Running Economy of African and Caucasian Distance Runners

Running economy of African and Caucasian distance runners

ADELE` R. WESTON, ZIPHELELE MBAMBO, and KATHRYN H. MYBURGH MRC/UCT Bioenergetics of Exercise Research Unit, Department of Physiology, University of Cape Town Medical School, SOUTH AFRICA; and School of Exercise and Sport Science, University of Sydney, AUSTRALIA

ABSTRACT WESTON, A. R., Z. MBAMBO, and K. H. MYBURGH. Running economy of African and Caucasian distance runners. Med. Sci. Sports Exerc., Vol. 32, No. 6, pp. 1130–1134, 2000. Purpose: Anecdotal evidence suggests an advantageous physiological endowment ˙ of the African endurance athlete. Higher fractional utilization of VO2max has been suggested but not measured directly, and investigations of running economy have been inconclusive. The aim of the current study was to measure a) running economy and b) ˙ fractional utilization of VO2max, in African and Caucasian 10-km runners of similar body mass. Methods: Eight African and eight Caucasian runners had no significant difference in mean race time (32.8 Ϯ 2.8, 32.0 Ϯ 2.5 min, respectively), body mass (61.4 Ϯ 7.0, 64.9 Ϯ 3.0 kg), age, body fat, or lean thigh volume. Caucasian runners were 6 cm taller (P Ͻ 0.05). Subjects completed a progressive ˙ ⅐ Ϫ1 treadmill VO2peak test. On a separate day, subjects completed two 6-min workloads (16.1 km h and 10-km race pace) separated by ˙ Ϯ Ϯ ⅐ Ϫ1⅐ Ϫ1 ϭ ⅐ Ϫ1 5 min. Results: Mean VO2peak was 13% lower in the Africans (61.9 6.9, 69.9 5.4 mL kg min , P 0.01). At 16.1 km h , the Africans were 5% more economical (47.3 Ϯ 3.2, 49.9 Ϯ 2.4 mL⅐kgϪ1⅐minϪ1, P Ͻ 0.05). This difference increased to 8% (P Ͻ 0.66 ˙ Ϯ Ϯ Ͻ 0.01) when standardized per kg . At race pace, the Africans utilized a higher %VO2peak (92.2 3.7, 86.0 4.8%, P 0.01) and had higher HR (185 Ϯ 9, 174 Ϯ 11 b⅐minϪ1, P Ͻ 0.05) and plasma [ammonia] (113.2 Ϯ 51, 60.3 Ϯ 16.9 ␮mol⅐LϪ1, P Ͻ 0.05). Despite the higher relative workload, the plasma [lactate] was not different (5.2 Ϯ 2.0, 4.2 Ϯ 1.7 mmol⅐LϪ1, NS). Conclusions: This study ˙ indicates greater running economy and higher fractional utilization of VO2peak in African distance runners. Although not elucidating the origin of these differences, the findings may partially explain the success of African runners at the elite level. Key Words: 10-KM RUNNERS, EFFICIENCY, RACE, ENDURANCE PERFORMANCE

frican runners dominate international distance run- economical than Scandinavian runners. To account for the ning events although the explanation for their dis- sizable difference in mean body mass between the two ˙ Ϫ0.75 Aproportionate success is unclear. Limited investiga- groups, VO2 was normalized per kg rather than per kg, ˙ tions have confirmed that VO2max is not unduly high in these as suggested by Svedenhag and Sjodin (16). This accentu- individuals (2,4) and therefore cannot explain their success. ated the difference in oxygen cost for a given running speed It is necessary to consider other contributory parameters between the Africans and the Caucasians. Furthermore, this ˙ such as fractional utilization of VO2max and running difference was more pronounced at faster speeds. Therefore, economy. the running economy data reported to date are conflicting Early investigators suggested that African individuals and are confounded by large differences in body mass may have greater economy of movement, albeit predomi- between groups. At present, it is not clear whether African nately with regard to stepping tasks (12,13,18). However, runners have advantageous running economy when com- investigations of running economy in African and