The main purpose of cyclic events, either short RELIABILITY AND VALIDITY OF THE MODIFIED CONCONI TEST ON CONCEPT II ROWING ERGOMETERS INTRODUCTION

Transcription

1 Journal of Strength and Conditioning Research, 2005, 19(4), National Strength & Conditioning Association RELIABILITY AND VALIDITY OF THE MODIFIED CONCONI TEST ON CONCEPT II ROWING ERGOMETERS ÖZGÜR ÇELIK, 1 ŞÜKRAN NAZAN KOŞAR, 2 FEZA KORKUSUZ, 1 AND MURAT BOZKURT 3 1 Middle East Technical University, Faculty of Education, Department of Physical Education and Sports, Ankara, Turkey; 2 Hacettepe University, School of Sport Sciences and Technology, Ankara, Turkey; 3 Dr. Muhittin Ulker Emergency Care and Traumatology Hospital, Ankara, Turkey. ABSTRACT. Çelik, Ö., Ş.N. Koşar, F. Korkusuz, and M. Bozkurt. Reliability and validity of the modified Conconi test on Concept II rowing ergometers. J. Strength Cond. Res. 19(4): The purpose of this study was to assess the reliability and validity of the modified Conconi test on Concept II rowing ergometers. Twenty-eight oarsmen conducted 3 performance tests on separate days. Reliability was assessed using the break point in heart rate (HR) linearity called the Conconi test (CT) and Conconi retest (CRT) for the noninvasive measurement of anaerobic threshold (AT). Blood lactate measurement was considered the gold standard for the assessment of the AT, and the validity of the CT was assessed by blood samples taken during an incremental load test (ILT) on ergometers. According to the results, the mean power output (PO) scores for the CT, CRT, and ILT were W, W, and W, respectively. The mean HR values at the AT for the CT, CRT, and ILT were b min, b min, and b min, respectively. Interclass correlation coefficient (ICC) analysis indicated a significant correlation between the 3 tests with one another. Also, Bland and Altman plots showed that there was an association between noninvasive tests and the ILT PO scores and HRs (95% confidence interval [CI]). In conclusion, this study showed that the modified CT is a reliable and valid method for determining the AT of elite men rowers. KEY WORDS. anaerobic threshold, heart rate break point, oarsmen, noninvasive field test INTRODUCTION The main purpose of cyclic events, either short or long distance, is to complete the competition distance in the shortest time. Sport scientists endeavor to improve physiologic and biomechanic factors that affect performance and prevent injuries. Each sport discipline requires different skills to achieve desirable results. These skills should be improved by training with various techniques. Race rowing is considered one of the most demanding endurance sports (28). The median results in Fédération Internationale des Sociétés d Aviron (FISA) regattas from 1974 to 1989 indicated an average race duration of 6.5 minutes (range, minutes) for elite men oarsmen and 6.6 minutes (range, minutes) for lightweights (24). Therefore, rowing can be classified as a medium- or long-duration endurance sport. The dominant energy system in race rowing is aerobic (23). Medium muscular endurance, long starting power, and maximum strength are also important factors in that sport, and rowing requires endurance and the ability to generate powerful strokes against the water (4, 27). According to this information, muscular endurance, starting power, and maximum strength should be training objectives. In this type of endurance sport, the anaerobic threshold (AT) is an important parameter in the achievement and maintenance of performance. The AT is defined as the maximal work intensity of oxygen intake (V O2 ) beyond which blood lactate concentration increases sharply, causing metabolic acidosis with associated alterations in gas exchange. The AT and other fatigue thresholds are used commonly for adjusting training intensity, monitoring the athletes adaptation to training, and predicting the performance of athletes (16, 19 21, 25). Direct methods have been developed to determine the AT; however, laboratory determination of the AT is limited by the requirements of expensive, sophisticated equipment and trained personnel (31). Field tests were therefore designed to predict the AT. The Conconi test (CT) is one of the field tests commonly used in running events. This test was created to determine the AT by forming a relationship between