The phenomenon known as the lactate paradox has been a topic of heated debate since it gained worldwide attention following Operation Everest in the early 1980s. What began as the simple finding that blood lactate (blood [La]) for a given sub-maximal workload or VO2 following acclimatization to high altitude is reduced compared with sea-level values, morphed into a complex set of parameters that have been redefined several times in the nearly 30 years that the lactate paradox has been researched. Though several strong hypotheses have been proposed to, to date, no one hypothesis has been able fully to explain the lactate paradox. The goal of the current article was to bring together the most prominent studies done on the lactate paradox and illuminate the details brought forth by each. In doing so we hope to stimulate new hypotheses and research studies that will further our understanding of the lactate paradox.

Discussions on growing more durable equine athletes often revolve around developmental orthopaedic disease (DOD). What degree of DOD, if any, is acceptable or inconsequential to long-term athletic performance remains a question. Much effort has been directed toward diminishing or eliminating the incidence and severity of DOD, but with limited success. We must continue to define what degree the management decisions affect the horse during different stages of its growth. It would seem logical that the most profound influence would occur during the most rapid periods of growth. Therefore, concerns have been directed toward the last third of gestation and the possible indirect effect that broodmare management has on foetal development. The early postnatal period is another period of rapid growth; thus, the effect of broodmare management on lactation may affect the growth of the foal. Management will also have a direct effect on growth and development from weaning through to maturity. Recent research in other species as well as in horse suggests that genetic predisposition is the primary determinant of DOD. Non-nutritional management also may have significant effects on articular and skeletal health and athletic durability. Much effort has been directed toward nutrition and potential impact on DOD, but with limited success. Except for overt deficiencies, or perhaps excessive imbalances, nutrition does not appear to be a primary determinant of DOD, particularly of focal lesions associated with osteochondrosis. However, nutrition may serve a secondary role in diminishing the severity and incidence of DOD in horses genetically predisposed to such conditions. Producing horses that will have long-term and productive athletic careers requires sound management that starts at conception and continues throughout the life of the horse.

This study tested the hypothesis that phenylbutazone would block the exercise-induced increase in cytokine markers of inflammation in blood. Blood samples were obtained from unfit Standardbred mares (age 10 ± 4 years, ~500 kg) before and after three different trials (standing control (CON), n = 9; exercise with phenylbutazone (EX-bute), n = 9; and exercise with water, n = 9). Comparisons were made for data collected in three trials, one where each horse underwent an incremental exercise test (graded exercise test (GXT)) where they were administered water as a placebo, a GXT following phenylbutazone administration (2 g given orally 2 h before the GXT) or standing parallel control where they stood quietly in stalls. During the GXT, horses ran on a treadmill (1 m s− 1 increases each min until fatigue, 6% grade). Blood samples were obtained 30 min before exercise, immediately after exercise and at 0.5, 1, 2, 4 and 24 h post-GXT or at matched time points during the parallel control trials. Samples were analysed using real-time PCR for measurement of mRNA expression of interferon-γ (IFN-γ), tumour necrosis factor-α (TNF-α) and interleukin (IL)-6 in samples collected during all three trials, and for IL-1 and IL-10 in samples collected for the CON and EX-bute trials. Data were analysed using ANOVA for repeated measures, and where appropriate, post hoc separation of means utilized the Student–Newman–Keuls test. The null hypothesis was rejected when P < 0.05. There were no changes (P>0.05) in IL-1, IL-6, IFN-γ or TNF-α during CON or following phenylbutazone administration. During the water trial, exercise resulted in significant increases in IFN-γ, IL-1 and TNF-α. It was concluded that high-intensity exercise results in a transient increase in the expression of inflammatory cytokines in blood that is blocked when phenylbutazone is administered to horses.

