© 2001 by The Society for Integrative and Comparative Biology
Modeling the Metabolic Energetics of Brief and Intermittent Locomotion in Lizards and Rodents1
1 Section of Integrative Physiology and Neurobiology, Environmental, Population, and Organismic Biology, University of Colorado, Boulder, Colorado 80309-0334
 |
SYNOPSIS |
|---|
TOP
 SYNOPSIS INTRODUCTIONDEFINING THE COST OF...COSTS OF INTERMITTENT ACTIVITY:...SUMMARYReferences |
|---|
When locomotor activity is brief, physiological steady state conditions are not attained. It is therefore difficult to model the energetic costs of intermittent activity using standard methods. This difficulty is addressed by considering as reflective of the metabolic costs of activity not only the oxygen consumed during the activity itself, but also the excess post-exercise oxygen consumption (EPOC) and any excess metabolites persisting at the end of EPOC. This paper briefly reviews the metabolic events associated with EPOC, and then examines how this approach can be applied to address questions of how behavioral variables associated with locomotion (activity duration, intensity, frequency) can influence the energetic costs to the animal per unit distance. Using data for lizards, mice, and others, EPOC can be shown to be the major component of energetic costs when durations are short, regardless of exercise intensity. Brief activity is much more expensive by this measure than is steady state locomotion, regardless of phylogeny or body mass. Three studies of intermittent locomotion provide evidence that brief behaviors can be undertaken at lower metabolic costs than predicted from single bouts of activity when repeated in a frequent, repeated pattern. Metabolic savings appear greatest when the pause period between behaviors is short relative to EPOC duration, the time for organismal metabolic rate to return to pre-exercise levels, although longer pause periods may increase endurance.
|
INTRODUCTION |
|---|
|
TOP
SYNOPSISINTRODUCTIONDEFINING THE COST OF...COSTS OF INTERMITTENT ACTIVITY:...SUMMARYReferences |
|---|
Estimating the metabolic costs of animal locomotor activities has been an active area of integrative biology for 25 yr. Most efforts derive from measurements of aerobic metabolism while animals are locomoting under steady state conditions, conditions which lend themselves well to laboratory measurement but often fall short in modeling animal behaviors that are intermittent or brief in duration. Intermittent locomotion is often characterized as behavior that is brief, sometimes strenuous, and repeated at frequent intervals. Examples include territorial defense and foraging behavior of many lizards (Huey and Pianka, 1981
), birds (McLaughlin, 1989), rodents (Madison, 1985), and migratory locomotion in ghost crabs (Weinstein and Full, 1984). O'Brien and his colleagues (1990) have characterized this type of foraging behavior as saltatory behavior, and classifies most animal behaviors in that category. In fact, a casual observer will note that many to most natural locomotor behaviors are intermittent in nature. We have recently engaged in several studies that attempt to model the terrestrial locomotor energetics of brief and intermittent behavior. In the context of this article, brief refers to running or walking behaviors of 5 sec to 60 sec in duration. In this review, we describe an alternative approach to estimating locomotor energetics when the behavior or the physiological response to the behavior violates steady-state assumptions. We then describe some recent studies that document how activity costs change as activity duration or activity intensity vary. Finally, we apply this approach to examine three examples of intermittent behavior. These three examples indicate that animals can incur high metabolic costs per unit distance when their behavior is a brief, unitary event, but that animals enjoy considerable energetic savings when the behavior is repeated as an intermittent behavior. These observations are relevant to ecologists and behaviorists attempting to model the costs of brief and intermittent field behaviors, and they also provide testable hypotheses of how intermittent behaviors might have evolved if energetic costs have been a selective pressure on behavior. |
DEFINING THE COST OF ACTIVITY, CACT |
|---|
|
TOP
SYNOPSISINTRODUCTIONDEFINING THE COST OF...COSTS OF INTERMITTENT ACTIVITY:...SUMMARYReferences |
|---|
Whereas traditional estimates of locomotor costs (Cost of transport, cost of locomotion, Mrun, Cmin, etc.) have considered the costs incurred during locomotion as a function of distance covered (Taylor et al., 1970
; Tucker, 1970; Schmidt-Nielsen, 1972; Full et al., 1990), the definition of the cost of activity attempts to include all metabolic costs incurred by the animal without regard to whether they are externally expressed during or after the cessation of activity. A common misunderstanding is that this represents an alternative way to measure the cost of locomotion. It does not. Estimating the cost of locomotion is most accurately estimated from steady state measurements as referred to above. The approach that we describe below is a way to estimate the energetic cost to the animal of engaging in a non-steady state activity. This is accomplished by evaluating both the metabolic elevation that occurs during the activity itself along with the metabolic elevation that follows. The components considered as representing activity costs are illustrated in Figure 1 for a 70 g lizard moderately walking for 1 min, followed by a post-exercise recovery period. Metabolic expenditure, reflected as the excess oxygen consumption incurred during steady-state exercise, is usually the metabolic cost considered in estimates of costs of steady state locomotion (Tucker, 1970; Full et al., 1990). This expenditure is equivalent to the volume of oxygen labeled EEOC (excess exercise oxygen consumption, Fig. 1). While this component is an adequate estimate of metabolic costs of a locomotor behavior when the animal's behavior and physiology are stable over a long period of time and anaerobic energy production is negligible, when behavior is brief or irregular the EEOC volume may significantly underestimate the actual metabolic expenditure of the animal.