running speed and heart rate (HR) during an incremental test. The speed vs. HR relationship has been described as linear from low-to-submaximal speeds and curvilinear from submaximal-to-maximal speeds. This transition from the linear to the curvilinear phase coincided with the beginning of blood lactate accumulation. The test was therefore developed as an indirect way to determine the AT in elite runners (10). However, values obtained for those exercise performances are dependent on multiple factors, including the type of physical exercise used in the testing procedure (18). Therefore, in subsequent years, researchers have reported different applications of this test to several age groups, various athletic abilities, and other physical activities (5, 9, 10, 26). However, in some applications, researchers could not find the deflection point, which was caused by some methodologic problems (13). Conconi et al. (10) also explained these problems in such a manner that an increase in intensity was achieved by applying a resistance in these studies (reply to Hofmann et al. [13]). Running is a natural skill of men, and intensity is changed mostly by movement frequency. Although rowing is a learned skill, stroke frequency is an important parameter for intensity. Therefore, the validity of the rowing modality of this test is not clear. Evaluation of elite athletes with their own sport equipment and in their own competition environment is another important factor in AT testing (28). Rowing ergometers are reliable equipment devices that are commonly used for testing oarsmen s performance (29). 871

2 872 ÇELIK, KOŞAR,KORKUSUZ ET AL. In recent years, the CT was modified to determine the AT of rowers on rowing ergometers. However, there were some concerns about these studies: the number of subjects was limited, test ergometers were not used commonly in training or competitions, or subjects of the study were not skillful enough to execute correct rowing movements. Related studies were not compatible with the newly established recommendations of the CT according to Conconi et al. (10) in In the present study, the original Conconi method for running was performed with rowers on rowing ergometers. General principles were implemented according to these recommendations. The duration of each stage was 1 minute, and increments did not increase the pulse rate of subjects more than 8 b min 1. It is hypothesized that with appropriate conditions, the modified CT will be a reliable and valid method to be used in rowing. The modified CT could be considered sport-specific. Although the testing of elite athletes performances with common equipment like treadmills may be easier, the biomechanics of specialized movements may not be precisely duplicated with this type of equipment. Since it is easy to apply and is highly economic, the CT was also accepted as a useful method. Although blood lactate is considered the gold standard to determine the AT, it requires a laboratory setting and is too costly. Therefore, the rowers and coaches could assess the AT under field conditions using an easy method if the CT method can be confirmed against the gold standard. Therefore, the goal of this study was to assess the reliability and validity of the modified CT in highly trained oarsmen on Concept II rowing ergometers. METHODS Experimental Approach to the Problem Treadmill and cycle ergometers are very common testing devices in most of the performance laboratory settings. Although the testing of athletes from different athletic branches on treadmills or cycle ergometers is more accessible, this is not parallel to the principle of specialization in training. Therefore, it is important to measure rowers with their training ergometers. Also, measuring blood lactate has been accepted as the gold standard in the assessment of the AT, but this method is limited to sophisticated laboratory conditions. The CT for the rowing modality seems to be a suitable method for determining the AT of rowers. Although use of the CT may make it easier to determine the AT, the reliability and validity of the test were not assessed for rowing. In the present study, the reliability of the CT for rowers was assessed by testretest measurements, and the validity was assessed by the blood lactate criterion test. Subjects Thirty well-trained