EFFECT OF PEDAL RATE ON DIURNAL VARIATIONS

IN CARDIORESPIRATORY VARIABLES

N. Bessot,1 S. Moussay,1 A. Gauthier,1 J. Larue,1 B. Sesboue,2

and D. Davenne1

1Centre de Recherches en Activites Physiques et Sportives (CRAPS UPRES EA2131),UFR Sciences et Techniques des Activites Physiques et Sportives; Universite de CaenBasse-Normandie, Caen Cedex, France2Institut Regional de Medecine du Sport, CHU Caen Cote de Nacre, Caen Cedex, France

Recently, it was observed that the freely chosen pedal rate of elite cyclists was signifi-cantly lower at 06:00 than at 18:00 h, and that ankle kinematics during cycling exhi-bits diurnal variation. The modification of the pedaling technique and pedal rateobserved throughout the day could be brought about to limit the effect of diurnalvariation on physiological variables. Imposing a pedal rate should limit the subject’spossibility of adaptation and clarify the influence of time of day on physiologicalvariables. The purpose of this study was to determine whether diurnal variationin cardiorespiratory variables depends on pedal rate. Ten male cyclists performeda submaximal 15 min exercise on a cycle ergometer (50% Wmax). Five test sessionswere performed at 06:00, 10:00, 14:00, 18:00, and 22:00 h. The exercise bout wasdivided into three equivalent 5 min periods during which different pedal rates wereimposed (70 rev . min21, 90 rev . min21 and 120 rev . min21). No significant diurnalvariation was observed in heart rate and oxygen consumption, whatever the pedalrate. A significant diurnal variation was observed in minute ventilation ( p ¼ 0.01).In addition, the amplitude of the diurnal variation in minute ventilation dependedon pedal rate: the higher the pedal rate, the greater the amplitude of its diurnalvariation ( p ¼ 0.03). The increase of minute ventilation throughout the day ismainly due to variation in breath frequency ( p ¼ 0.01)—the diurnal variation oftidal volume (all pedal rate conditions taken together) being non-significant—butthe effect of pedal rate � time of day interaction on minute ventilation specific tothe higher pedal rate conditions ( p ¼ 0.03) can only be explained by the increaseof tidal volume throughout the day. Even though an influence of pedal rate ondiurnal rhythms in overall physiological variables was not also evidenced, high

Submitted November 11, 2005, Returned for revision January 23, 2006, Accepted February 13,2006

Address correspondence to S. Moussay, UFR STAPS Caen, Universite de Caen Basse-Normandie, 2 Boulevard du Marechal Juin, 14032 Caen Cedex, France. E-mail: nicoaure@club-internet.fr

Chronobiology International, 23(4): 877–887, (2006)Copyright # Informa HealthcareISSN 0742-0528 print/1525-6073 onlineDOI: 10.1080/07420520600827178

877

pedal rate should have been imposed when diurnal variations of physiological vari-ables in cycling were studied.

Keywords Pedal rate, Time of day, Diurnal variation, Physiological variables, Cycling,Submaximal exercise

INTRODUCTION

Many physiological variables exhibit daily oscillations that reflect thedirect influence of an endogenous pacemaker clock located in the supra-chiasmatic nuclei (Ralph et al., 1990). Such rhythms have been compre-hensively described for heart rate (HR) (Reilly and Brooks, 1990;Taillard et al., 1990), core temperature (Gauthier et al., 1997; Racinaiset al., 2004; Reilly et al., 1997; Waterhouse et al., 2005), minute ventilation(VE) (Spengler et al., 2000), oxygen uptake (VO2) (Reilly and Brooks,1990), and testosterone and cortisol hormonal response (Bird and Tarpen-ning, 2004), as well as muscular performance (Guette et al., 2005; Pearsonet al., 2005; Racinais et al., 2005), and other physiological functions.

Studies focusing on diurnal variations in HR, VE, and VO2 in submaxi-mal exercise have reported equivocal results (Deschenes et al., 1998;Giacomoni et al., 1999; Hill et al., 1988; Reilly and Brooks, 1982; Reillyet al., 1984). It seems difficult to observe diurnal variation in cardiorespira-tory variables because of its low amplitude (less than 10%). Some test con-ditions, such as power output or pedal rate, could increase the amplitudeof variation to help detect diurnal differences.

