Journal Logo

Basic Sciences: Original Investigations

Biorhythmic influences on functional capacity of human muscle and physiological responses


Author Information
Medicine& Science in Sports & Exercise: September 1998 - Volume 30 - Issue 9 - p 1399-1407
  • Free


Several studies have sought to determine whether the functional capacity of human muscle varies according to the time of day that exertion occurs. Field tests of muscle power (standing broad jump, vertical leap), cycle ergometry, as well as isometric and isokinetic dynamometry have been employed to investigate the effects of circadian rhythms, where testing occurred at equally dispersed time points throughout the 24-h solar day (1,4-6,15,27). Those reports present conflicting results. For example, significant circadian effects were detected in standing broad jump (26,27) but not Margaria Step test performance (26). Isometric grip strength may (1) or may not (3) be subject to circadian influences. Significant circadian variation in back and leg strength as measured with an isometric dynamometer has been reported (6), yet no such fluctuation was identified in peak torque or rate of fatigue of the quadriceps when assessed with isokinetic dynamometry (4). The inconsistency in the literature regarding circadian effects on muscle performance can probably be ascribed to differences in the: 1) dispersion of the time points selected for study within the 24-h day, 2) methodologies employed to determine muscle performance, or 3) selection of performance indices quantified.

The investigation of circadian influences on muscle performance undoubtedly provides valuable basic science information. To our knowledge, however, no data are available concerning time of day influences on muscle performance limited to that segment of the day when exercise is typically engaged, i.e., 0800-2000 h. An investigation identifying the effects of such biorhythmic variation would provide novel information of an applied nature.

It has been clearly established that many physiological variables display circadian, within the 24-h day, fluctuation under resting conditions (22). In addition, several reports have confirmed the presence of circadian rhythms in the physiological responses to exercise (1,4,5,11,24-26). Most of those studies featured aerobic exercise as the stimulus to determine the sensitivity of physiological responses to circadian oscillation. Only a limited number(4,26) have investigated time of day influences on physiological responses to maximal muscle performance. Yet, similar to the assessment of maximal muscle performance itself, no data are currently available regarding the rhythmic variation in physiological responses to maximal muscle exertion during the portion of the 24-h day when people generally exercise. Thus, the purpose of the present study was twofold: 1) to assess the degree of biorhythmicity of muscle function at selected time points between 0800 h and 2000 h, and 2) to determine whether concomitant physiological responses to the stimulus of peak muscle exertion at those time points displayed significant biorhythmic variation.


Subjects. Ten healthy male university students (21.1 ± 0.6 yr, 181.7 ± 2.0 cm, 81.6 ± 3.3 kg; mean ± SE) volunteered to participate in the research project. The subjects were active but had not engaged in lower body resistance training for at least 6 months before the onset of the study. With the exception of the laboratory tests, the subjects performed no lower body resistance training for the duration of the study. After receiving a verbal description of the study, its potential risks, and the experimental procedures to be employed, the subjects provided informed consent. Each subject also completed a medical history form to ensure that no contraindications to participation existed. All experimental procedures were approved by the Committee for the Protection of Human Subjects at The College of William & Mary.

Experimental design. Subjects were initially provided with two familiarization trials 2-3 wk before the study began. During the actual testing, each subject completed a maximal effort resistance exercise protocol of isokinetic knee extensions and flexions at 0800, 1200, 1600, and 2000 h in a randomized fashion. At least 96 h separated any two consecutive data collection sessions. Exercise was not permitted for a minimum of 24 h before testing, and only water was consumed for 8-12 h before data collection. For each subject, all testing was completed within a 4-wk period, and subjects reported normal sleeping habits during that period. For every test session, subjects arrived in the laboratory 30 min before the scheduled exercise protocol. During the first session, the subject's body mass, height, and age were recorded. With each test the subject inserted a rectal temperature probe ∼150 mm beyond the external sphincter(30), and a 20-gauge Teflon catheter with a male adapter was placed in an antecubital vein and kept patent with an isotonic saline solution. A sphygmomanometer was fitted around the subject's arm, and a heart rate telemeter was secured around the subject's chest. Subsequently, the subject sat quietly in a chair for 20 min; at the end of this equilibration period, the first blood sample was drawn, and other preexercise physiological data were collected. The subject then performed a 5-min warm-up on an electrically braked cycle ergometer (Excalibur, Lode, Groningen, Holland) at 50 W.