oarsmen from 5 different sports clubs volunteered to participate in this study. The availability of subjects for the tests was the major inclusion criterion. Two subjects data were excluded because of the lack of a deflection point in the CT. The experimental protocol was approved by a local ethics committee. All subjects signed a consent form after being fully informed of the study s methods, possible side effects, and purpose. All participants had more than 2 years of rowing experience and performed 6 9 training sessions per week. They also performed almost 30% of their annual training on rowing ergometers. Tests were conducted in April and May at the end of precompetitive phase. Anthropometric Measurements Anthropometric parameters of participants were measured 2 days before the first performance test. Body height and weight were measured with the Seca anthropometer and beam-balance scale (Seca, Vogel & Haike, Hamburg, Germany). Body mass index was also calculated (in kilograms per square meter). Body composition of the participants was obtained using the bioelectrical impedance method (Omron BF 300 Body Fat Monitor, Matsusaka Co. Ltd., Matsusaka, Japan). Performance Test Procedures Participants performed 3 performance tests on 3 separate days. Two of the tests were the CT and Conconi retest (CRT) to confirm the reliability of the noninvasive AT measurement procedure. The last test was the incremental load test (ILT) to confirm the validity of the noninvasive anaerobic test measurement procedure. The intervals between the tests were 3 7 days. Tests were performed in the participants own training environment by the same ergometer. Participants were asked to drink at least 0.5 L of water 30 minutes before the tests to ensure that they were fully hydrated. The food consumption habits of participants were not altered. Participants were only instructed not to engage in strenuous physical activity the day prior to the testing. The testing procedure was as outlined in the following paragraphs. All performance tests were undertaken on a Model-C (Concept II, Inc., Morrisville, VT) rowing ergometer, the most widely used ergometer for training purposes (11). The Concept II ergometer measured the intensity of exercise as power output (PO) (in watts). During all tests, the vanes of the ergometer were kept fully closed, and the ergometers were self-calibrated. The HR was recorded every 5 seconds using a Polar Vantage NV instrument (Polar Electro, Kempele, Finland). CT (noninvasive test). Participants were allowed to warm up for 5 minutes. Participants decided the initial intensity of the warm-up period. After the warm-up period, participants began rowing at a self-selected cadence for the first minute at a steady PO of 75 W. Each minute thereafter, the PO was increased by 25 W (28). This load increment protocol was gradual to increase the corresponding HR by less than 8 b min 1 each minute according to the new specifications of Conconi et al. (10). Each stage was 60 seconds, and the test ended with the volitional exhaustion of the participant (10). Blood lactate was not measured during the tests, but it was measured at the beginning and end of the tests. ILT (invasive test). A continuous progressive load program of 3-minute stages was adapted. The test started with 150 W, and the PO was increased every 3 minutes by 50 W (6, 7, 30). The stroke rate during the tests varied between 20 and 34 strokes per minute. Capillary blood samples (5 l) were obtained from the fingertip at the beginning and during the 30-second breaks between the work stages. Blood lactate concentrations were measured with a Lactate Pro LT-1710 instrument (Arkray KDK Co., Kyoto, Japan). Lactate in the sample reacts with potassium ferricyanide and lactate oxidase to form potassium