Power output during exercise is usually viewed as being the mainfactor in diurnal variations under submaximal exercise conditions (Reillyand Brooks, 1982). Diurnal rhythms observed in physiological measure-ments decline as exercise intensity increases (Cable and Reilly, 1987; Hillet al., 1989). Taking the aerobic contribution into consideration, therange of the circadian rhythm in VO2 usually observed at rest (Cableand Reilly, 1987; Reilly and Brooks, 1990) clearly declined as exerciseintensity increases (Cable and Reilly, 1987).

It may be hypothesized that diurnal rhythms in physiological variablesare also affected by pedal rate. The influence of pedal rate on HR, VO2,and VE, when exercise is performed under submaximal aerobic con-ditions, has been clearly established (Boning et al., 1984; Chen et al.,1999). An increase in the values of cardiorespiratory variables wasobserved when the pedal rate was increased beyond 70 rev . min21 for agiven power output. This is explained by an increase in the metaboliccost associated with an increase in mechanical muscle work output tomove the legs faster (Francescato et al., 1995) and by a change in pedalingtechnique (Moussay et al., 2003). As for power output, pedal rate influe-nces the metabolic cost of cycling. Therefore, the magnitude of cyclic

N. Bessot et al.878

variations in HR, VE, and VO2 may depend not only on the power outputelicited during exercise but also on the pedal rate. There is no clear pictureas to what influence the pedal rate might have on the diurnal variation ofcardiorespiratory variables. However, the level of the imposed pedal ratemight be assumed to account in part for the diurnal variation. Consideringall of these aspects, the purpose of this study was to determine whether, fora given power output, diurnal variation in cardiorespiratory variablesdepends on pedal rates.

METHODS

Subjects

Ten healthy male cyclists (aged: 21.5 + 1.3 yrs; body mass:70.1 + 7.0 kg; height: 173.3 + 8.1 cm) adhering to a routine of diurnalactivity between 07:00 and 23:00 h alternating with nighttime sleep volun-teered to participate in this experiment. This study was granted ethicalapproval by the ethics committee, CHU Cote de Nacre, Caen, France,and complied with the expectations of the journal for the conduct ofresearch of human biological rhythms (Touitou et al., 2004). All subjectsprovided informed written consent after the study procedures wereexplained in detail. All subjects had at least three years of experience incompetitive cycling and usually trained 8 to 10 h per week. The subjectswere either “moderately morning” (n ¼ 4) or “neither type” (n ¼ 5) onthe basis of their answers to the self-assessment questionnaire of Horneand Ostberg (1976), which determines morningness-eveningness.

Experimental Procedure

Wmax was first assessed by a maximal graded exercise. The individualWmax was later used to set the individual power output for submaximalsteady-state exercise.

Two weeks later, subjects took part in five test sessions conducted at thefollowing times of day: 06:00, 10:00, 14:00, 18:00, and 22:00 h. To avoidthe effect of sleep deprivation, no test session was done at 02:00 h. Eachsubject performed each of the five test sessions, the ordering of whichwas randomized. For each subject, there was only one session per daywith a minimal period of 24 h between consecutive sessions. As proposedby Baxter and Reilly (1983), subjects ate a meal 2 h before the tests at14:00 h and 22:00 h, and they were awoken at 05:00 h for the 06:00 htest. Before this particular test, subjects were allowed to drink a glass ofwater. Also, subjects were asked to adhere as closely as possible to theirusual sleeping habits, with a minimum of 6 h sleep taken on the night

Effect of Pedal Rate on Diurnal Variations 879

preceding each test. Subjects were not engaged in any fatiguing exerciseduring the protocol.

During the conduct of the experimental protocol, the laboratory tem-perature (22.0 + 0.88C) and humidity (28.3 + 7.4%) were kept constant tominimize their effects on muscular function (Bergh and Ekblom, 1979).

In the first part of the protocol, the cyclists performed a maximal con-tinuous ramp test. They rode to warm up on a cycle ergometer (SRM HighPerformance Ergometerw, Julich, Germany) for 3 min at a work-load of50 W. Then the work-rate was increased every 1.5 min in steps of 25 Wuntil exhaustion. The last step completed by the subject was used to deter-mine Wmax.

During the second part of the protocol, each of the five test sessionsstarted with a 15 min rest period during which subjects remainedsupine. Then, each subject performed three 5 min submaximal steadystate exercises at a workload that elicited 50% of his Wmax. For each testsession, subjects were requested to ride a cycle ergometer (SRM High Per-formance Ergometerw, Julich, Germany) equipped with racing pedals andtoe clips. The ergometer was fully adjustable, both vertically and horizon-tally, to reproduce the exact measurements of each subject’s own bicycle.