The exchange of water between plasma and interstitial fluid (ISF) along the length of a capillary is attributed to a balancing of the Starling forces, site-specific differences in hydrostatic and osmotic pressures that theoretically determine directional fluid movement. The osmotic forces for water movement are attributed to the osmotic effects of proteins, colloid osmotic pressure (COP). Several physiological inconsistencies question the role of proteins and COP in fluid flux. A reconsideration of Hulett's insights concerning the osmosis of water provides substantial evidence that the effect of COP does not cause osmosis, and therefore another force is needed to explain plasma–ISF exchange. Review of whole-body tissue and blood ion concentrations and/or ion differences across isolated tissue or secretory epithelia from a variety of species indicates that the diffusion of bicarbonate and strong ions within plasma is the dominant osmotic effect returning ISF to the capillary. Conceptually, as these ions diffuse along physiological gradients, they alter the chemical potential of water through which they are diffusing (solute–solvent drag), creating an osmotic effect on plasma water, and explain plasma–ISF exchange. Considering venous–arterial differences, diffusing and strong ions give rise to a net osmotic force (~35 Torr) in venous end capillary plasma water that is coupled to ISF through pores in the endothelium. More importantly, diffusing and strong ions provide an incremental osmotic force (~150 Torr) that is essentially matched to any change in metabolic rate (e.g. muscular work) when CO2 output and water production are increased. The proposed diffusing ion osmotic force does not negate the necessity for colloidal proteins in volume regulation. Proteins can have an essential effect on fluid exchange in plasma when blood flow is intermittent or changes in protein concentration in the ISF such that proteins exert a force against a distensible boundary (i.e. the endothelium and basement membrane) as they are reflected by it or diffuse through the membrane due to changes in permeability.

The purpose of the present study was to investigate the effect of the horse's laterality on the symmetry of rein tension in right-handed riders. Eleven right-handed riders rode both a right-lateralized (RL) and a left-lateralized (LL) horse. Rein tension was measured during three circles of walk, trot and canter and four walk–halt transitions in each direction. Tensions were recorded continuously using a rein tension meter. The LL horse was ridden with significantly stronger mean tension in the left rein than in the right rein (1.5 vs. 1.4 kg; P = 0.0352). Significantly more tension was applied to the outside rein in a clockwise (1.4 vs. 1.2 kg; P = 0.0202), but not in a counterclockwise, direction (1.3 vs. 1.2 kg; P = 0.49). Less minimum tension (0.06 vs. 0.29 kg) and greater maximum (6.4 vs. 3.9 kg) and range of tension (6.3 vs. 3.6 kg) occurred in the left rein of the RL horse (P < 0.0001) and the right rein of the LL horse (0.37 vs. 0.08 kg, 4.8 vs. 7.4 kg, 4.3 vs. 7.3 kg respectively; P < 0.0001). The results of the present study indicate that the different utilization of both reins is likely to be influenced by the laterality of both horse and rider. These findings may have important implications for equine training, since consistency of reinforcement is an important factor for equine learning success.

Successful completion of endurance exercise is facilitated by conditioning that enhances the capacity of the cardiovascular and musculoskeletal systems to maintain a high level of aerobic activity. Insufficiency or disease in these systems may be diagnosed using routine haematology and serum chemistry testing. Pre-race blood samples obtained from 82 endurance-trained sled dogs were used to test the hypothesis that common haematological and serum chemistry values could identify subjects with occult disease, decreasing the likelihood of their finishing a multi-day exercise challenge. Multivariate logistic regression was used to identify the combinations of specific biomarkers that best predicted the eventual outcome of the exercise challenge for an individual dog. This analysis constructed six different predictive formulas. In all cases, the formulae contained the pre-race values for creatine kinase and some measurement of erythrocyte abundance. When a probability of 0.4 was used as a cut-off value, all dogs with values below the cut-off did not complete the race. These results suggest that in a population of healthy, aerobically conditioned subjects, routine clinical biomarkers can be used to identify subjects unsuited to performance for endurance exercise.

The fundamental tenets of exercise physiology are to describe energy transformations during physical work and make predictions about physical performance under different conditions. Historically, the most popular method to observe such responses during exercise has been the constant load or fixed-intensity protocol, based largely on the assumption that there is a certain threshold response of the organism under a given condition. However, constant load exercise does not fully allow for randomness or variability, as the biological system is overridden by a predetermined externally imposed load that cannot be altered. Conversely, in self-regulated (paced) exercise, there is almost an immediate reduction in power output and muscle recruitment upon commencing exercise. This observation suggests the existence of neural inhibitory command processes. This difference in regulation demonstrates the inherent importance of variability in the biological system; for in tightly controlled energy expenditure, as is the case during constant load exercise, sensory cues cannot be fully integrated to provide a more appropriate response to the given task. The collective evidence from conventional constant load versus self-regulated exercise studies suggests that energy transformations are indeed different, so that the inherent biological variability accounts for the different results achieved by the two experimental paradigms.

