|
|
||||||||
HIGHLIGHTED TOPICS
Role of Exercise in Reducing the Risk of Diabetes and Obesity
Exercise Physiology Laboratory, Department of Integrative Biology, University of California, Berkeley, California
Submitted 10 September 2004 ; accepted in final form 7 December 2004
| ABSTRACT |
|---|
|
|
|---|
O2 peak)] as well as a time-matched resting control trial (Con). Exercise bouts were matched for energy expenditure. Respiratory exchange ratios (RER) during exercise at 65%
O2 peak for both men and women (0.95 ± 0.01 and 0.93 ± 0.02) were significantly higher than 45%
O2 peak (0.89 ± 0.01 and 0.86 ± 0.02) and Con trials (0.86 ± 0.01 and 0.86 ± 0.02, respectively). During recovery, for men RER values were 0.78 ± 0.01 and 0.76 ± 0.01 after 45% and 65% exercise, respectively. For women, values were 0.79 ± 0.01 and 0.78 ± 0.01. These were significantly lower than during both the preexercise resting period and the corresponding no-exercise Con period (0.82 ± 0.01 and 0.83 ± 0.01, mean RER for men and women, respectively). Hence, the contribution of lipid oxidation to energy supply increased significantly during recovery compared with preexercise levels, and it was greater after exercise than during the time-matched, no-exercise Con period. It is concluded that, although carbohydrate is the major fuel source during moderate- to high-intensity exercise, 1) there is substantial postexercise lipid oxidation; and 2) lipid oxidation is the same during postexercise recovery whether the relative power output is 45% or 65% of
O2 peak when energy expenditure of exercise is matched. excess postexercise oxygen consumption; substrate utilization; crossover concept; exertion; energy expenditure
Whereas extensive studies have focused on the relative partitioning of lipids and carbohydrates (CHO) as fuels during exercise (2, 9, 34), less effort has been aimed at uncovering the metabolic events during postexercise recovery. Results show that the relative contribution of lipids decreases as exercise intensity increases [>60% peak oxygen consumption (
O2 peak)] (18). For instance, in healthy young men exercising at 65%
O2 peak for 60 min on a leg cycle ergometer, working muscle respiratory quotient [= carbon dioxide production (
CO2)/oxygen consumption (
O2)] approximated unity (i.e., 1.0), net free fatty acid uptake was small, glycerol release approximated zero, and there was no utilization of intramuscular triglyceride content (3). Still, even though participants in that study were encouraged to increase dietary energy intake to maintain body mass, exercise training resulted in body fat loss and accretion of lean body mass. Thus, notwithstanding the preferential use of CHO during moderate- to high-intensity exercise, it is possible to hypothesize that lipid oxidation predominates during recovery, especially after physical exercise leading to glycogen depletion.
Historically, studies of metabolism during postexercise recovery have focused on excess postexercise
O2 (EPOC, alternatively "oxygen debt") in terms of explaining lactate-glycogen interconversions (20). From a monumental body of literature dating to the beginnings of work physiology, the magnitude of EPOC is known to be related to the intensity and duration of a physical activity bout (6, 7). And, although it was observed that the respiratory exchange ratio (RER) becomes very low during recovery from prolonged exercise (14), energy partitioning after exercise was largely ignored. More recently, investigators have utilized exercise and EPOC measurements to assess the effects of physical activity on daily total energy expenditure (27, 30). Still, fewer have investigated substrate partitioning during postexercise recovery (1, 4). Moreover, limited data exist on energy substrate partitioning in women during the recovery period (13, 32, 34). Although Kiens and Richter (26) have provided data showing intramuscular triglyceride mobilization in the hours after high-intensity exercise, and Horton et al. (21) showed that lipid oxidation is promoted in men, but not women, during recovery from 2 h of mild- to moderate-intensity (40%
O2 peak) exercise, others (28) have concluded that physical activity bouts employing intensities and durations typically used in physical fitness and body weight management programs do not significantly impact lipid oxidation or daily total energy expenditure.