Subjects were required to pedal while randomly changing the cadenceevery 5 min to 70, 90, or 120 rev . min21. To maintain the imposed exer-cise condition, subjects were continuously provided with instantaneousvisual feedback regarding pedal rate. Subjects were given a 5 min restperiod between each exercise stage.

Data Collection

Submaximal aerobic test sessions were performed in order to detect atime-of-day effect on HR, VO2, VE, tidal volume (TV), and breath fre-quency (BF). During the test session, ventilatory and gas exchange variableswere monitored breath-by-breath by an open-circuit sampling system(Ergocard Medisoftw, Schiller, Belgium). HR was determined continuouslyfrom an electrocardiograph recording (Ergocard Medisoftw, Schiller,Belgium). For each stage, HR, VO2, VE, TV, and BF values were givenby the mean of the last minute of the exercise bout, as it would havetaken at least 2 min to reach a steady state of VO2 and VE at the given inten-sity. The difference between individual maximum and maximum valuesover the five time points was used to explore the impact of pedal rate onthe amplitude of the diurnal variation of each variable.

Statistical Analysis

Diurnal variations in HR, VE, VO2, TV, and BF were identified by atwo-way analysis of variance (ANOVA) with the factors “time of day” and

N. Bessot et al.880

“pedal rate condition” as repeated measures. Where appropriate, a PLSDFischer test was performed as a post-hoc analysis. A one-way ANOVA withthe “pedal rate” factor as the repeated measure was applied to the indivi-dual delta value calculated between test sessions to compare the magnitudeof diurnal variation of a single variable under the different pedal rate con-ditions. Probability ( p) values of less than 0.05 were taken to indicatestatistical significance.

RESULTS

Heart Rate

As shown in Figure 1, HR increased in line with pedal rate increase(F(2, 18) ¼ 331.67; p , 0.001) no matter the test session. HR did notvary significantly among the five time points of study (F(4, 36) ¼ 1.30;p ¼ 0.28). Moreover, the pedal rate effect on HR was unrelated to thetime of day (F(8, 72) ¼ 1.18; p ¼ 0.31).

Oxygen Uptake

As depicted in Figure 2, VO2 mainly depended on pedal rate. Inde-pendent of the time of day (F(2, 18) ¼ 97.18; p , 0.001), the higher thepedal rate, the higher the VO2. VO2 did not differ in a significantmanner according to the time of day when testing was performed(F(4, 36) ¼ 0.77; p ¼ 0.54), and there was no inter-relation between timeof day and pedal rate effects (F(8, 72) ¼ 1.85; p ¼ 0.08).

FIGURE 1 Heart rate related to time of day. Each point represents one pedal rate condition (mean+SD, n ¼ 9: 4 70 rev . min21; W 90 rev . min21 ; A 120 rev . min21).

Effect of Pedal Rate on Diurnal Variations 881

Ventilation

As shown in Figure 3, VE increased with pedal rate (F(2, 18) ¼ 76.21;p , 0.001). Mean values from the five test sessions rangedfrom 48.9 + 12.6 l . min21 at 70 rev . min21 to 70.8 + 12.4 l . min21 at120 rev . min21. The mean value measured at 120 rev . min21 was signifi-cantly higher than at any other frequency ( p , 0.001).

A significant chronobiological effect was found in VE (F(4, 36) ¼ 3.68;p ¼ 0.01; see Figure 3) for conditions of 70 and 120 rev . min21. Thehighest values of VE were recorded at 14:00 or 18:00 h on the basis ofthe pedal rate studied. Lowest values were always observed at 06:00 h.Ventilation showed significant time of day � pedal-rate interaction (F(8,

FIGURE 2 Oxygen uptake related to time of day. Each point represents one pedal rate condition(mean + SD, n ¼ 9: 4 70 rev . min21; W 90 rev . min21 ; A 120 rev . min21).

FIGURE 3 Minute ventilation related to time of day. Each point represents one pedal rate condition(mean + SD, n ¼ 9: 4 70 rev . min21; W 90 rev . min21; A 120 rev . min21).

N. Bessot et al.882

72) ¼ 1.93; p ¼ 0.03). The analysis of variance for repeated measures,done on the individual maximal delta values of data (maximum–minimum) for VE, demonstrated that the magnitude of the diurnal vari-ation increased with pedal rate (F(2, 16) ¼ 7.78; p ¼ 0.04). The magnitudeof the diurnal variation observed in VE at the imposed pedal rate of120 rev . min21 was significantly higher than the magnitude measuredunder the other pedal rate conditions.