Twenty-seven mature Standardbred mares (9–27 years, ~522 kg) were used to test the hypothesis that training and short-term cessation of training would alter total plasma total carbon dioxide (tCO2) concentrations. Plasma tCO2 concentrations were measured in blood samples (20 ml) collected at rest from the same ten unfit mares that were used in two separate studies that were conducted 1 month apart. Comparisons between the samples obtained from the two trials were made to demonstrate the consistency of plasma tCO2 concentrations in the untrained population. Another set of resting blood samples was collected from an additional 17 mares that were being utilized in a training study that was under way during the same period. All the mares were housed in groups on dry lots, and were fed approximately 12 kg of mixed alfalfa-grass hay divided into two feedings per 24 h period. During the ‘detraining’ period, the 17 horses were housed in 5 × 5 m stalls, and were fed the same hay ration. Water and trace-mineral blocks were available ad libitum. Blood samples were collected from the mares prior to the last session of their 12 weeks of training at 60% HRmax (maximum heart rate), as well as on the third day following 2 days of detraining (simulated quarantine). Plasma tCO2 concentrations were measured in duplicate using a Beckman EL-ISE analyser. For data analysis, t-tests with the a priori level of statistical significance set at P < 0.05 were used. Resting plasma tCO2 concentrations were lower (P < 0.05) in the trained horses (31.4 ± 1.9 mMol l− 1) than in the untrained horses (34.4 ± 0.9 mMol l− 1). There were no effects (P>0.05) due to cessation of training. It was concluded that training causes a decrease in plasma tCO2 concentrations that is not reversed by limited cessation of training.

Foot strike can be difficult to determine using kinematics alone, particularly when studying equine activities on more compliant surfaces, so this study was done with the aim of developing and validating a method to determine foot strike on an arena surface that can be used in conjunction with kinematics alone, and of applying the method in the context of measuring foot strike during jump landing on an arena surface. A low-cost contact mat was developed. The timing of the contact mat switching ‘on’ was compared to the timing of a force platform onset of 20 N, load and loading rate at foot strike. Two groups of 25 participants were used in two separate studies to validate the contact mat: the first measured the difference in timing with respect to two different activities (running and stepping down from a box), and the second measured the difference in timing with respect to 1- and 2-cm depths of an arena surface during running. In a third study, the mat was used to measure leading limb foot strike of six horses during jump landing, and these data were compared to kinematics from a palmar marker on the hoof wall. All data were recorded at 500 Hz. A consistent difference in delay was found between the mat and force platform onset, and as a result, no significant differences (P>0.05) in timing delay between different loading rates or depths were found. During jump landing, foot strike (determined from the mat) occurred after the vertical velocity minima and the acceleration maxima for the hoof marker, but it occurred before the point where the rate of vertical displacement began to reduce. In conclusion, further work is needed to enhance these techniques, but these preliminary results indicate that this method may be effective in determining foot strike for field-based applications.

The impact of body density (BD) on an animal's capacity to sustain a workload is important in the accuracy of evaluating swimming exercise in rats and the associated training protocols. The aim of this study was to quantify the influence of BD the maximal lactate steady-state (MLSS) workload in swimming rats. The BD of ten 90-day-old rats and sixteen 120-day-old rats was determined by underwater weighing, and their aerobic capacity was determined by the MLSS test. The MLSS blood concentration values were 4.11 ± 0.96 mmol l− 1 in the 90-day-old rats and 4.81 ± 1.49 mmol l− 1 in the 120-day-old rats. There was no significant (P>0.05) difference between these values. The older rats were more dense (P < 0.001) and showed a significantly increased (P < 0.001) absolute effort and relative-to-body-weight effort to keep themselves on the water surface when compared with the younger rats. BD can significantly affect an animal's capacity to sustain work within this age range, and individual fluctuation in effort should be determined to avoid mistakes when interpreting the results. This is particularly important in longitudinal studies in which the intervention or ageing process can modify the animal's body composition. Our results quantify the effects of BD on the performance of rats in the MLSS test.

The primary purpose of this study was to test the predictions of the dual-mode model using arm ergometry as the exercise modality. It was hypothesized that changes in affect during exercise would be greater in high (105% VT)- and low (80% VT)-intensity exercise conditions than in a control condition, while differences in affect would be observed between exercise conditions. Study participants were 24 physically active men. A within-subjects design was used. Affect was measured using the Activation–Deactivation Adjective Check List and the State Anxiety Inventory. Cognitive (i.e. self-efficacy and enjoyment) and physiological (i.e. heart rate, pain and perceived exertion) mediators of the exercise–affect relationship were also examined. Results showed that during exercise, changes in affect were greater in the exercise conditions than the control condition, and affective valence in the exercise conditions declined relative to the control condition. In partial support of the dual-mode model, self-efficacy mediated the relationship between below-VT exercise and affect, whereas pain mediated the relationship between above-VT exercise and affect. These findings highlight the importance of exercise intensity to the affective benefits of exercise.