3 THE MODIFIED CONCONI TEST FOR ROWING 873 FIGURE 1. Heart rate deflection point. FIGURE 3. Relationship heart rate deflection and blood lactate. TABLE 1. (n 28).* Age (y) Height (cm) Weight (kg) BMI (kg/m 2 ) BC (%) Selected physical characteristics of the subjects Mean * BMI body mass index; BC body composition. SD FIGURE 2. Power output at the 4-mmol/L blood lactate level. ferrocyanide and pyruvate. Upon the application of a given voltage, ferrocyanide is oxidized, releasing electrons and creating a current. This current is measured amperometrically and is directly proportional to the lactate concentration of the blood sample. An analyzer is supplied with a check strip (to confirm that the analyzer is operating correctly) and a calibration strip that provides a nonquantitative indication of instrument accuracy (22). The deflection point of HRs for the CT and PO corresponds to the concentration in millimoles per liter for the ILT and was evaluated by graphic analyses (Figures 1 3). Statistical Analyses The mean and SD were calculated for the HR and PO at the AT. Test-retest reliability of the Conconi method was determined by the ICC. The level of statistical significance was accepted as p The method of Bland and Altman (1, 3) was also used to determine the association between the noninvasive CT and the criterion invasive method for PO and HR parameters. For this method, differences between the tests were plotted for POs and HRs at the AT separately against their mean in each subject. These differences were averaged to show mean bias. Mean bias variability was used to calculate the limits of agreement between the tests. The confidence interval (CI) was determined for the mean bias and the upper and lower limits of agreement. From the Bland and Altman formula, 95% CI, t 0.05 was the critical value for 5% 2-tailed tests; t-distribution tables were used with n 1 degree of freedom. RESULTS Thirty well-trained oarsmen volunteered to participate in this study; however, the results of 28 oarsmen were used, as 2 subjects data were excluded because of the lack of a deflection point in the CTs. Measured physical characteristics of participants are presented in Table 1. Reliability and Validity Measurements All measured data appeared to be normally distributed. The associations between test and retest measurements were evaluated by ICCs. Mean PO scores for the CT, CRT, and ILT were W, W, and W, respectively. ICC results showed that test-retest results were significantly correlated with each other with respect to PO (Table 2). Mean HR responses for the CT, CRT, and ILT were b min 1, b min 1, and b min 1, respectively. Moderate-to-high positive relationships between the tests were obtained with respect to HR responses (Table 2).

4 874 ÇELIK, KOŞAR,KORKUSUZ ET AL. TABLE 2. Interclass correlation coefficient between Conconi test, Conconi retest, and incremental load test for power output and heart rate at anaerobic threshold.* Variables (n 28) CT/CRT CT/ILT CRT/ILT Power output (W) Heart rate (b min 1 ) * p ICC interclass correlation coefficient; CT Conconi test; CRT Conconi retest; ILT incremental load test. FIGURE 6. Difference between CRT and ILT for PO plotted against their mean. CRT Conconi retest; ILT incremental load test; PO power output. FIGURE 4. Difference between CT and CRT for PO plotted PO power output. FIGURE 7. Difference between CT and CRT for HR plotted HR heart rate. The Bland and Altman plots were constructed to give a visual interpretation of the data. The level of statistical significance was accepted as the 95% CI. Mean differences against the measurements were used to test the reliability of noninvasive measurements (Figures 4 6 for PO data and Figures 7 9 for HR data). According to the criteria of Bland and Altman, the plots showed good agreement, because a very low number of data points were outside 1.96 SD of the mean of measurements. FIGURE 5. Difference between CT and ILT for PO plotted against their mean. CT Conconi test; ILT incremental load test; PO power output. DISCUSSION The purpose of this study was to assess the validity and reliability of the modified CT on rowing ergometers. The results of the present study were discussed in the framework that includes PO and HR at the AT. It was found that the modified CT is a reliable and valid test for determining the AT of elite oarsmen. The subjects of this study were 28 trained oarsmen from 5 different sport clubs. However, the selection criteria were standard for all subjects, who had to have at least 2 years of rowing ergometer experience; also, they had to be regular competitors in rowing races. Evaluated test protocol in this study was a sport-specific test for

5 THE MODIFIED CONCONI TEST FOR ROWING 875 FIGURE 8. Difference between CT and CRT for HR plotted HR heart rate. FIGURE 9. Difference between CT and CRT for HR plotted HR heart rate. rowers; thus, we included only well-trained oarsmen in our study. The validity and reliability of the CT were evaluated by 2 different parameters. These were the PO and HR at the AT. In the present study, ICC values indicated that higher reliability was achieved with either PO or HR using the CT and CRT. Also, there was a significant correlation between the ILT and CT POs at the AT. In addition, the mean HR responses at the AT among the 3 tests were correlated with each other significantly but were not as high as the PO correlations. The differences between the CT and the ILT for the POs at the ATs were plotted against their means by the Bland and Altman method. Similarly, the differences between the CT and the ILT HRs at the AT were evaluated by plots. In all measurements, the CT showed good agreement either with each other or with the criterion test. A reliability of criterion test was not conducted in the present study. The ethical committee did not approve the test-retest of this invasive method. Although this was a limitation of the study, the same protocol has been used as a criterion test in different studies several times (6, 7, 30). These results demonstrate that the measurement of PO at the AT is repeatable in CTs. Ballarin et al. (2) applied the CT to healthy children and adolescents. Tests were implemented either outdoors on a track or indoors in a gymnasium. Test and retest correlation coefficients for speed and HR at the HR deflection point and slope of the linear part of the graph for outdoor measurements were 0.99, 0.82, and 0.95, and for indoor measurements, these values were 0.99, 0.84, and The findings of the present study were in agreement with the results of the Ballarin et al. study. Therefore, application of the CT to various sports may prove a useful tool for training and experimental purposes. Another finding of this study is that PO was a more reliable criterion than HR only for the determination of exercise intensity. Reliability and reproducibility of the deflection point in the CT was criticized previously (15). An incompatible test protocol, nutritional differences between test environmental conditions, and tiredness have been suggested as problems for reproducibility (10). In the present study, care was taken with the application of the test; tests were performed on reliable rowing ergometers, and increments between stages were gradual. Participants were told to be hydrated, and strenuous exercises the day before the performance tests were prohibited. In the current study, there was no significant difference between the HR obtained at the CT and ILT. However, interclass correlation between the HR for the CT and ILT was moderate. The results of the present study showed that PO was more reliable than HR measurements at the AT. As mentioned previously, the standardization and specialization of testing procedures were an important part of the current study. All tests were conducted at similar conditions and on the same ergometers designed for training and competition. Participants of the current study were at their recovery state. Differences in the outcomes of the present study and that of Gaisl and Wiesspeiner (12) could be related to the internal validity and execution criteria. In recent years, Bourgois and Vrijens (6) examined the validity of the CT during ergometry in young rowers. Investigators compared CT results with individual AT and 4-mmol/L blood lactate levels. The PO at the individual AT was significantly different from that at the 4-mmol/L blood lactate level and at the CT AT. The HR at the individual AT was also significantly different from that of the HR at the 4-mmol/L blood lactate level and at the CT AT. Correlation coefficients between the tests for PO and HR were low or not significant. Unlike the study of Bourgois and Vrijens, the present study showed that there was a high correlation between the CT and ILT, specifically with respect to PO at AT levels. Although the design of both studies and the content of test protocols were similar, Bourgois and Vrijens examined only 10 subjects. This number was relatively low for determining the validity of a test. Experience on rowing ergometers could also have affected the test results. Subjects of the present study were senior-a or senior-b oarsmen, and they had at least 2 years of rowing

