certain individuals, samples of marked armadillos were limited to once per week. In cases where we mistakenly violated this rule, or when we resampled an animal because the initial sample was incomplete, we averaged data for an animal within each week. Focal observations lasted a maximum of 10 min. Because many sessions did not last the full time, all data were transformed to percentages of total time observed for analysis. We arbitrarily decided that the minimum duration of a focal sample for inclusion in the data set was 3.0 min (average duration of focals was 423.89 ± 145.03 sec in 2007 and 449.45 ± 151.70 sec in 2008). A list and 590 Southeastern Naturalist Vol. 8, No. 4 Table 1. Definitions of behaviors recorded during instantaneous and focal sampling of wild Nine-banded Armadillos. BehaviorA Definition Amicable Social interaction in which both participants remain in close proximity to one another and engage in occasional physical contacts that are not hostile or related to reproduction. Bipedal sniff Animal stands on hind legs and elevates snout. Considered a component of vigilance (McDonough and Loughry 1995) whereby the animal investigates the environment for potential dangers. Chase One animal runs after another. Considered a component of aggressive interactions (see McDonough 1994). Dig Persistent use of the foreclaws to excavate soil, e.g., in creating a burrow. Does not include brief episodes associated with foraging. Feed Stationary or slow movement with snout to the ground, apparently occurring as the animal searches for prey, and associated with occasional brief bursts of digging as the animal attempts to capture prey items. Fight Often associated with chases as another component of aggression (Mc- Donough 1994). Animals attempt to kick one another with the back legs or scratch each other with the front claws. May also involve aerial tumbling (Denson 1979). In burrow Animal is observed going underground. Mate Two adults of opposite sex maintaining close proximity to one another and engaging in any of the pairing behaviors described by Mc- Donough (1997). Pause Animal is in a motionless quadrupedal stance; often the snout is elevated as if sniffing the air. Considered, along with bipedal sniffing, as a component of vigilance (McDonough and Loughry 1995). Run Rapid locomotion not associated with chases. Walk Consistent, slow locomotion as the animal moves from one location to another. Not a component of the brief movements associated with foraging. AAn additional category—“not visible”—was used during focal sampling for those instances when the animal could not be seen. Note that, for focal data, percentages of time engaged in each behavior were calculated based on the time visible, not the total duration of the observation session (i.e., time not visible was subtracted out of the total duration of the session prior to calculating percentages). definitions of the behaviors recorded in both instantaneous and focal sampling is provided in Table 1. Although our instantaneous data may seem substantial, with over 1000 total observations (see below), recent work by Wilson et al. (2008) suggest that even more observations per individual would be required to obtain precise estimates of armadillo activity budget parameters. Thus, our data may not represent the “true” values for armadillos in the wild. However, the fact that these data were largely congruent with those obtained using more detailed focal data (Table 2) suggests the instantaneous samples were reasonably accurate. 2009 K.A. Ancona and W.J. Loughry 591 Data analyses Instantaneous samples were compiled as proportions of individuals exhibiting each behavior. Because animals could move out of sight during focal sampling, we calculated percentages of time spent in each behavior based on total time visible, not total duration of the observation (see Table 1). For both sets of data, we calculated an overall time budget pooled across all animals and both years of the study (for focal data, multiple observations of the same animal within each year were averaged first). This calculation was necessary to generate a single value for comparison with other mammals. Such pooling of the data was further justified by the fact that Ancona (2009) found few differences in time budgets between individuals or between years of the study. However, the reader should be alerted that, because of some initial confusion in how to define behaviors, focal values for the behaviors walk and dig come from 2008 only. To assess the extent to which armadillos might be time-limited, we searched the literature for reports of time-budget data in other mammals. We limited our search to studies conducted in the last 30 years (i.e., since 1978). A total of 21 journals with emphases on animal behavior, ecology, evolution, and mammalian biology were examined (for a full list, see Ancona 2009). Some data were also obtained from chapters in Vizcaíno and Loughry (2008). For each study, we obtained the average total time spent feeding. To the extent possible, we determined time spent feeding as a percentage of total time active, excluding periods of inactivity such as sleeping. In cases where this was reported separately for different classes of individuals (e.g., males versus females) or different time periods, we calculated an average. Similarly, multiple studies of the same species were averaged to provide a single value. In instances where, for Table 2. Time-budget data from instantaneous and focal sampling of wild armadillos at Yazoo National Wildlife Refuge in 2007 and 2008. Behavior Instantaneous samplesA (n = 1171) Focal samplesB (n = 266) Amicable 0.0 0.03 ± 0.37 Bipedal sniff 0.43 1.11 ± 2.02 Chase 0.68 0.07 ± 0.56 DigC 0.17 0.00 Feed 77.11 89.32 ± 11.71 Fight 0.17 0.09 ± 1.07 In burrow 0.17 0.01 ± 0.13 Mate 4.10 0.12 ± 0.86 Pause 1.28 3.48 ± 5.29 Run 1.79 0.71 ± 2.19 WalkC 14.09 6.36 ± 10.91 AProportion of individuals observed engaged in each activity. BPercentages (± SD) of time animals were observed performing each behavior. CFocal data from 2008 only, n = 134. 592 Southeastern Naturalist Vol. 8, No. 4 various reasons, we could not generate a single averaged value, we present a range. Results General time budget of armadillos Time budget data from instantaneous and focal samples are presented in Table 2. Instantaneous data were derived from 611 observations in 2007 and 560 in 2008. Focal data were calculated from 132 samples in 2007 and 134 in 2008. Together, these data show that armadillos spent the vast majority of their time feeding. Indeed, the only other conspicuous component of the time budget was walking (Table 2), and if we make the reasonable assumption that most bouts of walking entailed moving from one foraging location to another, then virtually all of the active period was dedicated to activities associated with acquiring food. Table 3. Published values for percent of active time spent feeding for various species and orders of mammals. % time SpeciesA Order spent feeding References Yellow-bellied Glider Diprotodontia 85.0 Goldingay 1990 (Petaurus australis) Sugar Glider Diprotodontia 82.0 Jackson and Johnson 2002 (Petaurus breviceps) Nine-banded Armadillo Xenarthra 77.1–89.3 This study (Dasypus novemcinctus) Domestic sheep (Ovis aries) Artiodactyla 73.4 Newman et al. 1994 Black-tailed Prairie Dog Rodentia 71.6 Loughry 1993 (Cynomys ludovicianus) Squirrel Glider Diprotodontia 66.0–78.0 Jackson and Johnson 2002 (Petaurus norfolcensis) Townsend’s Ground Squirrel Rodentia 65.0–97.0 Sharpe and Horne 1998 (Spermophilus townsendii) Mahogany Glider Diprotodontia 40.0–77.0 Jackson and Johnson 2002 (Petaurus gracilis (de Vis)) Order Artiodactyla (n = 10) 10.0–57.0 Carnivora (n = 9) 6.0–60.0 Cetacea (n = 1) 16.0 Chiroptera (n = 1) 17.0 Diprotodontia (n = 6) 10.0–47.1 Perissodactyla (n = 2) 8.6–51.0 Primates (n = 53) 13.4–66.0 Proboscidae (n = 1) 18.9 Rodentia (n = 8) 8.75–54.0 Xenarthra (n = 2) 29.6–46.8 AIndividual species are listed in descending order of time spent feeding for those with values ≥70%. For brevity, the remaining values are given as ranges for all species within each order. See Ancona (2009) for the full list of species, individual values and all references. 2009 K.A. Ancona and W.J. Loughry 593 Comparison with other mammals We found time-budget data for 100 other species of mammals (Table 3). These data indicated that Nine-banded Armadillos spent more time feeding than most other species. Other than the Nine-banded Armadillo, the only species that spent ≥80% of their active time feeding were two marsupials and a ground squirrel (Table 3). Discussion This study represents the first detailed analysis of time budgets in armadillos and confirms the anecdotal impression that active armadillos do nothing but feed. While obviously this is not literally true, our data show that armadillo time budgets were dominated by foraging, with little time devoted to anything else. These findings are consistent with previous, more limited, studies that showed armadillos spend very little time vigilant (McDonough and Loughry 1995), and that, even during reproductive encounters, armadillos still spend much of their time feeding (>90%, McDonough 1997). We interpret these data, coupled with our own, as supporting the hypothesis that armadillos may be precluded from evolving complex social behaviors at least partly because of time constraints. Further support for the time-constraints hypothesis comes from our review of time-budget studies in other mammals. Not only did armadillos spend most of their time feeding, but their allocation of time to foraging was among the highest reported for any mammal. While it was not our intent here to analyze these data for broader phylogenetic patterns (as in Pollard and Blumstein 2008), the fact that armadillos ranked so highly in time spent feeding suggests they would be a prominent outlier in any such comparisons. Indeed, if we take the high end of the ranges reported, armadillos were second only to Spermophilus townsendii Bachman (Townsend's Ground Squirrel) in time spent feeding (Table 3). Of the 101 species examined, these two were the only ones to allocate ≥90% of their time to foraging. However, given the broad range of values for the ground squirrel, it is not clear how often the high end is reached. Thus, the more typical value for time spent feeding in this species might be considerably lower, perhaps even lower than that of armadillos. If so, then the only species with comparable amounts of time spent feeding were two marsupials (Petaurus australis Shaw [Yellowbellied Glider] and Petaurus breviceps Waterhouse [Sugar Glider], see Table 3), which, like armadillos, consistently allocated ≥80% of their time to this behavior. Whether such devotion to feeding leads to constraints on the evolution of sociality in these species, or in any of the others with relatively high values, remains unknown, but suggests a potentially interesting avenue for future work. While our comparative data are consistent with the time constraints hypothesis, we hasten to add that it is equally clear from these data that percent 594 Southeastern Naturalist Vol. 8, No. 4 of time spent foraging cannot be viewed as the sole necessary and sufficient cause that limits the evolution of sociality. As Table 3 documents, some highly social species, such as Cynomys ludovicianus (Ord) (Black-tailed Prairie Dogs) and Ovis aries L. (Domestic Sheep), also spend a great deal of time feeding. It seems likely that another contributing factor is the duration of the active period. For example, Black-tailed Prairie Dogs are active from dawn to dusk (Hoogland 1995). Thus, even if the animals spend over 70% of their time feeding, the long active period would still leave considerable time available for other activities. As mentioned above, such would not be the case for Nine-banded Armadillos, that are only active for a very brief period each day. Regardless, the more general point is that there are a number of potential infl uences on the evolution of sociality (Crespi and Choe 1997). The main value of our study is in highlighting one additional, and often overlooked, contributing factor, namely the time animals have available to engage in social interactions. Several indirect lines of evidence also support the time-constraints hypothesis. First, we collected our data during the summer, a period of time when time budgets might be expected to be the most diverse because of activities associated with reproduction and juvenile emergence from natal burrows. Yet, even so, feeding still dominated time budgets, suggesting armadillos do not reallocate much time to alternate activities. Second, some types of behavior were never observed, the most relevant being play. We observed no instances of play among juvenile armadillos, nor have any been reported in the literature. Burghardt (1988) proposed the excess energy hypothesis for the evolution of play, arguing that the high metabolic rate of endotherms, such as mammals, provides juveniles with an abundance of energy, some of which can be channeled into play. As stated above, armadillos have one of the lowest metabolic rates reported for any placental mammal (McNab 1980), so, unlike most other mammals, they may not be able to “afford” play or, more broadly, other forms of social behavior. Finally, further analyses of our data (Ancona 2009) revealed minimal variation in time budgets due to individual identity (age, sex, and female lactational status), time (year-to-year, seasonal, or daily effects) or the environment (various weather conditions). This limited variation might be interpreted as indicating all armadillos are similarly constrained such that they cannot afford to reduce time foraging to engage in other activities and cannot increase time feeding because they are already engaged maximally in this activity. Most other discussions about the influence of time constraints on the evolution of social behavior have focused on primates (review in Pollard and Blumstein 2008). In contrast to armadillos, where we have argued the important constraint is the time required for feeding, these studies have identified other sources of constraint. For example, Pollard and Blumstein (2008) identified time spent resting (during the day, this did not include 2009 K.A. Ancona and W.J. Loughry 595 nocturnal sleeping) as the most significant influence on sociality in diurnal primates, with increasing requirements for rest leading to smaller group sizes. It seems unlikely that such a scenario would apply in armadillos as there is little evidence that they rest much during the active period (K. Ancona, pers. observ.). Thus it appears that, although many animals may be influenced by time constraints, the exact nature of those constraints may vary considerably. The foregoing suggests a scenario in which the evolutionary history of armadillos serves to limit the potential for the evolution of complex social behavior. Specifically, armadillos possess a thick, tough carapace that may make them relatively free from predation (McBee and Baker 1982). Possession of armor is argued to allow for lower metabolic rates because less energy is needed for allocation to antipredator defenses (Lovegrove 2001). Lower metabolism might result in shorter active periods (McNab 1980) and exploitation of lower quality prey. Reliance on low quality (and, in this case, widely scattered) prey might then require that most of the active period be devoted to feeding, to the exclusion of most social interactions. We stress that this is highly speculative and we do not mean to imply any particular set of causal links. Rather, our intent is to point out that there seems to be a constellation of traits exhibited by armadillos that appear to work in concert to produce a relatively asocial animal in which complex sociality is precluded. Such traits are not unique to Nine-banded Armadillos, but rather are common features shared by most members of Cingulata (the order containing the 21 extant species of armadillos). We are unaware of any time-budget studies that have been conducted in other cingulates but, if the time-constraints hypothesis has general validity, then we predict that future such studies will reveal time budgets dominated by feeding, just as we have documented here. Because our study is entirely descriptive, we can offer no definitive support for the time-constraints hypothesis at present. However, the patterns we report are consistent with the notion that armadillos are time-limited. Further testing of the hypothesis will require experimental manipulations (e.g., food-supplementation studies) as well as more rigorous comparative analyses to identify possible trends among species. Nonetheless, our study suggests that, at least for armadillos, the timeconstraints hypothesis has some merit. We hope our findings will spur attempts to test this hypothesis more fully. Acknowledgments We thank the staff of the Yazoo National Wildlife Refuge for their very generous support of this project. Partial funding for this study came from a National Geographic grant, Valdosta State University (VSU) Faculty Research Awards, and a grant from the VSU Center for Applied Research (all to W.J. Loughry). We are extremely grateful to Jim Ha for creating the data-acquisition software and to Missy 596 Southeastern Naturalist Vol. 8, No. 4 Ard, Lynda Bernhardt, Rachel Morgan, and Brian Spychalski for their outstanding assistance in the field. Finally, we thank Mike Conner, Mitch Lockhart, Colleen McDonough, David Reed, and two anonymous reviewers for comments on earlier versions of this paper. Literature Cited Ancona, K.A. 2009. Time-budget analyses of wild Nine-banded Armadillos. M.Sc. Thesis. Valdosta State University, Valdosta, GA. 85 pp. Burghardt, G.M. 1988. Precocity, play, and the ectotherm-endotherm transition: Profound reorganization or superficial adaptation? Pp. 107–148, In E.M. Blass (Ed.). Handbook of Behavioral Neurobiology. Volume 9. Developmental Psychobiology and Behavioral Ecology. Plenum Press, New York, NY. 453 pp. Craig, S.F., L.B. Slobodkin, G.A. Wray, and C.H. Biermann. 1997. The “paradox” of polyembryony: A review of the cases and a hypothesis for its evolution. Evolutionary Ecology 11:127–143. Crespi, B.J., and J.C. Choe. 1997. Explanation and evolution of social systems. Pp. 499–524, In J.C. Choe and B.J. Crespi (Eds.). The Evolution of Social Behavior in Insects and Arachnids. Cambridge University Press, Cambridge, UK. 541 pp. Dawkins, R. 1976. The Selfish Gene. Oxford University Press, New York, NY. 224 pp. Denson, R.D. 1979. Aggression and tumbling among armadillos. Southwestern Naturalist 24:697–698. Dunbar, R.I.M. 1992. Time: A hidden constraint on the behavioural ecology of baboons. Behavioral Ecology and Sociobiology 31:35–49. Enders, A.C. 2008. Placentation in armadillos, with emphasis on development of the placenta in polyembryonic species. Pp. 172–180, In S.F. Vizcaíno and W.J. Loughry (Eds.). The Biology of the Xenartha. University Press of Florida, Gainesville, FL. 370 pp. Goldingay, R.L. 1990. The foraging behavior of a nectar-feeding marsupial, Petaurus australis. Oecologia 85:191–199. Griffin, A.S., and S.A. West. 2002. Kin selection: Fact or fiction. Trends in Ecology and Evolution 17:15–21. Hamilton, W.D. 1964. The genetic evolution of social behaviour (I and II). Journal of Theoretical Biology 7:1–52. Hoogland, J.L. 1995. The Black-tailed Prairie Dog: Social Life of a Burrowing Mammal. University of Chicago Press, Chicago, IL. 557 pp. Jackson, S.M., and C.N. Johnson. 2002. Time allocation to foraging in the Mahogany Glider Petaurus gracilis (Marsupialia: Petauridae) and a comparison of activity times in exudivorous and fl ivorous possums and gliders. Journal of Zoology 256:271–277. Layne, J.N., and D. Glover. 1978. Activity cycles of the Nine-banded Armadillo in southern Florida. P. 140, In R. Obrtel, C. Folk, and J. Pellantova (Eds.). Second International Theriological Congress. Institute of Vertebrate Zoology, Brno, Czechoslovakia. 461 pp. Layne, J.N., and D. Glover. 1985. Activity patterns of the Common Long-nosed Armadillo Dasypus novemcinctus in south-central Florida. Pp. 407–417, In G. G. Montgomery (Ed.). The Evolution and Ecology of Armadillos, Sloths, and Vermilinguas. Smithsonian Institution Press, Washington, DC. 451 pp. 2009 K.A. Ancona and W.J. Loughry 597 Lehmann, J., A.H. Korstjens, and R.I.M. Dunbar. 2007. Group size, grooming, and social cohesion in primates. Animal Behaviour 74:1617–1629. Loughry, W.J. 1993. Determinants of time allocation by adult and yearling Blacktailed Prairie Dogs. Behaviour 124:23–43. Loughry, W.J., and C.M. McDonough. 1996. Are road kills valid indicators of armadillo population structure? American Midland Naturalist 135:53–59. Loughry, W.J., and C.M. McDonough. 2001. Natal recruitment and adult retention in a population of Nine-banded Armadillos. Acta Theriologica 46:393–406. Loughry, W.J., P.A. Prodöhl, and C.M. McDonough. 2005. The inadequacy of observation: Understanding armadillo biology with molecular markers. Pp. 55–73, In S.G. Pandalai (Ed.). Recent Research Developments in Ecology. Transworld Research Network, Kerala, India. 73 pp. Lovegrove, B.G. 2001. The evolution of body armor in mammals: Plantigrade constraints of large body size. Evolution 55:1464–1473. McBee, K., and R.J. Baker. 1982. Dasypus novemcinctus. Mammalian Species 162:1–9. McDonough, C.M. 1994. Determinants of aggression in Nine-banded Armadillos. Journal of Mammalogy 75:189–198. McDonough, C.M. 1997. Pairing behavior of the Nine-banded Armadillo (Dasypus novemcinctus). American Midland Naturalist 138:290–298. McDonough, C.M., and W.J. Loughry. 1995. Infl uences on vigilance in Nine-banded Armadillos. Ethology 100:50–60. McDonough, C.M., and W.J. Loughry. 1997a. Infl uences on activity patterns in a population of Nine-banded Armadillos. Journal of Mammalogy 78:932–941. McDonough, C.M., and W.J. Loughry. 1997b. Patterns of mortality in a population of Nine-banded Armadillos, Dasypus novemcinctus. American Midland Naturalist 138:299–305. McDonough, C.M., and W.J. Loughry. 2005. Impacts of land-management practices on a population of Nine-banded Armadillos in northern Florida. Wildlife Society Bulletin 33:1198–1209. McDonough, C.M., and W.J. Loughry. 2008. Behavioral ecology of armadillos. Pp. 281–293, In S.F. Vizcaíno and W.J. Loughry (Eds.). The Biology of the Xenartha. University Press of Florida, Gainesville, FL. 370 pp. McNab, B.K. 1980. Energetics and the limits to a temperate distribution in armadillos. Journal of Mammalogy 61:606–627. Newman, J.A., P.D. Penning, A.J. Parsons, A. Harvey, and R.J. Orr. 1994. Fasting affects intake behavior and diet preference of grazing Sheep. Animal Behavior 47:185–193. Peppler, R.D. 2008. Reproductive biology of the Nine-banded Armadillo. Pp. 151–159, In S.F. Vizcaíno and W.J. Loughry (Eds.). The Biology of the Xenartha. University Press of Florida, Gainesville, FL. 370 pp. Pollard, K.A., and D.T. Blumstein. 2008. Time allocation and the evolution of group size. Animal Behaviour 76:1683–1699. Prodöhl, P.A., W.J. Loughry, C.M. McDonough, W.S. Nelson, and J.C. Avise. 1996. Molecular documentation of polyembryony and the micro-spatial dispersion of clonal sibships in the Nine-banded Armadillo, Dasypus novemcinctus. Proceedings of the Royal Society, London, Series B 263:1643–1649. 598 Southeastern Naturalist Vol. 8, No. 4 Queller, D.C., and J.E. Strassman. 1998. Kin selection and social insects. Bioscience 48:165–175. Rasmussen, G.S.A., M. Gusset, F. Courchamp, and D.W. Macdonald. 2008. Achilles’ heel of sociality revealed by energetic poverty trap in cursorial hunters. American Naturalist 172:508–518. Sharpe, P.B., and B.V. Horne. 1998. Influence of habitat on behavior of Townsend's Ground Squirrels (Spermophilus townsendii). Journal of Mammology 79:906–918. Sherman, P.W. 1977. Nepotism and the evolution of alarm calls. Science 197:1246–1253. Taulman, J.F., and L.W. Robbins. 1996. Recent range expansion and distributional limits of the Nine-banded Armadillo (Dasypus novemcinctus) in the United States. Journal of Biogeography 23:635–648. Vizcaíno, S.F., and W.J. Loughry (Eds.). 2008. The Biology of the Xenartha. University Press of Florida, Gainesville, FL. 370 pp. Wilson, R.R., B.D. Jansen, and P.R. Krausman. 2008. Planning and assessment of activity budget studies employing instantaneous sampling. Ethology 114:999– 1005.