Cauca- pared with their Caucasian counterparts. sian runners have proved inconclusive. Coetzer et al. (4) Success in distance running can also be explained by the found no difference in oxygen consumption at 17 km⅐hϪ1 ability to maintain a high proportion of one’s maximal when expressed relative to body weight. However, these oxygen consumption throughout the event. Previous studies subjects were not matched for primary race distance as the have indicated that African distance runners race at a higher ˙ Caucasians were predominantly middle-distance runners percentage of VO2max (2,4). Bosch et al. (2) investigated and the Africans were predominantly long-distance runners. physiological characteristics of African and Caucasian dis- Furthermore, the difference in body mass was large (mean tance runners while they ran a simulated marathon on the for African ϭ 56 kg vs Caucasian ϭ 70 kg). In contrast, treadmill. When running at the same percentage (ϳ87%) of Saltin et al. (14) found that Kenyan runners were more their best marathon time, the African runners ran at a higher ˙ Ͻ percentage of VO2max (76% vs 68%, P 0.05), a higher 0195-9131/00/3206-1130/0 heart rate and a higher RER. In addition, Coetzer et al. (4) ˙ ˙ MEDICINE & SCIENCE IN SPORTS & EXERCISE® extrapolated race VO2 from the VO2 vs treadmill speed Copyright © 2000 by the American College of Sports Medicine relationship obtained during a nonsteady state incremental Submitted for publication March 1999. exercise test. On this basis, the African runners raced 10 km ˙ Accepted for publication September 1999. at a higher percentage of their VO2max. Weston et al. (17) 1130 reported that African 10-km runners were able to continue testing. This included treadmill running at low and high running at a high percentage of peak treadmill velocity speeds and with and without the face mask. Further famil- (92%) for almost twice as long as Caucasian 10-km runners. iarization sessions were undertaken by those subjects who Although these studies all infer that fractional utilization of were less familiar with treadmill running if deemed neces- ˙ VO2max is greater in African runners, this has not been sary in the opinion of the investigator. All subjects then measured during steady state exercise at actual race pace. completed a peak treadmill velocity test with concurrent ˙ ˙ The mechanism for improved fractional utilization may measurement of VO2,VE, and RER using the same cali- be metabolic in origin. Metabolic investigations indicate brated on-line gas analysis system (Oxycon Alpha, Jaeger that African distance runners accumulate less plasma lactate Mijnhardt, Wuerzburg, The Netherlands) with HR measured (4,14,17) and less plasma ammonia (14) than their Cauca- throughout (Polar Sport tester, Kempele, Finland). At ex- sian counterparts at comparable workloads. However, no actly 3 min post exercise, a venous blood sample was studies to date have directly measured RER and plasma obtained for determination of plasma lactate concentrations. lactate and ammonia concentrations in African and Cauca- Samples were immediately centrifuged at 4°C at 3000 rpm sian distance runners while running at current race pace in and plasma was frozen for later spectrophotometric analysis a controlled laboratory setting. in duplicate (bioMerieux lactate PAP, Boehringer Mann- Therefore, the aim of this study was to investigate run- heim, Germany) using a Beckman DU-62 spectrophotome- ning economy, fractional utilization of maximal oxygen ter (Beckman Industries Inc., Palo Alto, CA). consumption and metabolite concentrations, at a steady state Submaximal exercise testing. On