6 876 ÇELIK, KOŞAR,KORKUSUZ ET AL. experience. Another important point was the compatibility of rowing ergometers with water rowing. Concept II rowing ergometers, used in the current study, were the most widely used ergometers for training purposes (11). Performance on water rowing could be predicted accurately by means of Concept II rowing ergometers (11, 28, 29). On the other hand, Bourgois and Vrijens used MicRow ergometers, which were mechanically controlled, and the PO could only be increased manually. Those differences might have created different results between these 2 studies. Sumsion et al. (27) claimed that there was no statistically significant breakdown in HR responses in the arm crank exercise for nonathletes. Jensen and Katch (14) also demonstrated that the force developed by the active muscle was sufficient to impede local blood flow without adequate rowing experience and that it restricted the subject s measured aerobic power. Lakomy and Lakomy (17) revealed that performance test results between rowers and nonrowers on rowing ergometers had to be considered separately. In that study, the most important differences were found for the submaximal speed V O2 relationship, which showed that the rowers were more efficient on the rowing ergometers than the nonrowers. Moreover, some researchers have compared experienced rowers test performances on rowing ergometers against other ergometers. Bunc and Leso (8) investigated the effect of increasing specific (rowing ergometer) and nonspecific (cycle ergometer) POs on parameters relating to the ventilatory threshold and work efficiency. These investigators concluded that both specific muscle mass and oxidative capacity of working muscles might be increased by specific training and that, therefore, sport-specific ergometer tests present more information on sport-specific abilities than nonspecific tests. Jeukendrup et al. (15) and Hofmann et al. (13) emphasized the problems of the CT protocol. The first problem was the short duration ( 1 minute) of each stage, during which adaptation of the circular system to a certain speed or PO would be incomplete, and the HR would start to lag behind progressively. Conconi et al. (10) replied that the HR adaptation time was seconds when the increase in speed was 0.5 km h 1 or less. The second problem with this specific test was a poor correlation between the HR deflection point and the 4-mmol/ L lactate level AT. The running speed for the deflection point for the lactate threshold was correlated in 10 subjects (r 0.99) (reply to Jeukendrup et al. [15]). The relationship between the HR deflection point and the lactate threshold varied with the nutritional status of the subject. In 1996, Conconi et al. stated that test subjects should be well rested and should not have performed hard physical work prior to testing. The third problem with the CT was that the HR deflection point could not be detected for all subjects. According to Conconi et al., there would be 2 major reasons for such a problem. Physiologically, humans and other mammals increase work intensity primarily by increasing the frequency of an action or its cadence. Cycle ergometer tests at a fixed cadence fail to respect this physiologic principle. Because of individual differences, increments in exercise intensity could be adjusted according to the characteristics of the subjects. In the present study, the increases in exercise intensity for the CT were flexible and adaptable to the subject s characteristics. PRACTICAL APPLICATIONS The results of this study show that the modified CT is a reliable and valid test for determining the ATs of elite men rowers. Measured PO at the AT by CT can be used to assess the performance level of rowers and the effect of training periodically. On the basis of these results, training programs can be designed. REFERENCES 1. ATKINSON, G., AND A.M. NEVILL. Statistical methods for assessing measurement error (reliability) in variables relevant to sports medicine. Sports Med. 26: BALLARIN, E., C. BORSETTO, M. CELLINI, M. PATRACCHINI, P. VITIELLO, P.G. ZIGLIO, AND F. CONCONI. Adaptation of the Conconi Test to children and adolescents. Int. J. Sports Med. 10: BLAND, J.M., AND D.G. ALTMAN. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 327: BOMPA, O.T. Periodization Training for Sports. Champaign, IL: Human Kinetics, BOURGOIS, J., P. COOREVITS, L. DANNEELS, E. WITVROUW, D. CAMBIER, AND J. VRIJENS. Validity of the heart rate deflection point as a predictor of lactate threshold concepts during cycling. J. Strength Cond. Res. 18: BOURGOIS, J., AND J. VRIJENS. The Conconi test: A controversial concept for the determination of the anaerobic threshold in young rowers. Int. J. Sports Med. 19: BOURGOIS, J., AND J. VRIJENS. Metabolic and cardiorespiratory responses in young oarsmen during prolonged exercise tests on rowing ergometer at power output corresponding to two concepts of anaerobic threshold. Eur. J. Appl. Physiol. 77: BUNC, V., AND J. LESO. Ventilatory threshold and work efficiency during exercise on a cycle and rowing ergometer. J. Sport Sci. 11: CAREY, D. Assessment of the accuracy of the Conconi test in determining gas analysis anaerobic threshold. J. Strength Cond. Res. 16: CONCONI, F., G. GRAZZI, I.CASONI, C.GUGLIELMINI, C.BOR- SETTO, E. BALLARIN, G. MAZZONI, M. PATRACCHINI, AND F. MANFREDINI. The Conconi test: Methodology after 12 years of application. Int. J. Sports Med. 17: COSGROVE, M.J., J. WILSON, D. WATT, AND S.F. GRANT. The relationship between selected physiological variables of rowers and rowing performance as determined by a 2000 m ergometer test. J. Sport Sci. 17: GAISL, G., AND G. WIESSPEINER. A noninvasive method of determining the anaerobic threshold in children. Int. J. Sports Med. 8: HOFMANN, P., R. POKAN, S.P. VON DUVIILARD, AND P. SCHMID. The Conconi test [Letter to the editor]. Int. J. Sports Med. 18: JENSEN, R.L., AND F.I. KATCH. A new approach to rowing ergometry: Establishing exercise intensity relative to maximum force output. Eur. J. Appl. Physiol. 1: JEUKENDRUP, A.E., M.K.C. HESSELINK, H. KUIPERS, AND H.A. KEIZER. The Conconi test [Letter to the editor]. Int. J. Sports Med. 18: JÜRIMAE, J., AND T. JÜRIMAE. Responses of blood hormones to the maximal rowing ergometer test in college rowers. J. Sports Med. Phys. Fitness 41: KENNEY, W.L., AND J.L. HODGSON. Variables predictive of performance in elite middle-distance runners. Br. J. Sports Med. 19: LAKOMY, H.K.A., AND J. LAKOMY. Estimation of maximum oxygen uptake from submaximal exercise on a Concept-II rowing ergometer. J. Sports Sci. 11:

7 THE MODIFIED CONCONI TEST FOR ROWING MAHLER, D.A., B.E. ANDREA, AND J.L. WARD. Comparison of exercise performance on rowing and cycle ergometers. Res. Q. Exerc. Sport. 58: MAUD, P.J., AND C. FOSTER. Physiological Assessment of Human Fitness. Champaign, IL: Human Kinetics, MILLER, J.M., T.J. HOUSH, J.W. COBURN, J.T. CRAMER, AND G.O. JOHNSON. A proposed test for determining physical working capacity at the oxygen consumption threshold (PWC vo2 ). J. Strength Cond. Res. 18: PERRY, S.R., T.J. HOUSH, G.O. JOHNSON, K.T. EBERSOLE, AND A.J. BULL. Heart rate and rating of exertion at the physical working capacity at the heart rate threshold. J. Strength Cond. Res. 15: PYNE, D.B., T. BOSTON, D.T. MARTIN, AND A. LOGAN. Evaluation of the Lactate Pro blood lactate analyzer. Eur. J. Appl. Physiol. 82: SHEPHARD, R.J. Science and medicine of rowing: A review. J. Sports Sci. 16: SHEPHARD, R.J., AND P.-O. ASTRAND. Endurance in Sport. Secher, London: Blackwell Scientific Publications, STEINACKER, J.M. Physiological aspects of training in rowing. Int. J. Sports Med. 13: SUMSION, J.R., D.E. HANSEN, AND K.T. FRANCIS. The relationship between anaerobic threshold and heart rate linearity during arm crank measurement. J. Appl. Sport Sci. Res. 3: SUSAN, P.W., E.G. CAROL, AND G.M. THOMAS. A comparison of exercise performance on bicycle and rowing ergometers in female master recreational rowers. J. Sports Med. Phys. Fitness 35: TOPSAKAL, N. Performance and blood parameter measurements before and after exercise on water and ergometer of oarsmen. Master s thesis. Marmara University, Istanbul, Turkey, URHAUSEN, A., B. COEN, B.WEILER, AND W. KINDERMAN. Individual anaerobic threshold and maximum lactate steady state. Int. J. Sports Med. 14: WELTMAN, A., D. SNEAD, R. SEIP, R. SCHURRER, S. LEVINE, R. RUTT, T. REILLY, J. WELTMAN, AND A. ROGOL. Prediction of lactate threshold and fixed blood lactate concentrations from 3200-m running performance in male runners. Int. J. Sports Med. 8: Address correspondence to Dr. Feza Korkusuz, feza@ metu.edu.tr.