The assessment of the effect of race training on the locomotory apparatus can be confounded by tissue responses to the increasing training load, such as dorsal metacarpal disease. The aim of this paper was to examine the changes in gait during the early stage of race training in a cohort of 19.2-year-old Thoroughbreds, 11 of which had previously been exposed to spontaneous pasture exercise and also to preconditioning exercise from an early age (CONDEX), and the remainder had been exposed to only spontaneous pasture exercise (PASTEX). Data were collected 52, 60, 77, 109 and 176 days after the horses had entered the racing stable. At each examination, the horses were trotted in-hand on a hard asphalt surface, and kinematic data were collected. The horses trotted slowest on the first examination day, with a significantly longer and slower stride and associated percentage stance time. On day 77, the CONDEX horses trotted faster, had a longer stride length and achieved mid-stance earlier than the PASTEX horses. Across both groups, the introduction of galloping was associated with a significant reduction in stride duration (0.69 ± 0.01 vs. 0.66 ± 0.01 s, P = 0.001), stride length (2.49 ± 0.03 vs. 2.38 ± 0.02 m, P = 0.004) and stance duration (27.3 ± 1.0 vs. 32.0 ± 1.0%, P = 0.001). The changes in gait appeared to be associated with the introduction of galloping and the associated tolerance of the CONDEX group to the musculoskeletal loading.

Athletes of various sports are required to utilize mouthguards during practice and competitions for protection against orofacial and dental injuries, regardless of the effects on performance. Recent advances in neuromuscular dentistry have led to the development of a mouthguard touted also to enhance the performance through jaw realignment. The purpose of this study was to compare the effects of a neuromuscular dentistry-based mouthguard to a standard, custom-fitted mouthguard (CFM) on muscular endurance, anaerobic power and anaerobic capacity in competitive athletes. Professional and Division I college athletes (n = 22, Mweight = 86.2 ± 3.1 kg) participated in this double-blind, crossover study. Subjects were randomly assigned to order of use of either the experimental (Pure Power Mouthguard (PPM)) or the traditional CFM. Subjects completed two separate sessions in which they completed three performance tests, which included vertical jump (VJ), bench press (BP) and a 30 s Wingate anaerobic test (WAnT)+eight 10 s intervals, while wearing the assigned mouthguard. Significantly better performance was found for PPM compared with CFM for VJ (67.6+9.4 cm vs. 65.3+8.6 cm; P = 0.003), 30 s WAnT peak power (11.6 ± 1.7 W kg− 1vs. 11.1 ± 1.5 W kg− 1, P = 0.038), average peak power for WAnT+intervals (10.6 ± 1.4 W kg− 1vs. 10.1 ± 1.2 W kg− 1, P = 0.025) and average mean power for WAnT+intervals (9.0 ± 1.1 W kg− 1vs. 8.7 ± 1.0 W kg− 1, P = 0.034). There were no significant differences for either BP or 30 s WAnT mean power (P>0.48). Compared with a CFM, a neuromuscular dentistry-based mouthguard appears to enhance peak power output, performance and repeated maximal efforts. When required to wear a mouthguard, athletes may benefit from wearing a neuromuscular dentistry-designed mouthguard compared with a CFM.

The objective of this experiment was to test the hypothesis that ageing and training alter plasma concentrations of the peptide hormone leptin. The rationale for the study is based on prior investigations performed at Rutgers University, where published reports documented that ageing disrupts the immune and endocrine responses to acute exercise. The training period for mature young (7.3 ± 0.6 years; n = 6) and old (22 ± 0.7 years; n = 6) (mean ± SE) Standardbred mares was the duration of the summer of 2009 at the Rutgers University Equine Exercise Physiology Laboratory. Mares exercise trained in groups of six in a free-stall motorized circular equine exercise machine for 30 min, three times per week. Each mare performed a graded exercise test before beginning the training, after 8 weeks and after 16 weeks of training. There was no difference in body mass due to age (P>0.10). Training caused an increased body mass in both old and young mares (P < 0.10). There was no effect of training on rump fat thickness; therefore, the increase in body mass was primarily due to an increase in fat-free mass (muscle mass). Old mares had significantly lower plasma leptin at all sample points. As a result of training, plasma leptin was reduced in both old and young mares (P < 0.10). In conclusion, age and training appear to alter the relationship between body composition and leptin.