a separate day, Ϫ1 absolute workload (16.1 km⅐h ) and at 10-km race pace in subjects returned to the laboratory 3 h after eating. Subjects African and Caucasian groups of runners who are matched did not ingest caffeine overnight or before the test. Subjects for current 10-km performance. In addition, to minimize the again warmed up at 14 km⅐hϪ1 for 5 min. Subjects then potential effects of body mass on running economy, the completed two submaximal workloads, one at 16.1 km⅐hϪ1 groups were as closely matched as possible for body mass. and the other at current 10-km race pace, with all subjects utilizing the same calibrated treadmill (Powerjog EG30, Birmingham, England). Each workload was undertaken for METHODS ˙ ˙ 6 min separated by a rest period of 5 min. VO2,VE, RER, Subjects and HR were measured continuously and results averaged over 15-s intervals. Mean values of the last 60 s were Sixteen well-trained distance runners (8 African, 8 Cau- designated as steady state values for data analysis. A venous casian), with a 10-km race time range of 29.2–37.0 min, blood sample was obtained exactly 1 min after exercise for were recruited for the study from local running clubs. Mean each workload for determination of plasma ammonia and Ϯ Ϯ ages were 28.0 7.0 for Africans and 25.9 5.3 for plasma lactate concentrations. Samples were immediately Caucasians. All reported their primary running distance to centrifuged at 4°C at 3000 rpm and plasma was obtained. be 10 km. The two groups were of the same performance Plasma ammonia was assayed spectrophotometrically in standard with no difference in mean current 10-km race time duplicate within2h(NH kit, Boehringer Mannheim), and Ϯ Ϯ 3 (African 32.8 2.8 min; Caucasian 32.0 2.5 min). Every plasma was stored for later spectrophotometric analysis of effort was made to recruit African and Caucasian athletes of lactate in duplicate (bioMerieux lactate PAP, Boehringer similar body mass. All subjects were seasoned competitors Mannheim). (at least 3 yr of competitive running) and were currently in Ϫ1 Running economy was defined as the steady-state oxygen training for a 10-km race (108 Ϯ 43 km⅐wk ). All subjects Ϫ Ϫ consumption in mL⅐kg 1⅐min 1 obtained at each workload. were informed of the possible risks of the experimental V˙ O was also normalized per kg0.66 (1,10). procedures, and all gave their written informed consent. The 2 study was approved by the Ethics and Research committee of the Faculty of Medicine at the University of Cape Town. Statistics The Student’s unpaired t-test was used to compare Afri- Procedures can and Caucasian runners, and Pearson’s correlation coef- Anthropometry. Height, weight, thigh length, and mid- ficient was used to investigate relationships between vari- thigh circumference were measured. Five skin-fold mea- ables for all individuals. Correlations were not calculated surements were made (tricep, bicep, suprailiac, subscapular, within groups due to insufficient numbers. mid-thigh) and % body fat calculated using the formula of Durnin and Wolmersley (9). Lean thigh volume was calculated using the assumption of the thigh as a truncated cone RESULTS and taking into account the thigh skin-fold measurement (11). Lean thigh volume data were unavailable from two of Anthropometric results showed no significant difference the African subjects. between the groups in body mass, % body fat or lean thigh ˙ VO2peak/peak treadmill velocity test. All subjects volume. However, the Caucasian subjects were 6 cm taller had undertaken at least one familiarization session before (Table 1).