Breath Frequency

Pedal rate was directly related to BF. We observed that BF increasedsignificantly under the 120 rev . min21 condition as compared to allother conditions (F(2, 18) ¼ 28.19; p , 0.001). Tidal volume and BFvalues were subjected to ANOVA to discover which of these variables con-tributed significantly to the diurnal variation of VE. The results indicatedsignificant diurnal variations for BF values (F(4, 36) ¼ 3.88; p ¼ 0.01; seeFigure 4). When the magnitude of the diurnal variations were comparedby calculating delta values between the test sessions, no significant differ-ence was found in pedal rate (F(2,18) ¼ 0.42; p ¼ 0.7).

Tidal Volume

TV increased with pedal rate (F(2, 18) ¼ 27.87; p , 0.001). The TVmeasured at 70 rev . min21 was significantly lower than for all otherpedal rate conditions. Maximal values of TV observed at 120 rev . min21

were significantly different from the ones measured under all the otherpedal rate conditions (see Figure 5). No significant time-of-day effect wasobserved (F(4, 36) ¼ 1.1; p ¼ 0.3). However a significant time of day �pedal rate interaction was shown in TV values (F(8, 72) ¼ 1.98;

FIGURE 4 Breath frequency related to time of day. Each point represents one pedal rate condition(mean + SD, n ¼ 9: 4 70 rev . min21; W 90 rev . min21 ; A 120 rev . min21).

Effect of Pedal Rate on Diurnal Variations 883

p ¼ 0.03). At 70 rev . min21, the only significant difference was observedbetween 06:00 and 10:00 h. At 90 rev . min21, the TV values at 06:00 hwere significantly lower than those at 10:00, 18:00, and 22:00 h. Themain significant diurnal variation was observed at 120 rev . min21, withstatistically significantly significant differences detected between the06:00, 10:00, and 14:00 h tests. Maximal values were observed at14:00 h and did not differ from the 18:00 h ones. The values at 18:00 hwere higher than the 06:00 and 10:00 h ones. A decrease in TV wasthen observed between 18:00 and 22:00 h. The delta values observedbetween the test sessions ranged from 4.2% at 70 rev . min21 to 8.4% at120 rev . min21.

DISCUSSION

The major aim of this study was to determine whether, for a givenpower output, diurnal variations in cardiorespiratory variables dependon pedal rate. Establishment of the influence of pedal rate on HR, VO2,and VE when exercise was performed under submaximal aerobic con-ditions confirmed earlier reports (Boning et al., 1984; Chen et al., 1999).An increase in the values for cardiorespiratory variables was observedwhen the pedal rate was increased for a given power output.

No significant time-of-day effect was detected in HR, regardless of thepedal rate studied. Although not statistically significant, mean values rosefrom 06:00 to 14:00 h and decreased from 14:00 to 22:00 h for each pedalrate condition studied. The finding of an increase in HR throughout theday is consistent with the results of previous reports (Reilly et al., 1984).

FIGURE 5 Tidal volume related to time of day. Each point represents one pedal rate condition(mean + SD, n ¼ 9: 4 70 rev . min21; W 90 rev . min21; A 120 rev . min21).

N. Bessot et al.884

If a measurement were made at 02:00 h, it is expected that the range in HRthroughout the 24 h would be similar to the reported circadian variation,and that statistical significance would have been attained.

No significant effect was observed in the pedal rate � time of dayinteraction. However, mean data rose by 1.5 bts . min21, 1.9 bts . min21,and 5.3 bts . min21 between 06:00 and 14:00 h when pedal rate wasimposed at 70, 90 and 120 rev . min21, respectively. Likewise, VO2 analy-sis failed to demonstrate any significant diurnal variation. Interestingly,the interaction between pedal rate and time-of-day effects tended to be sig-nificant ( p ¼ 0.08). This trend concerned the experimental data at120 rev . min21, which demonstrated a continuous but non-significantrise from 06:00 to 18:00 h and a decrease thereafter until 22:00 h.

Under submaximal exercise conditions, studies that focused on diurnalvariation in cardiorespiratory variables reported equivocal results. It hasbeen clearly demonstrated that the magnitude of the diurnal variationtends to decline as exercise intensity increases (Cable and Reilly, 1987;Hill et al., 1989). The exercise intensity chosen for this study wasperhaps too high, which might explain why the results failed to demon-strate any significant variation in HR and VO2 values.