Exercise testing. The exercise protocol involved performing alternating concentric muscle actions of the knee extensors and flexors on an isokinetic dynamometer (model 900-350, Biodex Co., Shirley, NY) using the right and then the left leg. A 7-min interval separated testing of the right and left legs. For each leg, five repetition sets were performed at 1.05, 1.57, and 2.09 rad·s−1, with a final set of fifty repetitions at 3.14 rad·s−1. Because time of day was the main independent variable of interest, the sequence of these velocities was maintained for each test session: sets performed by the same leg were dispersed by 3-min intervals. To control for confounding movements, the inactive leg was secured on a chair during testing, the subject's contracting leg and torso were stabilized with Velcro straps, and arms were crossed in front of the chest. The back rest was maintained in a position resulting in hip flexion of 80°, and the knee joint was aligned with the axis of the dynamometer. The limb weight was assessed by the dynamometer in preparation for testing so that performance variables could be corrected for that resistance. At each subject's first test session, he selected a range of motion that was adhered to for each subsequent test session. Verbal encouragement was provided throughout each exercise protocol. Immediately(<15 s) following completion of the exercise regimen, physiological variables were quantified, and a second blood sample was drawn.

Quantitation. Heart rates were measured with a portable telemetry unit (Cardiochamp, Sensor Dynamics Inc., Sacramento, CA). Rectal temperature was monitored with a thermistor connected to a digital thermometer (model 400, VWR Scientific, Bridgeport, NJ). Blood pressure was assessed via a sphygmomanometer (Welch Allyn Tycos, Tycos Instruments Inc., Arden, NC) and a stethoscope (Littmann Select, 3m Health Care, St. Paul, MN). Mean arterial blood pressure was calculated as the diastolic pressure plus 33% of the difference between the systolic and diastolic pressures. This value represents the average pressure driving blood into the tissue over the entire cardiac cycle (35). Blood samples were collected into heparin-treated vacutainers. Aliquots of whole blood were used immediately for hemoglobin and hematocrit analyses. Hematocrit was assayed in triplicate using microcapillary tubes after centrifugation at 4000 g for 5 min, while hemoglobin values were determined via the cyanmethemoglobin method. Exercise-induced changes in plasma volume were calculated from hematocrit and hemoglobin values according to Dill and Costill (9). The remaining whole blood was then centrifuged at 3000 g for 15 min at 5°C. The resultant plasma fraction was stored at −75°C until analyses for lactate and hormones were conducted.

Plasma lactate was measured in duplicate with an automated analyzer (YSI model 23 L, Yellow Springs Instruments, Yellow Springs, OH). Plasma testosterone and cortisol levels were determined in duplicate with solid phase 125I radioimmunoassays (ICN Biomedicals, Inc., Costa Mesa, CA). All samples were quantified in a single run; intra-assay variation was <2%, and assay sensitivities were 0.15 μg·dL−1 and 0.15 ng·mL−1 for cortisol and testosterone, respectively.

Muscle performance variables, i.e., peak torque, total work, maximal work for a single repetition, and average power, were calculated by software (Biodex Advantage) accompanying the dynamometer. The same software was used to determine the work fatigue index (% difference between first and last third of repetitions) during the 50-repetition sets. The fatigue index for peak torque of knee extensors (34) during the 50-repetition set of the right leg was calculated by conventional methods.

Statistical analysis. Standard descriptive statistics (mean ± SE) were conducted on physiological and muscle performance variables. To assess chronobiological variation in muscle performance, repeated measures, one-way ANOVAs were conducted for each parameter of each set across the four time points (0800, 1200, 1600, and 2000 h) selected for study. For physiological data, similar analyses of variance were completed on preexercise data and those taken immediately following the final set of the exercise protocol. When ANOVAs revealed significant F-ratios, Fisher PLSD post-hoc analyses were utilized to identify pairwise differences. Dependent t-tests(pre- to postexercise) were performed to determine the effects of maximal exertion on the physiological parameters of interest. Statistical significance was set atP ≤ 0.05.