There is a known relationship between the way athletes interpret the feelings of pre-competitive arousal and their subsequent performance. In equestrian sports, riders' interpretation of their horses' temperament may be an additional unique factor that influences the levels of riders' arousal and performance. The aims of this study were to investigate the effect of intensity and direction of arousal and self-confidence on dressage and showjumping performance, and to investigate how perception of equine temperament traits affects arousal and performance components. Twenty-two student riders (6 males and 16 females) competing on unfamiliar horses were asked to participate in the study. Prior to the competition, riders completed the Revised Competitive Sport Anxiety Inventory-2 with direction scales, and rated their horses' temperament traits using an adapted Five Factor Personality Questionnaire. Dressage performance scores were found to be positively correlated with cognitive arousal intensity (r = − 0.5, P < 0.05) and self-confidence intensity (r = 0.59, P < 0.05). Showjumping faults were negatively correlated with somatic arousal direction scores (r = − 0.5, P < 0.05). The ‘active’ equine temperament was positively correlated with somatic arousal intensity in showjumping (r = 0.5, P < 0.05) and negatively correlated with cognitive arousal direction in dressage (r = − 0.71, P < 0.005). These findings suggest that riders' self-confidence and perception of their horses' temperament are important factors in the relationship between arousal and performance.

Recent research has shown that more successful performers interpret pre-competitive symptoms of anxiety as more facilitative than less successful performers. The aim of the study was to examine the effect of both trait anxiety and pre-competitive arousal intensity and direction on intermediate riding performance in the disciplines of dressage, showjumping and eventing, including whether differences exist between male and female riders. Sixty-two German equestrian riders competing in the equestrian disciplines of dressage (N = 21), showjumping (N = 25) and eventing (N = 16) were asked to complete a revised version of the ‘Wettkampf–Angst-Inventar-State’, which included directional scales for all items and the ‘Wettkampf–Angst-Inventar-Trait’ prior to competing. Final placings were used as performance indicators. Most important findings revealed that the correlation between direction of somatic state arousal and competitive placing was nearing significance (rs = − 0.23, P =0.07) and loss of focus was positively correlated to competitive placing (rs = 0.26, P < 0.05). Conclusions may be drawn that in equestrian sports positive interpretation of physical symptoms of arousal as well as the ability to remain focused on the task at hand may lead to more successful ridden performance.

To examine the management and training practices of 2-year-old racehorses, cross-sectional survey data were collected from 55 racehorse trainers in the Central Districts (n = 15) and the Northern Districts (n = 40) of New Zealand during summer 2007–08. Trainers in the Northern Districts had a higher proportion of 2-year-olds in work at one time (0.31 vs. 0.22, P = 0.001) than trainers in the Central Districts. Group S (trainers with 10–19 horses in work) had a higher proportion of 2-year-olds in work than both groups M (20–30 horses in work) and L ( ≥ 40 horses in work) (0.38 vs. 0.24 vs. 0.29, respectively, P = 0.001). Most horses were broken prior to the start of the official 2-year-old racing year. Most trainers had a standard pattern of training, and they cited early education as the primary reason for training 2-year-olds; the primary training milestones used to evaluate the horses' training were the first training gallop and the entry in a race trial. Training programmes were geared towards 2-year-olds being ready for trialling and racing earlier in the Northern Districts than in the Central Districts. Training practices were affected more by the trainer's location than by the number of horses trained.

Nutritional management studies to improve the performance of Mangalarga Marchador (MM) horses during the marcha test are limited. This study was designed to test the hypothesis that chelated trivalent chromium (CR) feed supplementation may reduce the suitability of the length of the interval between concentrate feeding and the marcha test among MM horses. A total of 12 healthy mares (4.25 ± 0.62 years) were randomly assigned to one of six dietary treatments (0 or 10 mg Cr by concentrate, fed 0.5, 2 or 4 h before exercise), according to a completely randomized design, with a split-plot arrangement. The diet was Cynodon pasture and concentrate (50:50 ratio). The first 29 days of the trial were for diet, Cr and exercise adaptation; during the next 15 days, horses were submitted to three 50-min field marcha tests, once a week. Heart rate (HR) was measured before, during and until 25 min after the exercise. Respiratory rate and rectal temperature were measured; blood samples were collected before, at the end and 25 min after the test. There was no effect of Cr by concentrate feeding strategy on any physiological variables (P>0.05). Supplementation of Cr increased glycaemia before and soon after the second marcha test (P < 0.01). In addition, Cr reduced HR during the second marcha test and decreased the time to first post-exercise HR recovery (P < 0.05). Insulinaemia was greater when the concentrate was provided 2 h prior to the test (P < 0.05). Concentrate provided 0.5 and 2 h before the test reduced plasma triacylglycerol in the first and second tests, respectively. The interval between concentrate feeding and marcha tests should not be decreased in horses supplemented with Cr. Horses should be fed more than 2 h before that test. Cr supplementation during training may improve the cardiac performance of MM mares during the marcha test.