In terms of exercise tolerance and compliance when making physical activity recommendations for fitness or weight loss, issues regarding exercise intensity and duration need to be considered. Previously, Bergman and Brooks (2) showed that trained men can oxidize a slightly higher percentage of lipids than untrained men during mild- to moderate-intensity exercise, but only in the fasted state. Relative differences in substrate partitioning disappeared at high (>60%
O2 peak) compared with low relative exercise intensities or when men were recently fed a CHO-containing meal. As well, in assessing insulin sensitivity in women with Type 2 diabetes after low- (50%) or high- (75%
O2 peak) intensity exercise bouts expending 125 kcal, Braun et al. (5) found low- and high-intensity exercises to yield similar improvements in insulin sensitivity. Hence, on the basis of these previous findings, it was possible to suspect that exercise energy expenditure, rather than exercise intensity or duration per se, was important to consider in assessing the metabolic consequences of an exercise bout.
To evaluate the hypotheses that, in young men and women 1) lipid oxidation predominates during recovery from prolonged submaximal exercises of the type typically employed in fitness and weight loss programs, and 2) RER (
CO2/
O2) values and substrate partitioning during recovery from exercise are similar during moderate- and high-intensity exercises of similar energy expenditures [i.e.,
90 min at 45% vs. 60 min at 65% of
O2 peak], we determined pulmonary
O2 and RER in healthy young men and women during and after moderate- and high-intensity exercise bouts. Importantly, a time-matched control (Con) trial involving no exercise was employed, because time of day and eating may have effects that could mask the subtle effects of exercise intensity and duration on energy substrate partitioning during the EPOC period. Results confirm that moderate-intensity, but more prolonged physical activity is more effective than shorter, but more intense exercise bouts in terms of promoting lipid oxidation during activity, but both forms of exercise are equally effective in promoting lipid oxidation if the postexercise period is considered.
| METHODS |
|---|
|
|
|---|
Screening tests.
After medical history and physical examination, subjects performed two progressive exercise tests to reliably assess
O2 peak. Exercise was performed on a leg-cycle ergometer (Ergometric 818E, Monark, Vansbro, Sweden). A continual, progressive protocol was used to determine
O2 peak; subjects pedaled at a self-selected cadence beginning at 1 kp and increased 0.25 or 0.5 kp every 3 min until voluntary cessation. Respiratory gases were continuously monitored via Vista Mini-CPX O2/CO2 analyzer (VacuMed, Ventura, CA) and recorded every 30 s by a personal computer. In each trial, the system was calibrated twice before rest and exercise using room air and a certified calibration gas tank (16% O2 and 4% CO2). Heart rate was continuously monitored by a Quinton Q4500 electrocardiograph (Quinton, Bothell, WA) and blood pressure by auscultation using a stethoscope and sphygmomanometer. Values obtained during
O2 peak tests were accepted as maximal if heart rate reached within 10% of predicted maximal value and RER exceeded 1.1. Body composition was determined by skinfold measurements (23, 24).
Dietary controls. Participants were instructed to maintain constant diet and physical activity habits throughout the experimental period. Three-day dietary records were completed before each experimental trial to monitor the habitual dietary composition of each subject. During the morning of the first experimental trial, subjects were asked to eat the same breakfast for the remaining two trials. Each trial day, the subjects ate breakfast 2 h before reporting to the laboratory. Analysis of food records was performed using the Nutritionist III program (N-Squared Computing, Salem, OR).
Experimental trials.
The order of the experimental trials was randomly assigned. Each subject was tested under three conditions: exercise at 65% of
O2 peak (65%) for 1 h; 45% of
O2 peak (45%) for 8689 min, to obtain an isocaloric expenditure between the two intensities; and a control (Con) trial, where a 60-min resting period was substituted for exercise. Duration of the 45%
O2 peak test was determined through a prediction of caloric expenditure on the basis of corresponding RER values at 45% obtained during the preliminary
O2 peak assessments. Exercise power output was continually adjusted to maintain the desired relative exercise intensity. Each trial consisted of a preexercise resting period of 30 min, the exercise period of 6089 min, and a 3-h recovery period. During rest and recovery periods, subjects were seated at a table where they were allowed to read quietly. Subjects were transported in a wheelchair to the restroom in between preexercise rest and exercise as well as 120 min into recovery. Furthermore, heart rate and blood pressure were monitored throughout the experimental trial. Subjects were instructed to abstain from strenuous activity 24 h before each experimental trial, and trials were separated by at least 3 days.