Physiological variables. Under preexercise conditions, a significant biorhythmicity in rectal temperature was detected. From 0800 h to 2000 h, a gradual increase in temperature was observed, leading to a temperature difference of ∼0.5°C from morning to evening. At each of the four time intervals studied, the exercise protocol elicited a significant elevation in rectal temperature. However, these responses (pre- to postexercise differences) to maximal muscular work were not subject to biorhythmic regulation, as affirmed by similar increases in rectal temperature throughout the portion of the day examined. Consequently, postexercise rectal temperature exhibited significant time of day variation that was identical to that seen before exercise (Fig. 1).

Figure 1
Figure 1:
Pre- and postexercise rectal temperature. Units of measurement are degrees Celsius. Values are means ± SE, N = 10. * indicates significant difference (P ≤ 0.05) from preexercise rectal temperature at 1600 h and 2000 h. ** indicates significant difference (P ≤ 0.05) from preexercise rectal temperature at 2000 h. # indicates significant difference (P ≤ 0.05) from postexercise rectal temperature at 1600 h and 2000 h. ## indicates significant difference (P ≤ 0.05) from postexercise rectal temperature at 2000 h.

Unlike temperature, heart rate preceding exercise was similar among the time intervals studied. Although the exercise protocol caused significant increases in heart rate at 0800, 1200, 1600, and 2000 h, no significant biorhythmicity in this response (pre- to postexercise difference) was identified. Likewise, heart rates immediately following exercise demonstrated no significant differences among the time points studied (Fig. 2).

Figure 2
Figure 2:
Pre- and postexercise heart rate. Units of measurement are beats per minute. Values are means ± SE, N= 10.

The other cardiovascular variable examined, blood pressure, also was impervious to biorhythmic influences under resting conditions. The stimulus of resistance exercise resulted in significant increments in mean arterial blood pressure at each time point studied. There was evidence for biorhythmic oscillation in blood pressure responses to exercise, but these time of day differences did not achieve statistical significance (P = 0.09). However, when postexercise mean arterial blood pressures were analyzed, a significant biorhythmicity was detected where the lowest values were recorded at 0800 h (Fig. 3).

Figure 3
Figure 3:
Pre- and postexercise mean arterial blood pressure. Units of measurement are mm Hg. Values are means ± SE, N = 10.*indicates significant difference (P ≤ 0.05) from postexercise mean arterial blood pressure at 1200 h and 2000 h.

Plasma lactate values before, and immediately following the exercise protocol failed to exhibit significant oscillation over the segment of the day studied. Hence, although exercise evoked significant elevations in plasma lactate at each of the four time points observed, these responses did not demonstrate significant biorhythmicity (Fig. 4).

Figure 4
Figure 4:
Exercise induced response (pre- to postexercise differences) of plasma lactate concentrations. Units of measurement are mmol·L−1. Values are means ± SE, N = 10.

No significant biorhythmic effects on exercise-induced plasma volume shifts were noted. Consequently, hormone values were not corrected for plasma volume changes. Before exercise, plasma cortisol values were highest at 0800 h and lowest at 2000 h, whereas testosterone demonstrated a significant nadir at 2000 h. Similar to preexercise conditions, immediate postexercise cortisol concentrations varied significantly so that they were greatest at 0800 h and lowest at 2000 h. As noted at rest, postexercise plasma testosterone displayed a significant decline at 2000 h.

The fact that pre- and postexercise circulating cortisol and testosterone concentrations demonstrated similar patterns across the time points investigated are probably related to results indicating that, in general, exercise did not elicit significant alterations in hormonal values. Indeed, the resistance exercise protocol employed evoked a significant increase in plasma cortisol at 1200 h only, whereas an exercise induced elevation in testosterone was observed exclusively at 0800 h. Accordingly, hormonal responses to exercise (pre- to postexercise differences) failed to display significant sensitivity to biorhythmic effects. Plasma cortisol and testosterone data can be found in Table 1.

Plasma cortisol and testosterone concentrations before, and immediately following resistance exercise.

Plasma testosterone to cortisol ratios, both pre- and postexercise, were determined at each of the time points of interest. Significant time of day oscillation was established in this putative indicator of anabolic endocrine status (Fig. 5). Both before and after resistance exercise, the plasma testosterone to cortisol ratio was lowest at 0800 h, and greatest at 2000 h. Although exercise-induced responses (pre- to postexercise differences) of that ratio were resistant to biorhythmic influences, it was determined that only at 0800 h did resistance exercise alter (increase) testosterone/cortisol.