RUNNING ECONOMY OF AFRICAN DISTANCE RUNNERS Medicine & Science in Sports & Exerciseா 1131 TABLE 1. Anthropometric results. TABLE 3. Steady state exercise test results. African Caucasian African Caucasian Significance (8 ؍ N) (8 ؍ Significance (N (8 ؍ N) (8 ؍ N) Height (cm) 172.4 Ϯ 5.3 178.3 Ϯ 5.3 P Ͻ 0.05 16.1 km⅐hϪ1: Mass (kg) 61.4 Ϯ 7.0 64.9 Ϯ 3.0 NS HR (b⅐minϪ1) 168 Ϯ 16 160 Ϯ 15 NS % body fat 12.1 Ϯ 3.7 12.1 Ϯ 3.3 NS RER 0.94 Ϯ 0.03 0.94 Ϯ 0.04 NS Ϯ Ϯ ˙ ⅐ Ϫ1⅐ Ϫ1 Ϯ Ϯ Ͻ Lean thigh volume (cc) 2152 770* 2568 295 NS VO2 (mL kg min ) 47.4 3.2 49.9 2.4 P 0.05 V˙ O (mL⅐kgϪ0.66⅐minϪ1) 191.6 Ϯ 13.0 206.2 Ϯ 8.2 P Ͻ 0.01 * Indicates N ϭ 6; results are mean Ϯ SD. 2 ˙ ⅐ Ϫ1 Ϯ Ϯ VE (L min ) 90.0 7.7 89.8 12.1 NS 10-km race pace Ϫ Maximal treadmill test results are listed in Table 2. All HR (b⅐min 1) 185 Ϯ 9 174 Ϯ 11 P Ͻ 0.05 RER 0.99 Ϯ 0.05 0.99 Ϯ 0.06 NS subjects achieved a RER value of higher than 1.0 and a ˙ ⅐ Ϫ1⅐ Ϫ1 Ϯ Ϯ VO2 (mL kg min ) 57.2 6.7 59.5 5.9 NS ˙ ⅐ Ϫ0.66⅐ Ϫ1 Ϯ Ϯ maximal heart rate of above 90% of their age-predicted VO2 (mL kg min ) 230.8 23.3 245.8 26.0 NS %V˙ O 92.2 Ϯ 3.7 86.0 Ϯ 4.8 P Ͻ 0.01 Ϫ ˙ 2peak HRmax (220 age). VO2peak was considerably higher in the ˙ ⅐ Ϫ1 Ϯ Ϯ VE (L min ) 113.5 15.8 119.2 19.3 NS Ϯ Ϯ Caucasian athletes, even when corrected for body mass %HRmax 96 3927NS (13%, P ϭ 0.01); however, this was not reflected by a faster Results are mean Ϯ SD. mean 10-km race time. Mean peak treadmill velocity (PTV) was also higher in the Caucasian runners than in the African Caucasian group over 10 km, despite having a considerably runners, but this did not reach significance. Ϫ lower maximal oxygen uptake (13% lower, P ϭ 0.01). Results of the exercise test at 16 km⅐h 1 and at calculated Ϫ Physiological factors that may contribute to such a finding 10-km race pace are listed in Table 3. At 16.1 km⅐h 1, the are a) greater running economy or b) the ability to sustain a African runners had a 5% lower mean oxygen consumption higher percentage of their V˙ O throughout the event, or (P Ͻ 0.05, Fig. 1), and when this was expressed per kg0.66, 2max a combination of both. These two factors have been sys- the difference was increased to 8% (P Ͻ 0.01). The latter tematically investigated in the current study. individual data (lower panel) shows only one Caucasian The oxygen consumption per kilogram body weight of subject with an oxygen consumption below the African the Caucasian subjects investigated in the current study group mean and only one African subject with an oxygen Ϫ Ϫ Ϫ (mean 49.9 mL⅐kg 1⅐min 1 at 16.1 km⅐h 1, range 45.0– consumption greater than the Caucasian group mean. Run- Ϫ 52.6) is comparable to that previously reported in distance ning economy at 16.1 km⅐h 1 did not correlate with 10-km runners. Costill et al. (6), Daniels et al. (7), and Conley and performance in this homogeneous group. The data from the race pace workload indicate that the African runners were able to utilize a considerably higher percentage of their maximal oxygen uptake (Table 3) and to exercise at a higher heart rate. Considerable accumulation of ammonia occurred at race pace. Although highly variable between individuals, the mean was higher in the African runners than in the Caucasian runners. The race pace ammonia results include one African individual who recorded a particularly high plasma ammonia of 217 ␮mol⅐LϪ1 (Ͼ2 SD from the mean of the remainder of the group). However, if this individual was removed from the analysis, the mean (95.9 Ϯ 25.0) was still significantly higher than that of the Caucasians (P ϭ 0.01). Accumulation of lactate in the plasma was not higher in the African runners despite the higher relative exercise intensity (Table 4).