In selecting exercise intensities and durations for investigations, we were cognizant of findings in the most recent Institute of Medicine Macronutrient Report (22) in which a minimum of 60 min of exercise most days was recommended for healthy adults to maintain body weights in the healthy body mass index (19.924.9 kg/m2) range. However, because the physical activity estimates were based on results of studies using doubly labeled water technology, the results were nonspecific with regard to the intensity or duration of exercise necessary to oxidize body lipid stores. Hence, we decided on a 60-min activity duration as a baseline condition, but we sought to extend that period 50% while maintaining exercise energy expenditure constant.
Pulmonary gas determinations.
For determination of gas-exchange parameters, expired gas samples were taken at 15-min intervals during preexercise rest, exercise, and recovery. Pulmonary
O2,
CO2, and RER were determined using Vista Mini-CPX O2/CO2 analyzer as during assessments of
O2 peak. During exercise, expired air was constantly monitored for the first 17 min to ensure achievement of the target
O2. Subsequently, breath was sampled for 8 of every 15 min until the end of exercise. At the end of each exercise bout, subjects were instructed to stop cycling and to remain seated on the cycle ergometer for the first 15 min of recovery while expired air was constantly monitored. Afterward, the subject moved to a seat at a nearby table to rest, and again breath was sampled for 8 of every 15 min for time periods 30180 min of recovery. Respiratory values were averaged over the last 5 min of each time point. Throughout the study period when subjects were not wearing the mouthpiece, they were allowed to drink water ad libitum.
Calculations.
Energy derived from CHO and lipid oxidation was calculated (6):
![]() |
![]() |
![]() |
![]() |
![]() |
O2 is in liters per minute.
Calculations during the exercise period of the 45% of
O2 peak trial were made using the last 60 min of measurements to correspond to the 60-min exercise duration of the 65% of
O2 peak and resting control trials.
Statistics.
Data are presented as means ± SE. Within genders, paired t-tests were used to determine significance of differences between trials for physical characteristics. A two-way ANOVA with repeated measures was used to compare means for RER, absolute CHO, and lipid oxidation values, as well as relative contributions to energy expenditure on SPSS 11.5 Graduate Pack (SPSS, Chicago, IL). Significance of differences over time during exercise and postexercise recovery was determined by using one-way ANOVA with repeated measures. Statistical significance of mean differences was set at
= 0.05.
| RESULTS |
|---|
|
|
|---|
O2 peak, habitual physical activity levels, maximal workload, or maximal heart rate during assessment of
O2 peak. However, retrospective analysis showed dietary macronutrient composition differed between genders; women were on a lower fat diet (24.7 ± 1.5%) than men (33.5 ± 2.2% of energy). Nonetheless, study participant dietary energy intake and macronutrient composition did not vary significantly among trials.
|
|
|
|
|
|
| DISCUSSION |
|---|
|
|
|---|
RER data indicate predominance of lipid oxidation during postexercise recovery.
Our data corroborate results of previous studies showing that exercise raises metabolic rate and shifts the pattern of whole body metabolism toward lipid oxidation during the postexercise recovery period (4, 26, 28, 32). Although we did not sample tissue glycogen contents during our experiments, on the basis of extensive previous work, our data (Table 3) can be interpreted to suggest that muscle and liver glycogen stores are mobilized to support a majority of energy expenditure during exercise. It has been shown that the body prioritizes the resynthesis of glycogen during postexercise recovery (7, 19). The increase in relative lipid contribution during recovery from prolonged exercise in our study is consistent with the idea that the body will increase its lipid oxidation for energy, while sparing CHO to minimize further challenge to glycemia and to facilitate subsequent restoration of glucose homeostasis and glycogen repletion.
A few trends were notable throughout the postexercise recovery period. Although
O2 fell rapidly after exercise in both men and women, RER did not drop rapidly until 1530 min into recovery. However, we observed a postexercise nadir in RER, which stabilized
30 min into recovery. Also, there were upward spikes during recovery at 210 min, 30 min after subjects were wheelchair transported to and from the restroom. A similar upward spike is noticeable 150 min in the Con trial. These minor spikes associated with physical activity did not significantly alter the relative contribution of lipid and CHO oxidation in recovery because subsequent values decreased to baseline.