Figure 5
Figure 5:
Pre- and postexercise testosterone (μmol·L−1) to cortisol (nmol·L−1) ratios. Units of measurement are arbitrary. Values are means ± SE, N = 10. * indicates significant difference(P ≤ 0.05) from preexercise values at 1200, and 2000 h. # indicates significant difference (P ≤ 0.05) from postexercise values at all other time points. ## indicates significant difference (P ≤ 0.05) from postexercise value at 2000 h.

Performance variables. Statistical analysis of the data indicates that certain aspects of muscle performance are significantly affected by biorhythmic influences. For example, peak torque, which has been suggested to be the most accurate and reproducible indicator of isokinetic muscle function (16), was found to be greatest at 2000 h for right and left knee extensions, but only at 3.14 rad·s−1. No such significant variation in peak torque was identified for right or left knee flexors at any velocity of movement (Table 2).

Peak torque expressed during maximal concentric knee extensor and flexor muscle actions.

Total work performed per set also demonstrated significant biorhythmic effects, but similar to peak torque, only at greater speeds of movement (Table 3). Indeed, at 3.14 rad·s−1, total work performance of the right and left extensors was significantly lower at 0800 h than at all other time points measured. No similar sensitivity to biorhythmic regulation was detected in the knee flexors at 3.14 rad·s−1. However, the flexor muscles of the left knee were found to complete significantly less work at 0800 h when exercising at 2.09 rad·s−1. At no other speed of movement was the total work of the quadriceps or hamstrings significantly different among the times of day studied.

Total work performed during maximal concentric knee extensor and flexor actions.

With respect to the maximal amount of work performed during a single repetition, it was again determined that significant biorhythmic variation was manifested solely at higher velocities of movement. Specifically, the maximum repetition of work done at 3.14 rad·s−1 by the left knee extensors was lower at 0800 h compared with 1200, 1600, and 2000 h, and a similar but nonsignificant (P= 0.06) pattern was displayed by the left knee flexors at the same movement velocity(Table 4).

Maximal work expressed for a single repetition during maximal concentric knee extensor and flexor muscle actions.

The average power of muscle shortening movements displayed significant biorhythmicity that mimicked that observed in peak torque. Only at the highest speed of movement were significant differences detected, and the extensors, but not the flexors, of the right and left knee demonstrated power that was greatest at 2000 h (Table 5).

Average power expressed during maximal concentric knee extensor and flexor muscle actions.

Contrary to all other muscle performance parameters quantified, fatigability of the knee extensors and flexors was not found to significantly fluctuate according to time of day. In the present study, two different muscle fatigue indices were employed. Work fatigue was calculated as the percent difference in total work done during the first and last third of repetitions comprising a 50-repetition set at 3.14 rad·s−1. The second fatigue test was first described by Thorstensson and Karlsson (34), and evaluates the decrement in the averaged peak torque of knee extensions from the best 3 of the first 6 repetitions to the final 3 of 50 repetitions performed at 3.14 rad·s−1. Fatigue measured both in work output (Table 6) and peak torque (Table 7) were resistant to biorhythmic effects. Indeed, both measures of fatigue yielded very similar results.

Work fatigue during 50-repetition sets of maximal concentric knee extensor and flexor muscle actions.
Peak torque fatigue during 50-repetition sets of maximal concentric knee extensor muscle actions of the right leg.


Although several studies have previously investigated the circadian (within the 24-h day) regulation of muscle performance, they have rarely monitored physiological responses to maximal muscular efforts (4,26). To our knowledge, no reports are available concerning fluctuation in maximal muscle exertion, and physiological responses to that stimulus, during the portion of the day when exercise is commonly performed. In the present study, several muscle performance variables (peak torque, total work, maximum work/single repetition, average power, and fatigue) were compared at 0800, 1200, 1600, and 2000 h. In addition, pre- and postexercise data were collected on the physiological variables of heart rate, blood pressure, rectal temperature, as well as plasma lactate. Plasma anabolic (testosterone) and catabolic (cortisol) hormonal data were also recorded. Heart rate and blood pressure were monitored to assess the strain that resistance exercise imparts on the cardiovascular system and whether the responses of this system vary according to the time of day that such strain occurs. Rectal temperature was studied due to its relationship with muscle performance (30), and because at rest, it demonstrates biorhythmic variation (5,11). Plasma testosterone and cortisol concentrations were measured to provide insight into the time of day at which endocrine status may be most conducive for anabolic responses to resistance exercise to occur.