Strong diurnal rhythmicity was noted in VE, which would be in agree-ment with previous reports (Reilly and Brooks, 1982; Spengler and Shea,2000). This rise in VE throughout the day, during submaximal effort, maybe partly due to diurnal change in airway resistance and pulmonary diffu-sing capacity. A steady fall in pulmonary-diffusing capacity between 09:30to 21:30 h at rest was first observed many decades ago by Cinkotai andThompson (1966), and this has recently been put forward as an expla-nation for the diurnal variation in VE (Reilly and Brooks, 1990; Reillyet al., 1997). Variation in the amplitude of VE was significantly higherwhen a high pedal rate was imposed.

To better understand the respective influence of pedal rate, time ofday, and pedal rate � time-of-day interaction in breathing pattern, thestudy focused on TV and BF relations during the test. A significantincrease in BF was first observed when the pedal rate was increased, nomatter the time point of study—an increase previously documented byChen et al. (1999). Furthermore, no significant differences were foundin the magnitude of the diurnal variation between the pedal rates usedin the study. The lack of any significant pedal rate effect on the magnitudeof BF variation fails to explain the rise observed in the VE diurnal ampli-tude at high pedal rates (120 rev . min21). Tidal volume also increasedwith pedal rate, as classically observed by Chen et al. (1999). Thecurrent results also demonstrated a significant pedal rate � time-of-dayinteraction in tidal volume. The higher the pedal rate, the greater the mag-nitude of the diurnal variation in TV. This interaction between pedal rateand time-of-day effects in TV may explain why the variation in VE

Effect of Pedal Rate on Diurnal Variations 885

amplitude was higher with the imposed pedal rate of 120 rev . min21. Therise of VE throughout the day is mainly due to BF variation, the diurnalvariation of TV (all pedal rate conditions taken together) being non-signifi-cant. However, the effect of pedal rate � time-of-day interaction on VE canonly be explained by the increase of TV throughout the day, specific to thehigher pedal rate conditions.

Although no significant rhythmicity was observed in HR and VO2

values regardless of pedal rate, pedal rate cannot be ruled out as an influ-ence in chronobiological studies. The influence of pedal rate was con-firmed in the VE analysis, which showed a significant increase invariation amplitude at 120 rev . min21.

In conclusion, although no significant variation was observed in HRand VO2, no matter the pedal rate, it was noted that the differencesobserved between the time points of study increased as pedal rateincreased. The greatest influences were observed in the VE analysis,which showed significant interaction between time of day and pedal rateeffects. The higher the pedal rate, the greater the magnitude of thediurnal variation. Pedal rate as power output must be taken into accountwhen circadian variation of physiological variables is studied in cycling.Using a high pedal rate and low power output should increase the ampli-tude of the diurnal variation of cardiorespiratory variables. Many studiesconducted in the field of chronobiology have demonstrated diurnal fluctu-ations in performance in simulated competitions (Drust et al., 2005). Theinfluence of pedal rate on the diurnal fluctuation of cycling performancemust be investigated.

REFERENCES

Baxter, C., Reilly, T. (1983). Influence of time of day on all out swimming. Br. J. Sports Med. 17:122–127.

Bergh, U., Ekblom, B. (1979). Influence of muscle temperature on maximal muscle strength andpower output in human skeletal muscles. Acta. Physiol. Scand. 107:33–37.

Bird, S.P., Tarpenning, K.M. (2004). Influence of circadian time structure on acute hormonal responseto a single bout of heavy resistance exercise in weight-trained men. Chronobiol. Int. 21:131–146.

Boning, D., Gonen, Y., Maassen, N. (1984). Relationship between work load, pedal frequency, andphysical fitness. Int. J. Sports Med. 5:92–97.

Cable, T., Reilly, T. (1987). Influence of circadian rhythms on arm exercise. J. Hum. Mov. Stud. 13:13–27.

Chen, B., Jones, N.L., Killian, K.J. (1999). Is there a conflict between minimizing effort and energyexpenditure with increasing velocities of muscle contraction in human? J. Physiol. 518:933–940.

Cinkotai, F.F., Thompson, M.L. (1966). Diurnal variation of pulmonary diffusing capacity for carbonmonoxide. J. App. Physiol. 21:535–539.

Deschenes, M.R., Kraemer, W.J., Bush, J.A., Doughty, T.A., Kim, D., Mullen, K.M., Ramsey, K. (1998).Biorhythmic influences on functional capacity of human muscle and physiological responses. Med.Sci. Sports Exerc. 30:1399–1407.