After exercise, RER was substantially lower compared with preexercise levels in both intensities (P < 0.05); similarities in substrate partitioning during exercise recovery may be attributed to the isocaloric nature of the exercise tasks used. In this regard, results of our investigation are consistent with those of Al Mulla et al. (1), who matched exercise energy expenditure values for 40% and 60% of
O2 peak activity bouts for energy cost. Indeed, the present study design is most similar to that of Al Mulla and colleagues; however, they fasted study participants overnight, or possibly longer, making exercise RERs very low (0.79 ± 0.02 and 0.82 ± 0.04 for 40 and 60% exercises, respectively). Likely, Al Mulla and colleagues did not find a shift to lipid oxidation during recovery because the shift had occurred during activity. Additionally, their study lacked a time of day control. As a result, our data can be interpreted to support the assertion that RER decreases after prolonged exercise at low (1, 4) and high intensities (26, 33).
Substrate oxidation rates during exercise and recovery.
During exercise, CHO oxidation was significantly greater at 65% than 45%, which is consistent with results of previous studies (2, 34). At the same time, lipids were oxidized to a lesser extent during exercise at the higher (65%) power output. During the 3-h recovery period after 65% exercise, however, lipid oxidation was promoted such that total lipid oxidation during exercise and recovery for the two exercise tasks was not different (Table 3).
Considering exercise plus recovery, lipid oxidation was higher during postexercise recovery than during Con for men (Fig. 2, A and B, and Fig. 3A), but it did not reach significance for women. Results for men were in agreement with previous investigations (1, 4, 26). Similarly, CHO oxidation was significantly lower during postexercise recovery than in Con for men and women, which is consistent with the idea that the body shifts toward lipid oxidation when glycogen preservation or resynthesis becomes a priority.
Lipid oxidation for men during recovery accounted for
54% of total lipid oxidized throughout the 45% trial and 70% of total lipid oxidation throughout the 65% trial. For women, lipid oxidation in recovery was 46% of total lipid oxidized during the 45% trial and
60% of total lipid oxidized throughout the 65% trial. Although lipid oxidation occurred at generally higher rates after exercise, total caloric expenditure during exercise activity was about double that of recovery in both men and women. With regard to the harder (65%) exercise trial, it is noteworthy that, whereas CHO oxidation was favored during exercise, total lipid oxidation during the 3-h recovery period was greater than during the exercise itself.
Our study produced an unexpected result in that absolute lipid oxidation in the recovery period for women was not different between 45%, 65%, or Con trials (65% and Con, P = 0.07; Fig. 2, C and D, and Fig. 3B). Previously, our laboratory has observed relatively greater lipid oxidation in women than men during exercise (16, 17). However, others have also shown that women do not oxidize more lipid during postexercise recovery than during a time-matched resting control trial (21). Participants were asked to sit quietly during resting and recovery periods; however, nonexercise physical activity such as reading was allowed. Fidgeting may have limited contrasts of substrate oxidation during recovery periods between exercise and Con trials (21). However, retrospective analysis of dietary intake data indicated that women were on a relatively low-fat diet (
25% energy from fat), so it is possible that the unexpected difference in dietary macronutrient composition affected energy substrate partitioning during postexercise recovery. Regrettably, also, we did not control for menstrual cycle phase or oral contraceptive use, but these factors have, at best, small effects on lipid metabolism (11, 37, 38).
Irrespective of the two conditions of exercise intensity and duration we studied, results indicate that isoenergetic exercise tasks result in similar elevations in postexercise lipid oxidation. Such knowledge may be useful to practitioners and participants alike in weight management programs. In terms of exercise compliance, harder exercises result in higher levels of perceived exertion, and some participants may be more compliant with exercise prescriptions if easier, but more prolonged, exercise training tasks are practiced, especially when a training regimen is commenced.
Changes in pattern of substrate utilization over time.