The heart rate data gathered in this investigation closely emulate those in our study reporting biorhythmic effects on exhaustive aerobic performance and its physiological responses (8). In effect, between 0800 and 2000 h, neither pre- nor postexercise heart rates demonstrated significant fluctuation. This is true despite the fact that at each time interval studied, exercise caused significant elevations in heart rate. Several reports have confirmed significant biorhythmicity in heart rate before (5,11,25), during (28), and immediately following maximal exertion (25). However, such rhythmicity was established only by heart rate nadirs detected during the very early morning hours, i.e., 0200-0600 h, when exercise or athletic events do not typically occur.

In contrast to heart rate, the other cardiovascular variable of interest, blood pressure, did display significant biorhythmic variation upon completion of maximal resistance exercise. Specifically, the response of mean arterial blood pressure to resistance exercise was lowest at 0800 h. This finding can be attributed to a concomitant significant time of day effect on postexercise systolic blood pressure and a similar tendency (P = 0.06) in diastolic blood pressure response. These data are also in agreement with our previous study where it was determined that blood pressure response to exhaustive aerobic exercise was significantly lower at 0800 h (8). In that study, a similar nadir in plasma norepinephrine was determined, and across the four time points selected, a correlation of 0.78 was found between mean arterial blood pressure and plasma norepinephrine values. In the present study, plasma norepinephrine was not directly measured. However, because resistance exercise is known to elevate circulating catecholamine levels (19) and blood pressure responses recorded here were like those of our earlier study, it is possible that a similar time of day variation in circulating norepinephrine response to resistance exercise occurred and contributed to the present blood pressure data.

Contrary to plasma lactate responses to exhaustive aerobic exercise reported in our previous work (8), no significant biorhythmic sensitivity of lactate increases consequent to maximal resistance exercise was detected. However, similar to our findings with cycle ergometry, the current data indicate that lactate response to exercise is progressively enhanced throughout the segment of the day when exercise is typically conducted. The absence of significant rhythmicity is presumably related to the fact that plasma lactate responses to resistance exercise were tempered compared to those seen with maximal cycle ergometry. Thus, it appears that only with greater exercise-induced elevations in plasma lactate than those documented here can statistically significant time of day variation be established.

The present study does demonstrate, however, that resistance exercise elicits statistically significant elevations in plasma lactate and that this response occurs at each of the time points chosen for testing. This is consistent with previous studies which have documented that the stimulus of resistance exercise is sufficiently potent to cause increases in plasma lactate (17,32). However, postexercise lactate values recorded here were not as great as those observed following exhaustive aerobic exercise (8,20,31). This may be due to the intermittent, rather than continuous, nature of resistance exercise. Alternatively, instead of the rest periods interspersed within a resistance exercise regimen, the difference in lactate response between that stimulus and exhaustive continuous aerobic efforts may be related to the total volume of work performed. This assertion is supported by the correlational data of the present study between postexercise plasma lactate and muscle performance variables. Indeed, these analyses revealed that of peak torque, average power, maximal work performed in a single repetition, and total work done per set during the 50 repetition sets, lactate values were most strongly related to total work performed (r = 0.57, P < 0.05). This suggests that with a resistance exercise protocol involving more muscle groups and sets of muscle actions, postexercise plasma lactate values would be no less than those found following maximal aerobic exercise. In fact, there is evidence that following high-volume resistance exercise, lactate concentrations are similar to those noted after an exhaustive aerobic challenge (19,32).

Our finding that body temperature exhibits significant biorhythmic fluctuation under resting conditions is in agreement with the literature (5,11,25). The fact that exercise, irrespective of the time of day involved, results in significant increments in rectal temperature is also consistent with previously reported data (23,30). It is interesting to note that although preexercise rectal temperature progressively increased among the four time points selected for study, exercise-induced elevations in temperature were constant throughout those time intervals. This phenomenon had been noted before with aerobic exercise (8,25). This lack of biorhythmic fluctuation in body temperature response (difference between pre- and postexercise values) to exercise resulted in a significant rhythm in postexercise temperature that paralleled that observed before exercise. This, too, is consistent with previous reports (4,25,28).

