Successfully enduring a 5-set tennis match demands a sustained high level of technical, tactical, physiological and psychological capacity in often hostile environmental conditions. In terms of physiology, success is likely to depend on the player's ability to repeatedly generate power for explosive stroke production and for rapid court movement during extended matches. This paper outlines the effect of metabolic factors, and of playing in hot environments, on fatigue and tennis performance as well as physiological preparation to minimise these effects. Fitness training, nutrition and hydration are discussed and strategies are presented to help sustain physiological performance during a match and to optimise preparation for the next match.

Introduction

The margin between winning and losing a 5-set tennis match is frequently small. At the end of a match played over several hours differences in physiological performance between opponents of similar technical and tactical ability may be crucial to success. In long matches, the demand for energy to sustain work levels, imposed in a hot environment, provokes considerable physiological strain. To successfully endure tournament competition tennis players must be sufficiently prepared to meet this physiological challenge on several consecutive occasions. This paper discusses the scientific background to physiological performance in tennis and in particular the effect of metabolic factors, and of hot environments, on fatigue in tennis. Scientific findings are presented as the basis for recommendations on fitness training, nutrition and hydration to help players prepare and compete in extended matches and to recover during tournament competition. In each case scientific background is followed by a summary of specific recommendations for coaches and players.

The Scientific Background to Optimum Physiological Performance in Relation to Energy Metabolism, Hot Environments & Fatigue

High performance tennis is an extremely competitive environment. The small improvements in performance crucial to success are gained through increasingly sophisticated and specialised training and preparation techniques. For success in this atmosphere coaches and players need reliable information, based on scientific evidence, to continually improve training and preparation strategies. But in tennis the availability of direct scientific data is limited. Scientific information from other sports and studies of laboratory exercise must be interpreted and translated to the conditions of competitive tennis. This is a necessary, but imperfect process in establishing sound recommendations for training and preparation.

Seeking scientific information to establish the factors critical to limiting physiological performance is a first step in this process. These factors provide a target to focus training and preparation strategy. But before searching for the factors which might limit physiological performance, the actual nature of the performance must be established. For example, for the highly trained athlete successful endurance in a marathon involves sustaining an optimum running speed, whilst for the less trained competitor the concern is often with completing the distance (Williams & Chysanthoupolus, 1997).

What Constitutes Endurance in Tennis? Successfully enduring a 5-set tennis match demands a sustained high level of technical, tactical, physiological and psychological capacity in often hostile environmental conditions. For example, the the 1988 US Open best of five set men's singles final lasted 4.9 h (Chandler, 1995) and the 1994 Australian Open was played in 38oC (100oF) heat with centre-court surface temperature frequently 54oC (130oF; Bergeron, Armstrong, & Maresh, 1995). Under these conditions tennis players must accelerate, decelerate, change direction, move quickly, maintain balance and generate optimum stroke production repeatedly (Chandler). Aspects of physical fitness important here include cardiovascular endurance, flexibility, body composition, muscular endurance, strength and power as well as more specific factors such as acceleration, agility, balance and response time.

Physiological endurance is generally defined as the longest period that submaximal work (i.e. continuous jogging) can be sustained continuously (Sahlin, 1992). But in tennis the work level can be intense and is not constant (or continuous) so this definition is probably not relevant. In modern tennis effective stroke production involves explosive force particularly for groundstrokes and the serve (Elliott, 1999). In turn, rapid court movement is an essential response to the power developed on serve and groundstrokes by both players. These factors suggest that power output is a key physiological component in optimum tennis performance. The technical and tactical options available at any stage of a match may depend on the capacity to generate explosive bursts of power. As a result, success in the decisive rallies at the end of a long and demanding tennis match could be determined by the ability to repeatedly generate effective power output. Endurance in tennis may be defined by the ability to sustain effective power output through the 2-5 h length of a competitive match.

Endurance in Tennis Matches: Does Fatigue Limit Performance? Recently several studies have provided scientific evidence to support the common observation that fatigue impairs tennis performance through effects on stroke production and court movement. McCarthy, Thorpe, & Williams, (1997) studied the effect of a maximal tennis performance test (30 ball feeds.min-1; 4 min work, 40 s recovery) to exhaustion (35.4 ± 4.6 min) on stroke production. Groundstroke accuracy declined from the beginning to 75% of the performance test and at exhaustion. Backhand groundstroke (crosscourt) and service accuracy (right court) were lower after, compared to before, the maximal performance test. Mitchell, Cole, Grandjean, and Sobczak (1992) found that serve velocity was reduced following a 3 h competitive match. Recently a standardised testing procedure has been developed to determine the effect of fatigue on stroke production in situations specific to competitive tennis (Vergauwen, Spaepen, Lefevre, & Hespel, 1998). A strenuous standardised 2 h on-court training session (Vergauwen, Brouns, & Hespel, 1998) resulted in an increase in the percentage of errors on first serves (due mainly to errors on "wide", rather than "down the centre" serves). For groundstrokes, fatigue was most conspicuous when players were in a defensive (rather than neutral or attacking) rally situation and resulted in a 9% increase in error rate with more frequent errors on the backhand. As well as an increase in errors, the quality of successful strokes decreased with fatigue. Court movement is also affected by fatigue since the number of balls that could not be reached increased by 6% (Vergauwen, Brouns, et al., 1998). As well, the time to complete tennis pattern shuttle-runs increases with fatigue (Mitchell et al.; Vergauwen, Brouns, et al.). These results support the notion that improving physiological endurance, by minimising the effects of fatigue, is likely to benefit tennis performance. The following section outlines the scientific background to the factors which could limit physiological endurance in tennis (i.e. the capacity for repeated powerful efforts). As mentioned earlier, an appreciation of these factors will focus recommendations for training and preparation strategies.

Fatigue and Metabolism in Tennis: Several factors involving metabolism/nutrition (i.e. the supply of energy for muscle work) have been suggested as causes of fatigue: decreased muscle "immediate energy" (i.e. phosphocreatine) stores, increased muscle acidity, decreased muscle carbohydrate (i.e. glycogen) stores, a low blood glucose level and an increase in the ratio free tryptophan:branched chain amino acids in the circulation (Newsholme, Blomstrand, & Ekblom, 1992). The ratio of free tryptophan:branched chain amino acids may cause fatigue at the level of the brain (i.e. central fatigue) whilst the other factors are likely to contribute at the muscle level (i.e. peripheral fatigue; Newsholme et al.). All these factors are involved in some way with metabolism and the supply of energy for muscle work (i.e. the energy producing metabolic pathways). But the contribution of each factor to fatigue in tennis is unknown. To appreciate the potential significance of any one of these factors, it is important to determine how energy is supplied for muscle work in tennis.

The Role of Energy Producing Metabolic Pathways for Tennis - in Theory: Energy for muscle work is provided by instantaneous breakdown of the small amount of ATP present in the muscle cell. However, the muscle store of ATP can only provide energy for intense work of short duration (1-2s) and hence ATP must be continuously supplied during exercise. Supply of ATP for brief, intense exercise is provided by metabolism of: stored muscle phosphocreatine to produce creatine (the phosphocreatine system); and stored muscle glycogen to produce lactate (the glycolytic energy system). Supply of ATP in prolonged, less intense exercise is accomplished by aerobic metabolism of glycogen and fat stored in the muscle and also supplied to the muscle from the circulation. Phosphocreatine provides an instant reservoir for resupply of ATP (Newsholme, 1986). This energy system serves as the main source for ATP during single intense muscle contractions (Green, 1997). In theory these features, combined with the need for rapid movement and explosive stroke production implicate phosphocreatine as a key component of energy supply in tennis.

The Role of Energy Producing Metabolic Pathways in Tennis - in Practice: Conclusions from scientific studies concerning the role of the different energy producing metabolic pathways in singles tennis are conflicting (Christmass, Richmond, Cable, Arthur, & Hartmann, 1998). Contradictions as to the overall metabolic response (Bergeron et al., 1991; Christmass et al., 1998) and the level of lactate accumulation (Copley, 1984; Bergeron et al.; Therminarias, Dansou, Chirpaz-Oddou, Gharib, & Quirion, 1991; Christmass et al., 1998) are probably due to the lack of scientific studies on metabolism in singles tennis. There is little definite scientific information on the contribution of the energy producing metabolic pathways and the use of various fuels (phosphocreatine, carbohydrate, fat) for energy supply during a match. The potential of metabolic factors to influence fatigue in tennis must be interpreted on the basis of theory and of practical scientific information from laboratory studies, studies of related sports and the few available studies on tennis.

Metabolism and Fatigue - the Relevance of Metabolic Factors to Fatigue in Tennis: In theory, physiological performance in tennis could be limited by a decrease in power due to a fall in muscle phosphocreatine levels. Hultman and Sjoholm (1986) showed that direct stimulation of muscle (to avoid central fatigue) resulted in a rapid decline in power along with a decrease in phosphocreatine. Phosphocreatine levels return to about 50% within 60s of intense exercise resulting in near complete depletion but may take 5 min to fully recover (Soderlund & Hultman, 1991). International Tennis Federation rules allow 20s recovery between rallies and 90s between the change of ends. Generally a single rally would be unlikely to severely deplete the muscle phosphocreatine store but several consecutive intense rallies could perhaps reduce phosphocreatine levels enough to impair power output.

The conversion of muscle glycogen stores to lactate during intense exercise results in an increase in muscle (and blood) acidity. Increased muscle acidity is commonly associated with reduced muscle power, although the precise mechanism is unclear (MacLaren, Gibson, Parry-Billings, & Edwards, 1989). Generally, lactate levels and hence muscle acidity are thought to remain low during tennis (Bergeron et al., 1991; Copley, 1984) and contribute little to fatigue (Therminarias et al., 1991). However, increases in circulating lactate levels in tennis can occur (Christmass et al., 1998) and could influence fatigue in some cases (i.e. court surface, match characteristics).

The close association between fatigue and low muscle glycogen levels in prolonged continuous exercise is well established (Saltin & Karlsson, 1971). In humans the high rates of glycogen metabolism needed for power are not affected by glycogen levels when muscle glycogen stores are high (Vandenberghe, Hespel, Vanden Eynde, Lysens, & Richter, 1995). But when glycogen stores are low (i.e. 20-30 mmol.kg-1 w.w.; Costill, 1988) the supply of ATP from glycogen is reduced. This has relevance in soccer where low (9 mmol.kg-1 w.w.) muscle glycogen contents have been recorded at the end of a match, and players with low pre-match muscle glycogen stores spend more time walking and less time sprinting (Saltin, 1973). There is little direct information concerning the rate and extent of muscle glycogen depletion during singles tennis. However, carbohydrate intake has been found to improve explosive strength (Sargent jump) after 2 h of tennis (Burke & Ekblom, 1984). Low muscle glycogen levels could theoretically reduce optimum physiological performance in prolonged tennis.

In tennis, blood glucose levels do not decrease significantly (Mitchell et al., 1992) and in most cases actually increase (Christmass, Richmond, Cable, & Hartmann, 1995; Therminarias et al., 1991; Bergeron et al., 1991). Hypoglycaemia (i.e. low blood glucose) per se is therefore unlikely to be a factor in fatigue during tennis.

An increase in the ratio of (free) tryptophan:branched chain amino acids in the circulation could affect central drive and mood through changes in the level of neurotransmitters in the brain (Blomstrand, Celsing, & Newsholme, 1988). Four hours of singles tennis increased the (free) tryptophan:branched chain amino acid ratio more than 2.5-fold (Struder, Hollmann, Duperly, & Weber, 1995) suggesting that central fatigue (i.e. motivation, perception) could limit performance in long matches.

Fatigue and Dehydration in Tennis: Tennis is typically played in warm or hot environments. Many players probably compete with some deficit in the normal range of body water content (Bergeron et al., 1995). This condition is known as hypohydration (Sawka, 1992) but will be referred to by the more common term "dehydration". Few studies have systematically examined the effect of levels of dehydration on performance in tennis (Bergeron et al.). But dehydration is known to impair general exercise performance (Sawka & Pandolf, 1990). For a detailed review of fluid and electrolyte balance during tennis, and for the effects of dehydration on general exercise performance the reader is referred to Bergeron et al. (1995) and Sawka and Pandolf, respectively. Briefly, exercise performance is typically impaired by dehydration with increased effects with greater dehydration (Sawka & Pandolf) and at higher environmental temperatures (Sawka). Parameters of exercise performance potentially important to tennis such as muscular endurance, maximal aerobic power and physical work capacity are all adversely affected by dehydration (Sawka & Pandolf). Dehydration may also affect mental performance in tennis since short-term memory, arithmetic and motor speed and attention are all impaired in proportion to the degree of dehydration (Gopinathan, Pichan, & Sharma, 1988). Importantly, only a small degree of dehydration (1-2% body weight) may be needed to affect physical (Craig & Cummings, 1966) and mental performance (Gopinathan et al.).

Recommendations for Optimum Physiological Performance

The scientific background to physiological performance in tennis and the limitations to this performance provide a framework and focus for strategies to optimise pre-match preparation, competition and recovery. But in the same way it is important for specific recommendations and strategies to also be based on scientific evidence. Again, the lack of direct scientific information (metabolism/physiology) on tennis necessitates interpretation of data from laboratory studies and related activities. The following section outlines recommendations for preparation, competition and recovery for improved tennis performance. The scientific background to the strategies is outlined and is followed by a summary of practical recommendations.

Pre-match Preparation - Fitness Training and Acclimation for Long Matches in Hot Conditions: In terms of physiological fitness tennis players must be sufficiently powerful, and capable of rapid recovery between rallies. Acclimatising to hot environments may also benefit performance.

The Phosphocreatine Energy System: Since powerful stroke production and explosive movement are important factors in modern tennis (Elliott, 1999) a considerable proportion of fitness training should be focussed on developing the phosphocreatine energy system. The net use of phosphocreatine may be quantitatively small (Bangsbo, 1994), but it is important for producing and repeating rapid, powerful movements (Newsholme et al., 1992).

The Aerobic Energy System: Once power has been developed, the capacity to recover and to repeatedly generate power (i.e. endurance in tennis) is most likely to depend on high levels of aerobic fitness.

Recovery of phosphocreatine: Recovery of phosphocreatine stores after exercise stops when oxygen supply is blocked (Sahlin, Harris, & Hultman, 1979). The availability of oxygen is therefore probably a limiting factor for recovery of phosphocreatine levels (Sahlin et al., 1979). Since tennis players have a finite recovery time between rallies, effective oxygen delivery to muscle due to a high aerobic fitness level is likely to improve the ability to recover and repeat powerful efforts.

Lactic acid accumulation and muscle acidity: As mentioned earlier, lactic acid accumulation (and muscle acidity) are in most cases unlikely to contribute to fatigue in tennis (Therminarias et al., 1991). However, increased lactate levels result from longer work period durations in intermittent type exercise (Ballor & Volovsek, 1992) and also tend to occur in tennis players with more intense patterns of footwork (Christmass et al., 1998; i.e. take more steps per time). A combination of these effects (i.e. matches on slow surfaces such as clay) could conceivably result in an increase in lactate accumulation (and muscle acidity). Of importance here is that high aerobic fitness levels (and hence effective blood supply to muscle) enable more efficient buffering of muscle acidity (Newsholme et al.,1992) minimising possible effects on performance. Certainly, higher aerobic capacity is associated with greater recovery of muscle force as well as higher phosphocreatine and lower lactate levels after repeated bouts of heavy exercise (Jansson, Dudley, Norman, & Tesch, 1990).

Preserving muscle glycogen stores: In endurance exercise a high level of aerobic fitness results in an increased reliance on fat as a fuel which will help to spare glycogen reserves (Holloszy, 1990). The situation is less clear for intermittent exercise such as tennis. Higher rates of carbohydrate use occur in the presence of low muscle oxygen availability when work periods (i.e. rallies) are longer during intermittent treadmill exercise (Christmass, Dawson, & Arthur, 1999). More research is needed to determine whether increased aerobic fitness can spare muscle glycogen stores in long tennis matches.

Tolerating exercise in hot conditions: A high level of aerobic fitness and physical training in a cool environment improve physiological responses to exercise in a hot environment (i.e. a "partial" heat acclimation; Piwonka & Robinson, 1967; Pandolf, 1979). Therefore, increased aerobic fitness may benefit tennis performance in hot conditions when a period of acclimatisation is not possible.

Acclimatising for Tennis in Hot Conditions: The following recommendations are based on the reviews by Terrados and Maughan (1995), Sawka (1992) and Murray (1992). The adverse effects on physiological and performance variables as result of exercise in hot environments can be reduced by regular exposure (i.e. acclimatising) to the conditions (Terrados & Maughan). The principal adaptations are largely complete within 6-8 days of exposure to the new conditions (Armstrong & Maresh, 1991) including daily training or competition (Wenger, 1988).

For specific details of an overall conditioning program for tennis the reader is referred to Chandler (1995).

Pre-match Preparation - Nutrition and Hydration for Prolonged Matches in Hot Conditions: Maintaining an adequate diet in general as well as appropriate nutrient and fluid intake in the days prior to important matches is likely to benefit tennis performance.

The Tennis Player's Diet: Clearly sustaining physiological performance in tennis will be improved by high levels of physical fitness for the reasons outlined above. Such preparation generally requires large volumes of on- and off-court training for which an adequate diet is essential (Coyle, 1991). For a review of athlete nutrition and sports performance the reader is referred to C. Williams (1995) and Clarkson (1996). Briefly, a diet consisting of a wide range of foods in which 60-70% of total energy intake is derived from carbohydrate, 12% from protein and the remainder from fat is optimal for most sports (Devlin & Williams, 1991). A high level of carbohydrate intake is recommended on the basis of the demand placed on glycogen stores during training and competition and the relationship between depletion of these stores and fatigue (Costill & Hargreaves, 1992). Recommended dietary protein requirements for most athletes can be achieved within a balanced diet (Lemon, 1991a). Reducing fat intake to extremely low levels is not recommended on the basis that some lipids (i.e. essential fatty acids) are important to general health (C. Williams, 1995). Alternatively, consuming excess fat reduces the opportunity to maintain the recommended carbohydrate intake (C. Williams, 1995) and may have detrimental long-term effects on health (Clarkson, 1996).

Should Tennis Players Carbohydrate Load? The positive effect of a high carbohydrate diet in the days prior to competition on endurance in prolonged submaximal exercise is well established (Karlsson & Saltin, 1971). This effect is also apparent during brief high intensity continuous exercise (Maughan et al., 1997) as well as during intermittent exercise (Bangsbo, Norregaard, & Thorsoe, 1992). A gradual decrease in the volume of training in the week prior to a tournament combined with an increase in carbohydrate content of the diet (i.e. 3 days on a mixed diet containing 50% carbohydrate, followed by 3 days on a high carbohydrate diet containing 70% carbohydrate) has been recommended as an effective carbohydrate loading regime (Sherman, Costill, Fink, & Miller, 1981). The practice of dietary carbohydrate loading is relatively common for endurance events lasting more than 1-2 h and could be of benefit for tennis tournaments involving 5-set (i.e. long) matches. On the other hand, increases in muscle water content with carbohydrate loading can cause feelings of heaviness and stiffness (Fox & Mathews, 1981) which may impair tennis performance in some cases. There is little specific information to support or refute the benefits of carbohydrate loading in tennis. Short-term high fat diets have been suggested to improve endurance performance, however this has not been definitively confirmed and in general results indicate a possible detrimental effect on performance (Sherman & Leenders, 1995).

Pre-match Nutrition: The timing and composition of nutrient intake on the day of competition will also affect performance in prolonged exercise (Wright, Sherman, & Dernback, 1991). For sports in which the level of glycogen stores may impair performance ingestion of a high carbohydrate meal (200-300 g) 3-4 h before exercise is generally recommended (Coyle, 1991). High carbohydrate meals are usually easily and rapidly digested and absorbed compared to meals high in fat and protein (Williams & Chyssanthopoulos, 1997). In this context, a high fat meal 3-4 h before exercise should be avoided because digestion is slow which increases the potential for gastrointestinal distress (Williams & Chyssanthopoulos).

Carbohydrate Snacks: Irregular scheduling of matches during tournament tennis competition often makes it difficult to precisely time the pre-match meal. In this situation, supplementary carbohydrate snacks may be necessary. In the past, carbohydrate consumed in the hour prior to exercise has been suggested to reduce endurance capacity (Foster, Costill, & Fink, 1979), increase muscle glycogen depletion and cause a fall in blood glucose at the beginning of exercise (Costill et al., 1977). However, since these early findings several studies have found carbohydrate ingested in the hour before exercise does not result in increased muscle glycogen use and has either no effect, or improves, endurance performance (for review see Coyle, 1991). On this basis, Coyle (1991) concluded there was minimal support for the concept that carbohydrate ingestion in the hour prior to exercise impairs endurance performance. However, on the basis of possible differences between individuals in metabolic response to pre-exercise carbohydrate ingestion, experimentation with variations in pre-exercise carbohydrate intake during training is recommended (Costill & Hargreaves, 1992). In terms of the effects on blood glucose, the decrease at the onset of exercise may not reach levels associated with hypoglycaemia (2.5 mM; Chryssanthopoulos, Hennessy, & Williams, 1994) and in any case is frequently not perceived by athletes (Ferrauti, Weber, & Struder, 1997; Williams & Chyssanthopoulos, 1997) and does not cause muscle weakness (Coyle, 1991). Blood glucose concentration decreases significantly at the beginning of exercise in tennis with (Ferrauti et al., 1997) or without (Bergeron et al., 1991) carbohydrate ingestion immediately before activity. Despite the points outlined above Ferrauti et al. noted that some tennis players may perceive this response adversely (i.e. complain of hypoglycaemic symptoms) and suggested that players perform an extensive warmup so that any potential fall in blood glucose occurs in the pre-match period.

Pre-match Hydration: Players should consume 300-500 ml of water before a 5-set (i.e. long) match on the basis of an exercise duration of 1-3 h and provided a carbohydrate-electrolyte drink is consumed

during the match (Gisolfi & Duchman, 1992; see below).

During the Match - Nutrition and Hydration: Carbohydrate supplementation and fluid replacement during exercise are important in delaying fatigue (Maughan & Shirreffs, 1997). This is also likely to be the case for tennis players.

The Effect of Carbohydrate-Electrolyte Drinks on Tennis Performance: Consumption of carbohydrate-electrolyte drinks during competition will contribute to delaying fatigue due to dehydration but may also improve performance (Williams & Chryssanthopoulos, 1997). For example, in endurance exercise supplementation with carbohydrate can delay fatigue (Coyle et al., 1983) and may improve sprint performance at the end of prolonged exercise (Hargreaves, Costill, Coggan, Fink, & Nishibata, 1984). The effect of a carbohydrate beverage on tennis performance has been investigated although with contradictory findings (Burke & Ekblom, 1984; Mitchell et al., 1992; McCarthy, Thorpe, & Williams, 1995; Ferrauti et al., 1997; Vergauwen, Brouns, et al., 1998). Ferrauti et al. (1997) and Burke and Ekblom (1984) showed an improvement in court speed following supplementation with a carbohydrate drink in prolonged (4 h and 2 h, respectively) tennis matches although Mitchell et al. and McCarthy et al. (1995) found no effect. Ferrauti et al. (1997) concluded that the absence of an effect in these latter studies was due to too few subjects (McCarthy et al., 1995) and an inadequate testing procedure (Mitchell et al.). In terms of stroke production, several studies have found no effect of carbohydrate beverages (Ferrauti et al.; McCarthy et al., 1995; Mitchell et al.). However, Vergauwen, Brouns, et al. (1998) tested stroke production in specific situations relevant to match conditions. For example, comparisons were made with "down the centre" and "wide" serves and groundstrokes were compared across neutral, defensive and offensive rally situations (Vergauwen, Brouns, et al., 1998). In general this study showed that carbohydrate supplementation enabled players to produce more powerful and more accurate stroke production as well as to reduce error rate. On the basis of findings for prolonged continuous and intermittent exercise, and for tennis, it is possible that performance will benefit from consumption of a carbohydrate-electrolyte drink during 5-set tennis matches. However, the precise composition of a suitable carbohydrate-electrolyte drink is dependent on several factors (see review by Gisolfi & Duchman, [1992]).

Composition of Carbohydrate-Electrolyte Drinks: On the basis of a 5-set (i.e. prolonged) singles match, tennis players should follow the guidelines for fluid replacement for exercise durations from 1-3 h outlined by Gisolfi and Duchman (1992). Briefly, 800-1600 ml.hr-1 of a 6-8 % carbohydrate solution (5-15 oC) including 10-20 mM of sodium is recommended (Gisolfi, & Duchman). The rationale for the carbohydrate content has been reviewed previously (Lamb & Brodowicz, 1986) and this level satisfies the recommendations from Coyle (1991). The suggested volume range is large and some players may experience gastrointestinal discomfort with a volume more than 1.25 l.hr-1 (Coyle, & Montain, 1992). Precise volumes will depend on sweat rate which will vary between individuals (i.e. large males, small females) and with differences in environmental conditions.

During the Match - Physiological Considerations for Strategy and Tactics: International Tennis Federation rules provide for maximum standard recovery in tennis (see above). The recovery of muscle metabolism is partly a time-dependent process (Harris et al., 1976). On this basis it can be assumed that optimum physiological performance will benefit from securing complete advantage of available recovery time. Again, the finding that higher lactate levels occur with longer work periods in intermittent exercise (Ballor & Volovsek, 1992) and with more intense footwork patterns in tennis (Christmass et al., 1998), provides some support for the perception that players may modify match strategy to "preserve energy" late in 5-set (i.e. long) matches. Specifically, match strategy to limit the length of rallies is understandable from a perspective of maintaining physiological performance. Players with intense footwork patterns may also benefit from tactics to modify this characteristic at particular stages in prolonged matches. For example, in baseline rallies, limit footwork recovery to a court position related to the opponent's specific technical/tactical options rather than repeated recovery to the central baseline position.

Between Matches - Recovery for Repeated Performance: Success in a tennis tournament involves winning several consecutive matches frequently separated by less than 24 h. Tennis players must develop effective strategies for post-exercise recovery so that high level performance can be repeated. In addition to restoration of fluid and electrolyte balance and energy stores, post-exercise recovery may include the use of massage (Cafarelli & Flint, 1992), hydrotherapies (i.e. float tanks, contrast temperature baths), sauna and decompression chambers as well as electrical physiotherapy modalities (Calder, 1990). However, there is relatively little scientific data concerning the effectiveness of most of these techniques with the exception of nutritional strategies for recovery (Burke, 1996; Maughan & Shirreffs, 1997). For a review of general processes available for recovery, and specifically the role of massage, the reader is referred to reviews by Calder and Cafarelli and Flint, respectively. The current paper will focus on recommendations for fluid and nutrient recovery in the post-exercise period.

Nutrient Recovery: Nutritional recovery includes replenishment of glycogen stores which may be depleted in prolonged matches (Bergeron et al., 1995), restoration of fluid and electrolyte balance, and repair of cell/tissue damage (Burke, 1996). The relationship between glycogen stores, carbohydrate supplementation and performance (see above) emphasises the importance of replacing glycogen stores in the post-exercise period. On average 20-24 h are required to restore normal muscle glycogen levels following exhaustive exercise (Coyle, 1991). Costill and Miller (1980) originally hypothesised that inadequate carbohydrate intake during repeated days of exercise would result in gradual loss of glycogen stores and reduced endurance performance. Although this hypothesis has been the subject of debate, current nutritional recommendations promote a high carbohydrate diet during periods of prolonged competition to support maximal recovery of glycogen levels (Burke). Coyle has suggested that the rate of glycogen storage is generally optimised by consuming 50 g of glucose (carbohydrate with a high glycaemic index) every 2 h in the early stages of recovery (total > 600 g for a 70 kg body mass). The rate of glycogen storage is more rapid in the initial 2 h post-exercise period (Ivy, Katz, Cutler, Sherman, & Coyle, 1988b) but this is unlikely to be physiologically significant (Burke). Instead the recommendation that carbohydrate intake commence as soon as practical following a match is based on maximising available recovery time for glycogen storage (Coyle). Provided total carbohydrate intake is adequate, muscle glycogen storage is apparently not influenced by the frequency and size of carbohydrate meals (Costill et al., 1981; Burke et al., 1996). Therefore, in cases where consumption of carbohydrate every 2 h in the time between matches is impractical (i.e. sleep) a low fat, low protein meal containing sufficient carbohydrate for the period without food will suffice (i.e. 50 g.2h-1, Coyle). Moderate to high glycaemic index foods should form the basis of carbohydrate intake post-exercise (Coyle). Since appetite is often lower in the period immediately post-exercise, carbohydrate intake in the form of fluids may be useful at this time (Keizer, Kuipers, van Kranenburg, & Geurten, 1986).

Fluid-Electrolyte Recovery: Recent research indicates that the provision of fluids is likely to be as important as carbohydrate for maintaining exercise performance (Below, Mora-Rodriguez, Gonzalez-Alonso, & Coyle, 1995). A planned program of fluid intake is important (particularly when recovery is limited and dehydration may be ~ 2-5% body mass; Burke, 1996) because of the combined effects of failure to rehydrate when fluids are readily available (i.e. "involuntary dehydration"; Nadel, Mack, & Nose, 1990) and continued loss of fluid through sweating and urination after exercise (Burke). Generally, the volume of fluid consumed for rehydration must be larger (i.e. 1.5-2.0 fold) than the volume lost in exercise to allow for fluid lost as sweat and urine in post-exercise recovery (Shirreffs, Taylor, Leiper, & Maughan, 1996). However, fluids with a low sodium content (i.e. water) will cause high urinary losses so that even consumption of large volumes (i.e. 2.0-times the exercise sweat loss) will not sustain fluid balance in recovery (Shirreffs et al., 1996). Rehydration drinks should generally contain moderately high levels of sodium (50 mM) and perhaps some potassium (Maughan & Shirreffs, 1997). For optimum rehydration, when sodium is the sole electrolyte source consumed, increasing the sodium content of rehydration fluids above that of commonly available drinks (i.e. GatoradeTM 20 mM; Gonzalez-Alonso, Heaps, & Coyle, 1992) may be beneficial (Maughan & Leiper, 1995). Finally, water may suffice for rehydration when solid food, consumed during recovery, adequately replaces electrolytes (Maughan & Shirreffs).

Creatine Supplementation: At present creatine does not appear to have adverse effects and is not banned by the IOC (Clarkson, 1996). Oral creatine supplementation can increase muscle levels of creatine and phosphocreatine (Harris, Soderlund, & Hultman, 1992) as well as increasing the recovery of phosphocreatine after intense exercise (Greenhaff, Bodin, Soderlund, & Hultman, 1994) which would be expected to improve and help maintain power output (see above). Balsom, Ekblom, Soderlund, Sjodin, and Hultman (1993a) showed that creatine supplementation improved performance during brief, high-intensity, intermittent cycling. In general, research supports an ergogenic effect of creatine supplementation in high-intensity intermittent exercise (M.H. Williams, 1995). Dietary creatine supplementation has been shown to improve performance in prolonged (80 min) intermittent, intense exercise (Preen et al., 1997) and therefore might improve tennis performance in long matches. Presently, there is no evidence of an effect in tennis and research is also needed to determine whether effective creatine supplementation regimes can be developed for the specific conditions of tournament tennis.

Caffeine Supplementation: Caffeine is a stimulant drug and is restricted by the IOC (Clarkson, 1996). Research suggests that long duration aerobic performance is improved by caffeine supplementation at levels below current IOC regulations (Graham & Spriet, 1991). Caffeine supplementation in low doses has been shown to improve visual choice reaction time (Lieberman, Wurtman, Emde, Roberts, & Coviella, 1987) which might be expected to contribute to improved tennis performance. In the study by Vergauwen, Brouns, et al. (1998) carbohydrate compared to placebo supplementation improved tennis performance (see above). However compared to carbohydrate only, carbohydrate plus caffeine supplementation did not alter the decline in stroke quality observed at fatigue or add to the ergogenic effect of carbohydrate for shuttle run performance (Vergauwen, Brouns, et al., 1998). Ferrauti et al. (1997) studied the effect of supplementation with carbohydrate and caffeine separately on tennis performance. In this study post-exercise urinary caffeine levels were below IOC limits and there was no effect of caffeine on shuttle run performance (Ferrauti et al.). Caffeine had no effect on stroke production or playing success in male subjects although performance improved in female players (Ferrauti et al.). At present, caffeine supplementation appears to have minimal ergogenic benefit for tennis.

Clarkson (1996) and M.H. Williams (1995) have presented general reviews of the numerous dietary supplements suggested as possible ergogenic aids for athletes. In most cases results are equivocal as to an ergogenic effect and further research is needed (Clarkson; M.H. Williams, 1995). The previous discussion is presented with the intention of directing coaches to current scientific opinion in this controversial field rather than an advocacy for nutritional ergogenic supplementation. The reader is referred to M.H. Williams (1994) for discussion of the ethical issues involved in the use of nutritional ergogenic aids in sport.

Summary & Conclusion

Fatigue impairs tennis stroke production and court movement and hence improving physiological performance is likely to benefit players. In terms of physiological performance, successfully enduring a 5-set (i.e. long) tennis match is likely to depend in part on the capacity to repeatedly generate power across several hours of play. Fitness training for tennis should focus on developing the phosphocreatine energy system (i.e. various forms of training for power). But a high level of aerobic fitness is also crucial to facilitate recovery between rallies and to a lesser extent to improve tolerance of hot conditions in non-acclimated players. Hot environments and dehydration impair physical and mental performance variables relevant to tennis. Where necessary, players should acclimatise by training daily for 6-8 days in the hot environment. Pre-match dietary carbohydrate intake to maintain glycogen stores is important but dehydration may cause fatigue before glycogen stores are depleted. Carbohydrate drinks during competition may improve physiological performance and will also contribute to delaying fatigue due to dehydration. Recent evidence indicates that carbohydrate-electrolyte drinks may improve tennis-specific performance. For long matches (3 h) players should consume 800-1600 ml.h-1 of a 6-8% carbohydrate solution containing 10-20 mM sodium (this volume range is large and specific intake will depend on individual sweat rate and environmental conditions). For effective post-match recovery players should begin consumption of carbohydrate (50 g.2 h-1, total > 600 g for 70 kg body mass) and fluid (sodium content 50 mM; 1.5-2.0-fold the exercise fluid loss) as soon as possible after the match.