Drust, B., Waterhouse, G., Atkinson, G., Edwards, B., Reilly, T. (2005). Circadian rhythms in sportsperformance—an update. Chronobiol. Int. 21:21–44.

N. Bessot et al.886

Francescato, M.P., Girardis, M., di Prampero, P.E. (1995). Oxygen cost of internal work during cycling.Eur. J. Appl. Physiol. Occup. Physiol. 72:51–57.

Gauthier, A., Davenne, D., Gentil, C., Van Hoecke, J. (1997). Circadian rhythm in the torquedeveloped by elbow flexors during isometric contraction: effect of sampling schedules. Chronobiol.Int. 14:287–294.

Giacomoni, M., Bernard, T., Gavarry, O., Altare, S., Falgairette, G. (1999). Diurnal variations inventilatory and cardiorespiratory responses to submaximal treadmill exercise in females. Eur.J. Appl. Physiol. Occup. Physiol. 80:591–597.

Guette, M., Gondin, J., Martin, A. (2005). Time of day effect on the torque and neuromuscular pro-perties of dominant and non-dominant quadriceps femoris. Chronobiol. Int. 22:541–558.

Hill, D.W., Cureton, K.J., Collins, M.A., Grisham, S.C. (1988). Diurnal variations in responses to exer-cise of “morning types” and “evening types.” J. Sports Med. Phys. Fitness. 28:213–219.

Hill, D.W., Cureton, K.J., Collins, M.A. (1989). Effect of time of day on perceived exertion at work ratesabove and below the ventilatory threshold. Res. Q. Exerc. Sport. 60:127–133.

Horne, J.A., Ostberg, O. (1976). A self-assessment questionnaire to determine morningness-evening-ness in human circadian rhythms. Int. J. Chronobiol. 4:97–110.

Moussay, S., Bessot, N., Gauthier, A., Larue, J., Sesboue, B., Davenne, D. (2003). Diurnal variations incycling kinematics. Chronobiol. Int. 20:879–892.

Pearson, S.J., Gladys, N.L., Onambele, N.L. (2005). Acute change in knee extensors torque, fiber pen-nation, and tendon characteristics. Chronobiol. Int. 22:1013–1027.

Racinais, R., Hue, O., Blonc, S. (2004). Time-of-day effect on anaerobic muscular power in a modera-tely warm environment. Chronobiol. Int. 21:485–495.

Racinais, R., Connes, P., Bishop, D., Blonc, S., Hue, O. (2005). Morning versus evening power outputand repeated sprint ability. Chronobiol. Int. 22:1029–1039.

Ralph, M.R., Foster, R.G., Davis, F.C., Menaker, M. (1990). Transplanted suprachiasmatic nucleusdetermines circadian period. Science 247:975–978.

Reilly, T., Brooks, G.A. (1982). Investigation of circadian rhythms in metabolic responses to exercise.Ergonomics 25:1093–1107.

Reilly, T., Robinson, G., Minors, D.S. (1984). Some circulatory responses to exercise at different timesof day. Med. Sci. Sports Exerc. 16:477–482.

Reilly, T., Brooks, G.A. (1990). Selective persistence of circadian rhythms in physiological responses toexercise. Chronobiol. Int. 7:59–67.

Reilly, T., Atkinson, G., Waterhouse, J. (1997). Biological Rhythms and Exercise; Oxford: Medical Publi-cations, p. 109.

Spengler, C.M., Shea, S.A. (2000). Endogenous circadian rhythm of pulmonary function in healthyhumans. Am. J. Respir. Crit. Care. Med. 163:1038–1046.

Spengler, C.M., Czeisler, C.A., Shea, S.A. (2000). An endogenous circadian rhythm of respiratorycontrol in humans. J. Physiol. 526:683–694.

Taillard, J., Sanchez, P., Lemoine, P., Mouret, J. (1990). Heart rate circadian rhythm as a biologicalmarker of desynchronization in major depression: a methodological and preliminary report.Chronobiol. Int. 7:305–316.

Touitou, Y., Portaluppi, F., Smolensky, M.H., Rensing, L. (2004). Ethical principles and standards forthe conduct of human and animal biological rhythm research. Chronobiol. Int. 21:161–170.

Waterhouse, J., Drust, B., Weinert, D., Edwards, B., Gregson, W., Atkinson, G., Kao, S., Aizawa, S.,Reilly, T. (2005). The circadian rhythm of core temperature: origin and some implications forexercise performance. Chronobiol. Int. 22:207–225.

Effect of Pedal Rate on Diurnal Variations 887