The significant biorhythmicity of rectal temperature may have important implications for our muscle performance data. Our results indicate the presence of significant biorhythms in most isokinetic muscle performance variables measured. However, it appears that these time of day influences are manifested only at higher speeds of movement. With a single exception, all instances of significant biorhythmic variation occurred during movement velocities of 3.14 rad·s−1; the other positive finding was detected at 2.09 rad·s−1. These data suggest that biorhythmic variation in concentric muscle actions of the lower body is speed specific. Such specific rhythmicity was noted in peak torque, average power per set, work performed per set, and maximal work during a single repetition.

Because the set of muscle actions completed at 3.14 rad·s−1 consisted of 50, rather than 5 repetitions as at all other velocities tested, it was considered that this difference may have been responsible for our speed specific findings. Indeed, it was revealed that in 16 of a total of 80 sets completed, peak torque during knee extensions was expressed beyond the 5th, but no later than the 14th, repetition performed. However, it is unlikely that this contributed to our significant time of day effect in knee extensor torque, because these 16 cases were equally distributed among the four data collection time points.

Our data on peak torque are in agreement with previous reports. That is, in vivo peak torque during concentric muscle actions is greatest at slower speeds of movement and gradually diminishes as the velocity of movement increases. This is true of knee extensors (10,13,33), knee flexors(36), and plantar flexors (36). It has been postulated that the speed specificity of peak torque is related to motor unit recruitment patterns such that at higher speeds of movement only fast-twitch motor units substantially contribute to muscle force production (16,33,34). Although slow-twitch motor units are recruited during high velocity movements, they are unable to affect force generation during these conditions due to their contractile characteristics. This phenomenon, coupled with the significant biorhythmicity of temperature, may underlie our data concerning time of day variation in muscle performance.

It has been determined that within physiological limits, muscle function improves as temperature increases (21,29). It has also been demonstrated that during exercise, muscle and rectal temperature respond in a parallel fashion, clearly establishing a relationship between these two variables (30). Moreover, the data presented here and elsewhere have confirmed that rectal, and thus muscle temperature, is lowest in the morning hours and progressively increases throughout the afternoon and evening hours (8,25). Close inspection of our data reveals that at virtually every speed of movement and parameter of performance quantified, muscle function is impaired in the morning compared with the evening. Yet, our only significant time of day differences in muscle performance occurred at the fastest velocities of movement employed. Like muscle, the function of the nervous system is temperature dependent, resulting in a positive, linear relationship between conduction velocity and temperature(2,29). Accordingly, significant biorhythmicities of nerve conduction velocity, sensitivity, and neuromuscular efficiency have been identified that mimic the time of day fluctuation of core temperature (12).

Because muscle performance at greater velocities of movement depends primarily on fast-twitch motor units and the function of these neuromuscular units are affected by temperature, it is reasonable that our results indicate significant time of day variation in muscle performance only at the highest velocities of movement measured. Further support for this assertion is provided by our data which show that across the four time points investigated, a correlation of 0.88 exists between peak torque generated at 3.14 rad·s−1 and rectal temperature.

In contrast to every other performance variable assessed, muscle fatigue, which was determined at 3.14 rad·s−1, was resistant to biorhythmic influences. At first glance, these data seem to contradict our explanation for the observation of biorhythmicity at fast velocities of movement. However, the fatigue tests employed consisted of 50 repetitions with a duration of∼1 min. Because these fatigue tests measured the relative decrement in muscle performance during a 50-repetition set, the anaerobic metabolic capacity of the involved musculature would be of greater importance than its innervation characteristics. An earlier investigation directly assessing time of day effects on anaerobic capacity utilizing Wingate test performance failed to identify significant fluctuation (26). This constancy in the anaerobic capacity of leg muscles supports our results concerning muscle fatigue during a 50-repetition test, without conflicting with our explanation of speed specific biorhythmicity in all other muscle performance variables of interest.

Circulating concentrations of the androgen testosterone and the glucocorticoid cortisol were measured to determine biorhythmic variation in the endocrine regulation of the endogenous anabolic/catabolic milieu and, in turn, to assess the responsiveness of these hormones to resistance exercise at different times of the day. Our results affirm previous work (7) demonstrating that both of these steroid hormones tend to peak in the morning hours and diminish throughout the day leading to the expression of their lowest values in the evening. In the present study, plasma testosterone and cortisol concentrations were highest at 0800 h and lowest at 2000 h, both before and immediately following resistance exercise.