|
Inclusion of the excess post-exercise oxygen consumption (EPOC, sensu Gaesser and Brooks, 1984
; Fig. 1) in the estimate of metabolic expenditure associated with an activity includes all costs reflected in oxidative metabolism between the onset of exercise and the end of the recovery period. The addition of EPOC volume is significant, as EPOC is often 
9x the EEOC volume when activity is 60 sec or less in duration (Baker and Gleeson, 1998; Hancock et al., 2001). The approach of considering both EEOC and EPOC expenditures as reflective of locomotor costs has been used by exercise physiologists for decades (see Bahr, 1992 for introduction) and by comparative physiologists when non-steady state conditions prevail (Full and Herreid, 1984). An earlier attempt in our lab to express these costs per unit distance traveled for non-sustainable activities (Wagner and Gleeson, 1996) was recently expanded to apply to any activity regardless of the duration of activity or the sustainability of the effort (Baker and Gleeson, 1998, 1999).
Metabolic costs of activity reflected in the EPOC volume component vary depending upon species and activity level. High energy phosphate resynthesis, replenishment of oxygen stores, metabolite removal and re-cycling, regulated gene expression, and hormonal stimulation are all thought to contribute to the EPOC volume. Several recent accountings of EPOC components that contribute to EPOC in mammals can be found (Gaesser and Brooks, 1984; Bahr, 1992; Fitts, 1994; Neufer, 1999). For purposes of this article, it matters not so much how each of the specific components of EPOC contribute to total costs as does the recognition that those costs, such as phosphocreatine resynthesis for example, are legitimate energetic expenses incurred by the animal because of its activity. To the extent that these expenses are reflected in the oxidative metabolism of the animal, they are accounted for by including EPOC in an energetic assessment. Only if activity-induced metabolite accumulations or depressions are not remedied during the period of time defined by the EPOC duration is an additional computation required. Most EPOC durations are long relative to the time period generally ascribed to phosphagen replenishment (EPOC duration = 7–19 min in small rodents, Baker and Gleeson, 1998, 1999; Hatta et al., 1994; 15–120 min in lizards; Gleeson, 1979; Wagner and Gleeson, 1996; Hancock et al., 2001). The relatively long period covered by EPOC measurement leaves residual lactate accumulation as the most common source of additional costs not reflected in EPOC, and then generally only in ectotherms where lactate removal rates are long relative to endotherms (Gleeson, 1991, 1996).
Thus the cost of activity (Cact, energy expended per unit distance traveled) can be expressed as
|
|
Where EEOC and EPOC represent the energetic equivalence of the mass-specific oxygen consumption elevated above resting levels during exercise and post-exercise, respectively, 
[X] represents the energetic contribution reflected in any difference in concentration of metabolite X between the pre-exercise state and the state of the animal at the end of EPOC, and distance traveled is the distance moved during the locomotor behavior under consideration. Cost of activity can have different units of energy utilization and distance (kcals/g/m, kJ/g/km, etc.), however the more familiar term cost of locomotion is generally expressed as mls O2/g/km, and that convention is continued here.