DISCUSSION The results indicate that the African athletes in the current study were able to achieve the same performance as the

TABLE 2. Maximal treadmill test results. African Caucasian Significance (8 ؍ N) (8 ؍ N) -Peak treadmill velocity (km⅐hrϪ1) 21.8 Ϯ 1.7 23.0 Ϯ 1.6 NS Figure 1—Oxygen consumption when running at 16.1 km⅐h؊1 in in ˙ ⅐ Ϫ1 Ϯ Ϯ Ͻ VO2peak (L min ) 3.8 0.4 4.6 0.5 P 0.01 dividual African (F) and Caucasian (E) runners. The subjects are ˙ ⅐ Ϫ1⅐ Ϫ1 Ϯ Ϯ ϭ VO2peak (mL kg min ) 60.7 7.5 69.9 5.4 P 0.01 displayed from most economical to least economical in each panel; ˙ ⅐ Ϫ0.66⅐ Ϫ1 Ϯ Ϯ Ͻ VO2peak (mL kg min ) 250.2 24.5 289.2 25.5 P 0.01 Ϫ therefore, an individual subject may not hold the same position in both HR (b⅐min 1) 194 Ϯ 3.7 189 Ϯ 5.7 P Ͻ 0.05 ˙ ⅐ ؊1⅐ ؊1 max panels. The upper panel indicates VO2 (mL kg min ) and the lower Ϯ Ϯ ؊0.66 ؊1 RERmax 1.10 0.06 1.13 0.05 NS panel indicates V˙ O (mL⅐kg ⅐min ). Mean values for Africans and ⅐ Ϫ1 Ϯ Ϯ 2 Plasma [lactate] (mmol L ) 10.7 2.6 12.6 2.9 NS Caucasians are indicated with the solid and dashed lines, respectively, Results are mean Ϯ SD. with the level of significance indicated.

1132 Official Journal of the American College of Sports Medicine http://www.msse.org TABLE 4. Plasma ammonia and lactate concentrations. as previously reported (17), in data that were inclusive of African Caucasian some of these subjects. Significance Despite the significant difference in fractional utilization (8 ؍ N) (8 ؍ N) ⅐ Ϫ1 ␮ ⅐ Ϫ1 Ϯ Ϯ 16.1 km h NH3 ( mol L ) 77.3 39.9** 54.9 6.1* NS ˙ ␮ ⅐ Ϫ1 Ϯ Ϯ Ͻ of VO2peak between the two groups in this study, the range Race pace NH3 ( mol L ) 113.2 51.2** 60.3 16.9* P 0.05 Ϫ1 Ϫ1 ˙ 16.1 km⅐hr lactate (mmol⅐L ) 2.8 Ϯ 1.7 2.1 Ϯ 1.6 NS of fractional utilization of VO2peak reported for both African Race pace lactate (mmol⅐LϪ1) 5.2 Ϯ 2.0 4.2 Ϯ 1.7 NS and Caucasian runners is similar to those previously re- * Indicates N ϭ 5, ** indicates N ϭ 7; results are mean Ϯ SD. ported for runners (6,15). The mean value of ϳ89% for all runners in the current study is only slightly higher than that ˙ Krahenbuhl study (5) reported a mean VO2 of 51.7 reported by Costill et al. (6) and Scrimgeour et al. (15). Ϫ Ϫ Ϫ Ϫ mL⅐kg 1⅐min 1, 50.5 mL⅐kg 1⅐min 1, and 50.3 These authors reported the mean %V˙ O sustained during Ϫ Ϫ 2max mL⅐kg 1⅐min 1, respectively, in highly trained distance 10-km and 10-mile races, respectively, was 85.8% and Ϫ runners when running at 16.1 km⅐h 1. 86.1%. The values reported by Davies and Thompson (8), of However, the findings of the current study showed a 5% four ultra-distance world record holders, are more similar to greater running economy in the African runners when com- those in the current study (ϳ89%). Therefore, the subelite pared with the Caucasian runners (P Ͻ 0.05) and a mean African runners in the current study are not outside the Ϫ1 Ϫ1 (47.4 mL⅐kg ⅐min ) that is lower than the previously normal range for well-trained distance runners but rather are reported values above. Only one Caucasian subject had an representative of the higher end of the range. oxygen consumption below the mean African value (see At race pace, both plasma lactate and plasma ammonia Fig. 1). The difference in mean body mass between the two concentrations were considerably elevated, with mean groups in the current study was relatively small and statis- plasma lactate concentrations above the classically de- tically insignificant (3.5 kg). scribed “onset of blood lactate accumulation” value of 4 Running economy is regarded as the metabolic power mmol⅐LϪ1. This indicates that exercise lasting as long as 30 required to run at a particular velocity and metabolic power min can be maintained despite substantial lactate accumu- 0.66 has been reported to be proportional to mass (10). There- lation. Ammonia accumulation in the plasma