If plasma testosterone by itself is considered the primary endocrine regulator of muscle mass, it appears that the early morning (0800 h) is the optimal time for resistance exercise. Not only are pre- and postexercise testosterone concentrations greatest among the time points studied, but only at 0800 h does exercise evoke significant increases in the blood-borne levels of this steroid. Conversely, if the plasma testosterone to cortisol ratio, suggested to reflect the overall anabolic status of men (14,18), does in fact more accurately indicate the potential for protein accretion in muscle than testosterone alone, then our pre- and postexercise data imply that among the time intervals examined, the early evening (2000 h) is most conducive to resistance exercise. In contrast, however, the responsiveness of plasma testosterone/cortisol to the stimulus of resistance exercise is greatest at 0800 h, the same time at which significant exercise-induced testosterone elevations occur.

In conclusion, the present study identified significant biorhythmic variation in blood pressure, rectal temperature, plasma testosterone, and cortisol concentrations, but not heart rate or plasma lactate, following a protocol of maximal effort isokinetic resistance exercise of the knee extensors and flexors. With the exception of fatigue, each of the muscle performance variables quantified (peak torque, average power, maximal work done for a single repetition, and total work performed per set) were significantly influenced by chronobiological effects. Yet, significant results were demonstrated only at the fastest velocities of movement utilized. This speed specificity in the time of day variation of muscle performance may be accounted for by a similar biorhythmicity in body temperature, i.e., lowest at 0800 h, greatest at 2000 h. Thus, it appears that in the segment of the day when exercise is typically performed, the functional capacity of muscle is lowest in the early morning and peaks in the early evening hours.


1. Atkinson, G., A. Coldwells, and T. Reilly. A comparison of circadian rhythms in work performance between physically active and inactive subjects. Ergonomics 36:273-281, 1993.
2. Bagley, R. S., S. J. Wheeler, and J. M. Gay. Effects of age on temperature-related variation in motor nerve conduction velocity in healthy chickens. Am. J. Vet. Res. 56:819-821, 1995.
3. Baxter, C., and T. Reilly. Influence of time of day on all-out swimming. Br. J. Sports Med. 17:122-127, 1983.
4.Cabri, J., B. Dewitte, and J. P. Clarys. Circadian variation in blood pressure responses to muscular exercise. Ergonomics 31:1559-1565, 1988.
5. Cohen, C. J. Human circadian rhythms in heart rate response to a maximal exercise stress. Ergonomics 23:591-595, 1980.
6. Coldwells, A., G. Atkinson, and T. Reilly. Sources of variation in back and leg dynamometry. Ergonomics 37:79-86, 1994.
7. Cooke, R. R., J. E. McIntosh, and R. P. McIntosh. Circadian variation in serum free and non-SHBG-bound testosterone in normal men: measurements, and simulation using a mass action model. Clin. Endocrinol. 39:163-171, 1993.
8. Deschenes, M. R., J. V. Sharma, K. T. Brittingham, D. J. Casa, L. E. Armstrong, and C. M. Maresh. Chronobiological effects on exercise performance and selected physiological responses. Eur. J. Appl. Physiol. 77:249-256, 1998.
9. Dill, D. B., and D. L. Costill. Calculation of percentage changes in volumes of blood, plasma, and red cells in dehydration.J. Appl. Physiol. 37:247-248, 1974.
10. Dudley, G. A., R. T. Harris, M. R. Duvoisin, B. M. Hather, and P. Buchanan. Effect of voluntary vs. artificial activation on the relationship of muscle torque to speed. J. Appl. Physiol. 69:2215-2221, 1990.
11. Faria, I. E., and B. J. Drummond. Circadian changes in resting heart rate and body temperature, maximal oxygen consumption and perceived exertion. Ergonomics 25:381-386, 1982.
12. Freivaldis, A. Investigation of circadian rhythms on select psychomotor and neurological functions. Doctoral Dissertation. University of Michigan, Ann. Arbor, MI, 1979.
13. Froese, E. A., and M. E. Houston. Torque-velocity characteristics and muscle fiber type in human vastus lateralis. J. Appl. Physiol. 59:309-314, 1985.
14. Hakkinen, K., A. Pakarinen, M. Alen, and P. V. Komi. Serum hormones during prolonged training of neuromuscular performance. Eur. J. Appl. Physiol. 53:287-293, 1985.
15. Ilmarinen, J., R. Ilmarinen, O. Korhonen, and M. Nurminen. Circadian variation of physiological functions related to physical work capacity. Scand. J. Work Environ. Health 6:112-122, 1980.
16. Kannus, P. Isokinetic evaluation of muscular performance: implications for muscle testing and rehabilitation. Int. J. Sports Med. 15:S11-S18, 1994.
17. Kraemer, R. R., J. L. Kilgore, G. R. Kraemer, and V. D. Castracane. Growth hormone, IGF-I, and testosterone responses to resistive exercise. Med. Sci. Sports Exerc. 24:1346-1352, 1992.
18. Kraemer, W. J. Endocrine responses to resistance exercise. Med. Sci. Sports Exerc. 20:S152-S157, 1988.
19.Kraemer, W. J., B. J. Noble, M. J. Clark, and B. W. Culver. Physiologic responses to heavy-resistance exercise with very short rest periods. Int. J. Sports Med. 8:247-252, 1987.
20. Lehmann, M., and J. Keul. Free plasma catecholamines, heart rates, lactate levels, and oxygen uptake in competition weight lifters, cyclists, and untrained control subjects. Int. J. Sports Med. 7:18-21, 1986.
21. Lutz, G. J., and L. C. Rome. Muscle function during jumping in frogs. I. Sarcomere length change, EMG pattern, and jumping performance. Am. J. Physiol. 271:C563-C570, 1996.
22. Mills, J. N. Human circadian rhythms. Physiol. Rev. 46:128-163, 1966.
23. Nadel, E. R., E. Cafarelli, M. F. Roberts, and C. B. Wenger. Circulatory regulation during exercise in different ambient temperatures. J. Appl. Physiol. 46:430-437, 1979.
24. Reilly, T., and C. Baxter. Influence of time of day on reactions to cycling at a fixed intensity. Br. J. Sports Med. 17:128-130, 1983.
25. Reilly, T., and G. A. Brooks. Selective persistence of circadian rhythms in physiological responses to exercise. Chronobiol. Int. 7:59-67, 1990.
26. Reilly, T., and A. Down. Investigation of circadian rhythms in anaerobic power and capacity of the legs. J. Sports Med. Physiol. Fitness 32:343-347, 1992.
27. Reilly, T., and A. Down. Circadian variation in the standing broad jump. Percept. Mot. Skills 62:830, 1986.
28. Reilly, T., G. Robinson, and D. S. Minors. Some circulatory responses to exercise at different times of day.Med. Sci. Sports Exerc. 16:477-482, 1984.
29.Rutkove, S. B., M. J. Kothari, and J. M. Shefner. Nerve, muscle, and neuromuscular junction electrophysiology at high temperature. Muscle Nerve 20:431-436, 1997.
30. Saltin, B., and L. Hermansen. Esophageal, rectal, and muscle temperature during exercise. J. Appl. Physiol. 21:1757-1762, 1966.
31. Schwarz, L., and Kindermann, W. β-endorphin, adrenocorticotropic hormone, cortisol and catecholamines during aerobic and anaerobic exercise. Eur. J. Appl. Physiol. 61:165-171, 1990.
32.Tesch, P. A., E. B. Colliander, and P. Kaiser. Muscle metabolism during intense, heavy-resistance exercise. Eur. J. Appl. Physiol. 5:362-366, 1986.
33. Thorstensson, A., G. Grimby, and J. Karlsson. Force-velocity relations and fiber composition in human knee extensor muscle. J. Appl. Physiol. 40:12-16, 1976.
34. Thorstensson, A., and J. Karlsson. Fatiguability and fibre composition of human skeletal muscle. Acta Physiol. Scand. 98:318-322, 1976.
35. Vander, A. J., J. H. Sherman, and D. S. Luciano. Human Physiology, 6th ed. New York, NY: McGraw-Hill, 1994, p. 431.
36. Wickiewicz, T. L., R. R. Roy, P. L. Powell, J. J. Perrine, and V. R. Edgerton. Muscle architecture and force-velocity relationships in humans. J. Appl. Physiol. Respirat. Environ. Exerc. Physiol. 57:435-443, 1984.


©1998The American College of Sports Medicine