Duration and intensity of activity can influence Cact. Cost of activity is highest at the shortest activity durations, and decreases logarithmically as the activity duration increases. At very short activity durations in lizards (5–15 sec), costs can be 4–7 times more expensive than predicted from traditional steady state approaches (Hancock et al., 2001). In mice the costs may be 60–200 times sustained locomotor costs when durations are 5–15 sec (Baker and Gleeson, 1998). Preliminary data from mammals ranging from mice to horses (Keng et al., 1999; unpublished data) support this general finding that short duration activity is relatively costly per unit distance. Figure 2 summarizes the relationship between cost and duration for lab mice.
|
In most cases the intensity of locomotor activity, that is, how fast an animal locomotes, has no significant effect on Cact when duration is held constant. The exception may be in mice. In laboratory mice when activity durations are brief (15–60 sec) and intensity low (25–100% of maximal aerobic speed), there is no real change in excess oxygen consumption as running intensity increases (Baker and Gleeson, 1999
). Since they run farther as they run faster however, Cact decreases several-fold as activity intensity increases. The more general response, however, is that total metabolic costs and distance increase in proportion, rendering Cact independent of intensity. This is the case in lizards, horses, and humans running for short durations of 100 sec or less (Zanconato et al., 1991; Keng et al., 1999; Hancock and Gleeson, unpublished data).
In summary, the inclusion of all metabolic costs into an analysis of locomotor activities suggests that activities that are short in duration are costly per unit distance relative to estimates based on continuous locomotion. The energetic analysis of single bouts of brief activity such as those summarized above would predict that the cost of repeated brief activity would be extraordinarily high. For example, Kenagy and Hoyt (1989) report data which suggest that golden-mantled ground squirrels run (for periods of 15 sec or less) approximately 6.5 times per day. This intermittent behavior is compressed into a two hour activity period. A conservative, mass corrected, extrapolation of the data from mice (Baker and Gleeson, 1998) to squirrels predicts an energetic cost equivalent to three times the estimated daily energy expenditure of the squirrel (Kenagy and Hoyt, 1989). Clearly, costs of this magnitude cannot be possible. As this simple example illustrates, intermittently performed behavior must be performed at lower costs than those predicted from single bouts of similar activity. The remainder of this paper summarizes three studies that address this apparent economy associated with intermittent locomotion.
|
COSTS OF INTERMITTENT ACTIVITY: THREE EXAMPLES |
|---|
|
TOP
SYNOPSISINTRODUCTIONDEFINING THE COST OF...COSTS OF INTERMITTENT ACTIVITY:...SUMMARYReferences |
|---|
To test whether or not the number of repeated bouts would influence energetic costs, Edwards recently conducted an experiment with mice forced to run maximally for 15 sec either once, twice, 3, 5, 9 or 13 times within a 6.25 min period, after which the animals were allowed to recover fully (Edwards and Gleeson, 2001
). Distance run and 
O2 were measured throughout each experiment. She found that although animals ran farther as the number of bouts increased, the total metabolic cost, represented by the sum of EEOCs and the EPOCs of the inter-bout and the final recovery periods, actually remained unchanged. As a result metabolic expenditure per unit distance traveled declined. Energetically, the costs were very similar to the costs incurred by mice running the same distances in single bouts of longer duration (Baker and Gleeson, 1998). Figure 3 summarizes these findings, which illustrates two features of intermittent locomotor energetics. The first is that the energetic costs of frequent activity are less than predicted by the summed costs of several individual bouts of the same activity performed separately. The second is that intermittent activity of this type appears to be no more costly per unit distance than would a period of constant activity that covered the same cumulative distance. That is, an animal may expend no more energy to travel between points A and B whether it travels the distance intermittently or all at one time. It is important to qualify these conclusions by recognizing that the pause period between periods of activity in this experiment were short (15–345 sec) relative to the roughly 19 min required for mice to fully recover from 5 sec bursts of activity (Baker and Gleeson, 1998). As a result of this short pause period, mice do not incur the full cost of EPOC during each pause period.
|
A second example comes from a laboratory experiment that modeled the intermittent field behavior of foraging prairie voles. Voles of the genus Microtus exhibit a characteristic foraging behavior whereby they intersperse periods of intermittent locomotor behavior with periods of quiescence (Madison, 1985
). Using the prairie vole (M. ochragaster), we attempted to reproduce this behavior in the laboratory in order to assess the costs of intermittent behavior in a more behaviorally relevant context. Madison tracked the activity pattern of tagged voles in the field over three 24 hr periods (Fig. 1 of Madison, 1985). This field behavior predicts a pattern of seven brief bouts of locomotion dispersed over a two hour period. We have modeled that behavior as seven 5 sec bouts of running distributed 15 min apart, with a 30 min pause arbitrarily placed between active bout 4 and 5 to represent the irregular pattern seen in the field (Table 1). Oxygen consumption was measured throughout as described by Baker and Gleeson (1998). The rate of oxygen consumption of 6 animals subjected to the protocol described in Table 1 was elevated with each bout of brief activity, and would return to levels not different from rest during the 15 min rest period between bouts (Fig. 4). M. ochragaster demonstrated EPOC durations of 5.3 ± 1.8 min, a rapid recovery relative to laboratory mice (Baker and Gleeson, 1998). Hence, unlike the mouse example described above, voles fully recovered between each bout of activity. Although such a scenario should result in maximizing costs because they experience a full recovery associated with each activity bout, voles also demonstrate an energetic economy relative to the cost of a single bout of activity. Summarized in Table 2, prairie voles significantly reduce their metabolic costs per unit distance by about one third relative to the costs of a single 5 sec bout of activity (5 vs. 8.4 mls O2/g/km, Table 2). The mechanism for this savings appeared to be in reduced EPOC volumes seen following repeated behaviors relative to that after a solitary behavior. The interbout and recovery O2 of the voles engaged in intermittent activity is significantly less than predicted from the O2 associated with a solitary bout of 5 sec activity (paired t = 2.42, P = 0.03). This observation has also been reported in lizards (Scholnick and Gleeson, 2000). These costs are still significantly greater than the cost of locomotion predicted for an animal of their body mass (43 g; 1.9 mls O2/g/km, Taylor et al. 1970), but which is based upon an assumption of continuous locomotion and consequently does not consider the recovery (EPOC) component. The ecological implication of these data is that the estimate of 24 hr foraging costs for these animals is about 2.6x higher than previously predicted from estimates of the cost of locomotion rather than the cost of activity. Like the mouse example, this example also demonstrates that activity costs become more economical when brief behaviors are consolidated into a short time period. This protocol differed from the mouse example described earlier in that the pause period in this example was long (15 min) relative to the period required for vole EPOC recovery (5.3 min) and the voles incurred the full cost of EPOC with each bout of activity. As a consequence, although the overall Cact is low relative to predictions from a single bout of activity, it is higher than we would predict for an animal active continuously for 35 sec (seven times 5 sec of activity). This suggests that an important element in characterizing intermittent locomotion from an energetic perspective may be the consideration of the pause period relative to the period required for EPOC recovery.
|
|
) interspersed between 15 min of recovery. Animals recover between each bout of running. Mean of 6 animals, 43.4 ± 2.2 g mass
|
The third example compares the impact of intermittent locomotion on both endurance characteristics and Cact in the lizard Dipsosaurus dorsalis. This example builds on the work of Randi Weinstein and Bob Full who have shown that intermittent locomotor behavior of different run-pause intervals influences the locomotor endurance of both crabs (Weinstein and Full, 1992
) and lizards (Weinstein and Full, 1999). As Weinstein discusses elsewhere in this volume (Weinstein, 2000), animals appear to have run-pause optima that maximize locomotor endurance or distance traveled. How optimization of the pattern of Intermittent behavior relates to the recovery costs that animals incur is unknown for any animal. This relationship was examined in the desert iguana, which does not locomote intermittently in nature, but readily does so under laboratory conditions. In this context Dipsosaurus is used merely to model the behavior. Using a moderately vigorous running intensity of 4x maximum aerobic speed for the species (John-Alder and Bennett, 1981), animals were tested at run-pause ratios of 1:1, 1:2, 1:4, and 1:8, while the run period was set to either 5, 15, or 30 sec duration. Endurance was determined as the length of time each animal was able to run while maintaining the pre-set speed on the treadmill. Increasing the run-pause duration when animals were forced to run for periods of either 15 sec or 30 sec resulted in significant but behaviorally very modest increases (
30%) in endurance (data not shown). In contrast, when lizards were run for only 5 sec durations and pause duration was increased, increases in endurance of 300% were measured (Fig. 5). It is probably not coincidental that the average run duration of Dipsosaurus in the field is approximately 5 sec (Hancock et al., 2001). With 5 sec run times identified as optimal, animals were then fitted with small masks and rates of gas exchange measured following the techniques of Hancock et al. (2001). Gas exchange was measured during six 5 sec bursts of running (30 sec total) followed by interspersed pause periods of 1, 4, or 8 times the run period (5, 20, or 40 sec, respectively). We found that the metabolic costs of intermittent locomotion increased as the pause period lengthened in a manner similar to the increase observed in endurance (Fig. 5). We speculate that the increased endurance as pause periods lengthen is due in part to the more complete recovery of muscle high energy phosphate stores disturbed with each bout of activity, as is seen in muscles of humans subjected to run-pause periods of different lengths (Saltin and Essen, 1971). These increased metabolic costs may reflect the aerobic costs of recovery during the longer pause phase, and this hypothesis is currently being investigated.
|
As was found with the mice and voles, six repeated bouts of 5 sec duration in lizards resulted in lower overall activity costs (3.5–5 ml O2/g/km depending upon run-pause ratio) than predicted from a single bout of 5 sec activity at the same intensity (
6.4 ml O2/g/km; Hancock and Gleeson, unpublished data), and like the mice Cact was not different than that predicted from a single 30 sec period of exercise at the same intensity followed by recovery. This finding in an ectotherm suggests that the conclusion reached earlier from studies of mice, namely that intermittent activity characterized by short pause periods may be no more costly to an animal than when the animal conducts the locomotor behavior as a single, longer bout, is a general characteristic of behavioral energetics. |
SUMMARY |
|---|
|
TOP
SYNOPSISINTRODUCTIONDEFINING THE COST OF...COSTS OF INTERMITTENT ACTIVITY:...SUMMARYReferences |
|---|
The data indicate that the metabolic cost of brief (
60 sec) locomotor activities is significantly more costly per unit distance traveled to animals than are costs estimated from steady state locomotion. The high cost of activity (Cact) is dominated by the costs reflected in the excess post-exercise oxygen consumption (EPOC) that immediately follows a burst of activity. The cost of activity is sensitive to activity duration, being highest at the shortest durations, but is relatively insensitive to activity intensity in most animals. Although the high Cact for single periods of brief activity would suggest that intermittently repeated behavior would be prohibitively expensive, the data from mice, voles, and lizards indicate otherwise. Both rodents and lizards demonstrate an economy of intermittent locomotion that suggests that it may often be no more expensive to travel between points A and B in a series of frequently repeated bursts of locomotion than it is to travel the entire distance as a single event. This is particularly true when the pause period between activity periods is short relative to the length of time required for post-exercise oxygen consumption to return to pre-exercise levels. Longer pause periods may result in greater costs of activity, but may afford the animal greater endurance capacity. We currently lack data that can explain the physiologic explanation for why intermittent locomotion is less expensive than predicted from single bouts of activity, and this is worthy of study. The data from horses and rodents and from a reptile also suggest that there are allometric and phylogenetic differences to the response to brief activity that merit further evaluation so that these data may be interpreted more generally and applied in a more behavioral context.
|
ACKNOWLEDGMENTS |
|---|
We thank Dr. Randi Weinstein, Kelly Kirlin, and Dr. Carlos Crocker for their significant contributions to the design, conduct, or interpretation of some of the work discussed, and we thank Emily Baker for sharing her unpublished data. We also thank Dr. Bruce Wunder for assistance in obtaining M. ochragaster and directing us to the appropriate literature on their ecology. Supported by NSF 97240140.
|
FOOTNOTES |
|---|
1 From the Symposium Intermittent Locomotion: Integrating the Physiology, Biomechanics and Behavior of Repeated Activity presented at the Annual Meeting of the Society for Integrative and Comparative Biology, 4–8 January 2000, at Atlanta, Georgia.
2 E-mail: gleeson{at}colorado.edu
|
References |
|---|
|
TOP
SYNOPSISINTRODUCTIONDEFINING THE COST OF...COSTS OF INTERMITTENT ACTIVITY:...SUMMARYReferences |
|---|
Bahr, R. 1992. Excess post-exercise oxygen consumption—magnitude, mechanisms and practical implications. Acta Physiol. Scand, 605:(Suppl.)1-70.[Medline]
Baker, E. J., and T. T. Gleeson. 1998. EPOC and the energetics of brief locomotor activity in Mus domesticus. J. Exp. Zool, 280:114-120.[CrossRef][Web of Science][Medline]
Baker, E. J., and T. T. Gleeson. 1999. The effects of intensity on the energetics of brief locomotor activity. J. Exp. Biol, 202:3081-3087.[Abstract]
Edwards, E. J., and T. T. Gleeson. 2001. Can energetic expenditure be minimized by performing activity intermittently? J. Exp. Biol, 204:599-605.[Abstract]
Fitts, R. H. 1994. Cellular mechanisms of muscle fatigue. Physiol. Rev, 74:49-94.
Full, R. J., and C. F. Herreid. 1984. Fiddler crab exercise: The energetic cost of running sideways. J. Exp. Biol, 109:141-161.
Full, R. J., D. A. Zuccarello, and A. Tullis. 1990. Effect of variation in form on the cost of terrestrial locomotion. J. Exp. Biol, 150:233-246.
Gaesser, G. A., and G. A. Brooks. 1984. Metabolic bases of excess post-exercise oxygen consumption: A review. Med. Sci. Sports, 16:29-43.
Gleeson, T. T. 1991. Patterns of metabolic recovery from exercise in amphibians and reptiles. J. Exper. Biol, 160:187-207.
Gleeson, T. T. 1996. Post-exercise lactate metabolism: A comparative review of sites, pathways, and regulation. Ann. Rev. Physiol, 58:565-581.[CrossRef][Web of Science][Medline]
Hancock, T. V., S. C. Adolph, and T. T. Gleeson. 2001. Effect of duration on EPOC and costs of short vigorous activity in the desert iguana (Dipsosaurus dorsalis). J. Comp. Physiol. (In press).
Huey, R. B., and E. R. Pianka. 1981. Ecological consequences of foraging mode. Ecology, 62:991-999.[CrossRef][Web of Science]
John-Alder, H. B., and A. F. Bennett. 1981. Thermal dependence of endurance and locomotory energetics in a lizard. Am. J. Physiol, 241:R342-R349.
Kenagy, G. J., and D. F. Hoyt. 1989. Speed and time-energy budget for locomotion in golden-mantled ground squirrels. Ecology, 70:1834-1839.[CrossRef]
Keng, P., S. J. Wickler, T. T. Gleeson, E. J. Baker, and T. V. Hancock. 1999. The cost of brief locomotor activity: A measure of excess post-exercise oxygen consumption (EPOC) in Equus callabus. FASEB Journal, 13: A1050.
Madison, D. M. 1985. Activity rhythms and spacing. In R. H. Tamaroin (ed.), Biology of new world Microtus, pp. 373–419. Spec. Publ. 8., American Society of Mammalogy.
McLaughlin, R. L. 1989. Search modes of birds and lizards: Evidence for alternative movement patterns. Am. Nat, 133:654-670.[CrossRef]
Neufer, P. D. 1999. Contractile activity and skeletal muscle gene expression. In M. Hargreaves and M. Thompson (eds), Biochemistry of exercise X. Human Kinetics, Publ., Auckland.
O'Brien, W. J., H. I. Brownman, and B. I. Evans. 1990. Search strategies of foraging animals. Am. Nat, 133:654-670.[CrossRef]
Saltin, B., and B. Essen. 1971. Muscle glycogen, lactate, ATP, and CP in intermittent exercise. In B. Pernow and B. Saltin (eds.), Advances in experimental medicine and biology: Muscle metabolism during exercise, pp. 419–424. Plenum Press, New York.
Schmidt-Nielsen, K. 1972. Locomotion: Energy cost of swimming, flying, and running. Science, 177:222-228.
Scholnick, D. A., and T. T. Gleeson. 2000. Activity before exercise influences recovery metabolism in the lizard Dipsosaurus dorsalis. J. Exp. Biol. (In press).
Taylor, C. R., K. Schmidt-Nielsen, and J. L. Raab. 1970. Scaling of energetic cost of running to body size in mammals. Am. J. Physiol, 219:1104-1107.
Tucker, V. A. 1970. Energetic cost of locomotion in animals. Comp. Biochem. Physiol, 34:841-846.[Medline]
Wagner, E. L., and T. T. Gleeson. 1996. Low temperature and exercise recovery in the desert iguana. Physiol. Zool, 69:168-190.
Weinstein, R. B., and R. J. Full. 1992. Intermittent exercise alters endurance in an eight-legged ectotherm. Am. J. Physiol, 262:R852-9.
Weinstein, R. B., and R. J. Full. 1999. Intermittent locomotion increases endurance in a gecko. Phys. Biochem. Physiol, 72:732-739.
Zanconato, S., D. J. Cooper, and Y. Armon. 1991. Oxygen cost and oxygen uptake dynamics and recovery with 1 min of exercise in children and adults. J. Appl. Physiol, 71:993-998.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||