Data in Fig. 3 show that stable metabolic rates were achieved by men and women after 30 min of rest. However, although metabolic rates were stable during the control trial, relative lipid contribution to energy expenditure increased continuously over the 4.5-h period. Although there was a decrease in absolute CHO oxidation and increase in absolute lipid oxidation, energy expenditure remained constant. It is important to note that during rest, just as during exercise, in fasting persons the body will increase its lipid oxidation over a period of time as endogenous CHO energy sources are reduced and glucose homeostasis becomes increasingly dependent on gluconeogenesis (38). The results obtained were those expected on the basis of extensive previous work. However, the control resting trials were necessary for purposes of comparison with results of exercise trials.
Dietary influences.
Application of dietary control is difficult and can influence the results and conclusions. For instance, in our study using two levels of isoenergetic exercise and standard techniques of indirect calorimetry, we observed exercise to result in increased lipid oxidation during postexercise recovery. In contrast, using two levels of isoenergetic exercise (40 and 70%
O2 peak) and room indirect calorimetry, Melanson et al. (28) did not observe an effect of exercise intensity on either 24-h energy expenditure, lipid oxidation, or CHO oxidation. However, in their trials, subjects were fed an extra 500 kcal to cover the energy costs during (400 kcal) and after (100 kcal) activity. In contrast, we did not allow study participants to eat during the observation period. Overall, we view results of Melanson et al. to be similar to ours but with differences in emphasis and interpretation. Physical exercise results in an EPOC, the magnitude of which depends on the intensity and duration of activity (7, 8). By definition, EPOC implies an increase in energy expenditure over rest. Both Melanson et al. and we report that physical activity results in increments in energy expenditure over sedentary, nonexercise control conditions, but we observed a preferential use of CHO during activity and preferential use of lipid during recovery. However, Melanson et al. conducted a different experiment; the additional nutrition provided to participants by Melanson et al. on activity days (500 kcal, 55% CHO, 30% fat, 15% protein) likely would have influenced the subtle effects of exercise on energy substrate partitioning we observed during the EPOC period.
In the effort to quantify magnitude of lipid oxidation after exercise, like Melanson et al. (28), others (4042) have also added energy to the diet by providing a postexercise meal. Such treatment is appropriate if one considers that most of us live in a postprandial state. However, matching or surpassing exercise energy output with dietary input masks the effect of exercise on total energy expenditure and postexercise lipid oxidation. Nonetheless, results of such studies serve as an important reminder that the economy of human locomotion is great. Historically, it was probably an advantage that the energy expenditure of physical activity, including that for postexercise recovery, could be accommodated by subtle changes in behavior and relatively minor increments of dietary intake. In contemporary society, however, the lesson may be that both physical activity and dietary guidelines are needed to prevent and manage body fat accretion.
Limitations.
There are several limitations to our study. First, the experimental trials were performed on moderately active, but not highly trained individuals. Thus the responses observed may not be representative of either very physically fit or sedentary individuals. Similarly, we studied normal-weight individuals, who may respond differently during exercise and recovery than overweight, obese, or Type 2 diabetic individuals. Next, although there is less of a learning curve for riding a leg-cycle ergometer for 6090 min than running, the use of a cycle ergometer with friction braking may have led to variances in substrate oxidation. Furthermore, length of the recovery period led to extraneous movements from subjects, who were reminded to remain still during the periods of quiet rest. Thus we suspect that fidgeting, especially in the Con trial, minimized opportunity to find significant contrasts between energy expenditures or substrate utilization patterns during recovery periods from exercise and time-matched Con conditions (27).
As well, we did not assess or control for the possible effect of menstrual cycle phase (10, 38) or oral contraceptive use (11) on substrate utilization and energy expenditure (12, 37). Additionally, beyond broad categories of CHO and lipid, the precise identities of fuels oxidized could not be determined. For example, we could not differentiate between intramuscular triglycerides, free fatty acids, and lipoproteins. Similarly, we could not discriminate among CHO sources such as oxidized muscle glycogen, blood or muscle lactate, or blood glucose. Importantly, we did not investigate the potential effects of meal size, composition, and timing on lipid oxidation during postexercise recovery. Like physical activity, as well as by definition, nutrition is known to affect metabolism. And, as already indicated, differences in dietary macronutrient composition between men and women may have affected results. Finally, we acknowledge that, given intersubject variability and the small number of study participants, our study may have lacked power to document small (e.g., gender-specific) effects, whereas more robust (e.g., exercise) effects could be identified as different.