To help resolve the mechanistic bases for haematological adaptations (~28% increase in red blood cell volume) of equids to high altitude (3800 m, barometric pressure Pb, 487 mm Hg) and exercise, plasma erythropoietin concentration ([EPO]) was measured at rest and following exercise in six, moderately fit equids (four Arabians, one Quarter Horse and one Shetland Pony; four females and two males; age 9.0 ± 4.5 years (mean ± SD)). [EPO] was measured on 2 days at 225 m (i.e. ~sea level; Pb, 743 mm Hg), over the course of a 10-day altitude exposure, and then again for 2 days after return to sea level. A standard track exercise test (submaximal, speed set-to-heart rate of 110 (trot), 150 (canter), 180 (gallop) bpm) was performed 2 days pre-high-altitude exposure and on three separate days at high altitude. In addition, a maximal incremental exercise test was performed on a high-speed motor-driven treadmill at sea level and 2 days following return to sea level from high altitude. Resting [EPO] increased from 28 ± 29 at sea level to 144 ± 46 mU ml− 1 (P < 0.05) on the first day at high altitude. By day 2 at high altitude, [EPO] had returned to baseline (31 ± 24 mU ml− 1, P>0.05 vs. pre-high altitude) and did not change over the remaining 8 days at high altitude nor over the 2 days after return to sea level. [EPO] was not significantly altered by acute exercise at sea level or at 3800 m. These results indicate that [EPO] increases rapidly (though transiently) in response to hypobaric hypoxia but not to acute exercise, and that exercise does not appear to potentiate the altitude response. Thus, if any [EPO]-derived haematological adaptations to high altitude are present, these appear to result from a transient ~4-fold elevation of [EPO] rather than any sustained increase in this signalling mechanism, at least in the equid.

The present study tested the hypothesis that lactate concentration ([La− ]) would differ between sample sites and between assay techniques that used different analytical substrates. Six clinically normal adult (two Thoroughbreds, three Standardbreds and one Quarter Horse) mares weighing between 435 and 560 kg were used in the study. Each mare performed an incremental exercise test (graded exercise test, GXT) where it ran on a treadmill at a fixed 6% grade. The GXT started at 3 m s− 1 for 1 min with increased in speed by 1 m s− 1 every 60 s until the horses completed the final 10 m s− 1 step. Jugular vein, pulmonary arterial and carotid arterial blood samples (14 ml) were collected before exercise and during the last 10 s of each step of the GXT. [La− ] was measured in whole blood (WB, no manipulations), total blood (TB, where the red blood cells were lysed) and plasma. Data were used to calculate the velocity to produce [La− ] of 4 mmol l− 1 (VLA4) and 10 mmol l− 1 (VLA10). Statistical analysis utilized a three-way ANOVA and, where appropriate, the Holm–Sidak or the Student Neuman–Keuls method for post hoc comparisons. The null hypothesis was rejected when P < 0.05. There was an effect of exercise intensity on [La− ] for all three methods (P < 0.001) with all means during exercise significantly greater than the resting mean, and there were differences due to method (i.e. analytical substrate) (P < 0.001) and sample site (P = 0.043). Comparisons of least-squared means (LSM ± SE) within site revealed that there was a difference (P < 0.05) between jugular vein (5.41 ± 0.24) and carotid artery (6.24 ± 0.24) and between carotid and pulmonary artery (5.98 ± 0.24). There was no difference (P>0.05) between jugular vein and pulmonary artery. Within method, there was a difference (P < 0.05) between WB (6.54 ± 0.36) and TB (5.06 ± 0.36) and between TB and plasma (6.04 ± 0.64), but there was no difference (P>0.05) between WB (6.54 ± 0.36) and plasma (6.04 ± 0.64). Further analysis of the data demonstrated that the method and sample site influenced (P < 0.05) VLA4 and VLA10.