was higher in fore, when calculating running economy, it is appropriate to the African runners than the Caucasians. Because there were 0.66 scale oxygen consumption per kg rather than per kg. In no significant differences in lactate accumulation, this ˙ 0.66 the current study, when VO2 is normalized per kg , the means that the ammonia accumulation relative to lactate results are accentuated and running economy is 8% better in accumulation is higher in the Africans. As it is thought that Ͻ the Africans than the Caucasians (P 0.05). Our finding of the accumulation of ammonia and lactate generally occur in a significant difference in running economy is in agreement parallel (3), this suggests the lactate removal may be en- with the findings of Saltin et al. (14) in Kenyan runners. The hanced in the African runners. previous studies that do not concur were studies in South In summary, the results indicate that the African runners African runners (2,4) where there was a sizable difference in in this study were more economical than the Caucasian mean mass between the subject groups (12 kg and 14 kg, 0.66 runners. Furthermore, the African runners raced 10 km at a respectively) and results were not scaled relative to mass . ˙ higher percentage of their VO2peak, whereas only accumu- The results further indicate that the African runners eval- lating a similar concentration of plasma lactate. These char- uated in this study were able to race 10 km at a significantly acteristics explain the ability of these African runners to higher relative intensity than Caucasian runners. At 10-km achieve a comparable race performance despite a consider- race pace, the Africans utilized a considerably higher per- ably lower V˙ O . If these advantageous characteristics centage of their maximal oxygen uptake than the Caucasians 2peak Ͻ are present in elite African distance runners, these findings (92.2% vs 86.0% respectively, P 0.01). Running at a may assist in the explanation of the disproportionate success higher relative intensity elicited a higher HR and a higher of this racial group in distance running events. concentration of plasma ammonia. This could potentially be the result of either a physiological or motivational advan- tage. However, if this was solely the result of a difference in The authors wish to thank the subjects for their participation and motivation, one would also expect the RER and plasma Julia Goedecke and Judy Belonje for their technical assistance. lactate to be higher, but this was not the case. Despite higher While in Cape Town, Ade` le Weston was supported by the J. W. relative intensity at race pace, the African athletes had the Jagger Gift Scholarship for Foreign Students and the R. A. Noakes Medical Research Fellowship. The Bioenergetics of Exercise Re- same mean RER as the Caucasian group. Therefore, the search Unit is funded by the Medical Research Council of South African athletes had the same carbohydrate flux despite Africa and the Foundation for Research Development. running at an intensity closer to V˙ O . It is tempting to Current address for Dr. Weston is the School of Exercise and 2peak Sport Science, University of Sydney, PO Box 170, Lidcombe, NSW speculate that race pace is in fact determined by carbohy- 1825, Australia and for Dr. Myburgh is the Department of Human drate flux. The fact that the African athletes only achieved and Animal Physiology, University of Stellenbosch, Private bag XI, the same carbohydrate flux and subsequent plasma lactate Matieland 7602, South Africa. Address for correspondence: Dr Ade` le Weston, School of Exer- accumulation at a higher relative intensity of exercise may cise and Sport Science, University of Sydney, PO Box 170, Lid- be due to higher skeletal muscle oxidative enzyme activity combe, NSW 1825, Australia; E-mail: [email protected].

RUNNING ECONOMY OF AFRICAN DISTANCE RUNNERS Medicine & Science in Sports & Exerciseா 1133 REFERENCES

1. BERGH, U., B. SJODIN,A.FORSBERG, and J. SVEDENHAG. The rela- M. Nimmo, T. Reilly, and C. Williams (Eds.). London: E& FN tionship between body mass and oxygen uptake during running in Spon, 1987, pp. 205–219. humans. Med. Sci. Sports Exerc. 23:205–211, 1991. 11. KATCH, V. L., and F. I. KATCH. A simple anthropometric method 2. BOSCH, A. N., B. R. GOSLIN,T.D.NOAKES, and S. C. DENNIS. for calculating segmental leg limb volume. Res. Q. 45:211–214, Physiological differences between black and white runners during 1974. a treadmill marathon. Eur. J. Appl. Physiol. 61:68–72, 1990. 12. LEARY, W. P., and C. H. WYNDHAM. The capacity for maximum 3. BUONO, M. J., T. R. CLANCY, and J. R. COOK. Blood lactate and physical effort of Caucasian and Bantu athletes of international ammonia ion accumulation during graded exercise in humans. class. S. Afr. Med. J. 39:651–655, 1965. J. Appl. Physiol. 57:135–139, 1984. 13. ROBINSON, J., D. B. DILL,P.M.HARMAN,F.G.HALL, and J. W. 4. COETZER, P., T. D. NOAKES,B.SANDERS, et al. Superior fatigue WILSON. Adaptation to exercise of Negro and White sharecrop- resistance of elite black South African distance runners. J. Appl. pers: in comparison with Northern Whites. Hum. Biol. 13:139– Physiol. 75:1822–1827, 1993. 158, 1941. 5. CONLEY, D. L., and G. S. KRAHENBUHL. Running economy and 14. SALTIN, B., H. LARSEN,N.TERRADOS, et al. Aerobic exercise distance running performance of highly trained athletes. Med. Sci. capacity at sea level and at altitude in Kenyan boys, junior and Sports Exerc. 12:357–360, 1980. senior runners compared with Scandinavian runners. Scand. 6. COSTILL, D. L., H. THOMASON, and E. ROBERTS. Fractional utiliza- J. Med. Sci. Sports 5:209–221, 1995. tion of the aerobic capacity during distance running. Med. Sci. Sports Exerc. 5:248–252, 1973. 15. SCRIMGEOUR, A. G., T. D. NOAKES,B.ADAMS, and K. H. MYBURGH. The influence of weekly training distance on fractional utilisation 7. DANIELS, J. T., G. KRAHENBUHL,C.FOSTER,J.GILBERT, and S. of maximal aerobic capacity in marathon and ultramarathon run- DANIELS. Aerobic responses of female distance runners to submaximal and maximal exercise. Ann. N. Y. Acad. Sci. 301:726– ners. Eur. J. Appl. Physiol. 55:202–209, 1986. 733, 1977. 16. SVEDENHAG, J., and B. SJODIN. Body mass modified running econ- 8. DAVIES, C. T. M., and M. W. THOMPSON. Aerobic performance of omy and step length in elite male middle- and long-distance female marathon and male ultramarathon athletes. Eur. J. Appl. runners. Int. J. Sports Med. 15:305–310, 1994. Physiol. 41:233–245, 1979. 17. WESTON, A. R., O. KARAMIZRAK,A.SMITH,T.D.NOAKES, and K. H. 9. DURNIN, J. V. G. A., and J. WOMERSLEY. Body fat assessment from MYBURGH. African distance runners exhibit greater fatigue resis- total body density and its estimation from skinfold thickness: tance, lower lactate accumulation and higher oxidative enzyme measurements on 481 men and women aged from 16 to 72 years. activity. J. Appl. Physiol. 86:915–923, 1999. Br. J. Nut. 32:77–97, 1974. 18. WYNDHAM, C. H., N. B. STRYDOM,J.F.MORRISON, et al. Differ- 10. FREDERICKS, E. C. Biomechanical aspects of endurance. In: Exer- ences between ethnic groups in physical working capacity. J. Appl. cise-Benefits, Limits and Adaptations, D. Macleod, R. Maughan, Physiol. 18:361–366, 1963.

1134 Official Journal of the American College of Sports Medicine http://www.msse.org