Although simple in design, results of the present investigation serve as a reminder that the energy expenditure of exercise includes components of both exercise and recovery. Although the elevation in energy expenditure during recovery is typically significantly less than that during activity, there occurs a shift in the pattern of endogenous substrate utilization during recovery such that it represents a period of predominant and substantial lipid oxidation. And, whereas catabolic processes predominate during activity, anabolic processes predominate during recovery (6). This knowledge is important not only at a basic level but also it offers possibilities for the design of different postactivity interventions to benefit diverse populations ranging from obese individuals to athletes.
Summary.
Although CHO is the major fuel source of exercises requiring moderate and greater exercise intensities (9, 17, 18), 1) there is substantial postexercise lipid oxidation, and 2) lipid oxidation is the same during postexercise recovery whether the relative workload was moderate (45%
O2 peak) or hard (65%
O2 peak) when energy cost of exercise is matched. These results are supportive of the hypothesis that physical exercise is useful in the mobilization and utilization of lipid energy depots in normal weight, healthy young men and women. Some lipid is oxidized during physical activity itself, especially if exercise is of moderate intensity and duration is prolonged (Table 3) (9). However, moderate and greater exercise intensities set into play mechanisms that shift the pattern of substrate utilization away from CHO-derived energy sources toward lipid so that lipid becomes the predominant energy source during recovery (Fig. 2). Hence, on the basis of evidence in this experiment, it is reasonable to conclude that physical activity raises total energy expenditure, resulting in some lipid oxidation during activity, especially if intensity is mild to moderate, and, moreover, that it shifts the pattern of substrate utilization during postexercise recovery when lipids become predominant energy sources.
| GRANTS |
|---|
|
|
|---|
| ACKNOWLEDGMENTS |
|---|
|
|
|---|
| FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
G. C Henderson, J. A Fattor, M. A Horning, N. Faghihnia, M. L Johnson, M. Luke-Zeitoun, and G. A Brooks Glucoregulation is more precise in women than in men during postexercise recovery Am. J. Clinical Nutrition, June 1, 2008; 87(6): 1686 - 1694. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. C. Henderson, J. A. Fattor, M. A. Horning, N. Faghihnia, M. Luke-Zeitoun, and G. A. Brooks Retention of intravenously infused [13C]bicarbonate is transiently increased during recovery from hard exercise J Appl Physiol, November 1, 2007; 103(5): 1604 - 1612. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. C. Henderson, J. A. Fattor, M. A. Horning, N. Faghihnia, M. L. Johnson, T. L. Mau, M. Luke-Zeitoun, and G. A. Brooks Lipolysis and fatty acid metabolism in men and women during the postexercise recovery period J. Physiol., November 1, 2007; 584(3): 963 - 981. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. A. Wallis, A. L. Friedlander, K. A. Jacobs, M. A. Horning, J. A. Fattor, E. E. Wolfel, G. D. Lopaschuk, and G. A. Brooks Substantial working muscle glycerol turnover during two-legged cycle ergometry Am J Physiol Endocrinol Metab, October 1, 2007; 293(4): E950 - E957. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. L. Friedlander, K. A. Jacobs, J. A. Fattor, M. A. Horning, T. A. Hagobian, T. A. Bauer, E. E. Wolfel, and G. A. Brooks Contributions of working muscle to whole body lipid metabolism are altered by exercise intensity and training Am J Physiol Endocrinol Metab, January 1, 2007; 292(1): E107 - E116. [Abstract] [Full Text] [PDF] |
||||
![]() |
Kevin. A. Jacobs, R. M. Krauss, J. A. Fattor, M. A. Horning, A. L. Friedlander, T. A. Bauer, T. A. Hagobian, E. E. Wolfel, and G. A. Brooks Endurance training has little effect on active muscle free fatty acid, lipoprotein cholesterol, or triglyceride net balances Am J Physiol Endocrinol Metab, September 1, 2006; 291(3): E656 - E665. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |