The majority of research on the effect of age on physiological processes is carried out on subjects that are not physically active. A deficiency with this is that the negative effects of inactivity are implicitly ignored 1 resulting in sub-optimal values being obtained. In order to eliminate this deficiency the use of master athletes, who are less likely to show these negative effects, has been suggested1 as a more objective model of inherent human ageing. Furthermore, as a result of the methods traditionally used to depict physiological changes with age and the use of cross-sectional study designs, it is widely assumed that physiological function shows a linear decline with age. This study set out to test this assumed linear relationship between two aspects of functionality and age on a cohort of endurance trained (but not elite) master cyclists. 57 male subjects (55-79 years) were recruited from an amateur long distance cycling club where the objective of each rider is to complete events in a specified time, irrespective of age. Subjects performed an incremental exercise test to exhaustion on a cycle ergometer, with breath by breath recording of oxygen uptake performed throughout the test (OxyconPro, Jaeger, Netherlands). VO2max was determined as the highest VO2 recorded over a 20s period. Peak (explosive) power output during sprint cycling was also determined using an inertial testing system 2. Both VO2max and peak power were normalised to fat free mass obtained from DXA scans. When analysed using regression analysis, a modest linear relationship was found between VO2max , power and age (Fig 1 A and C, r = -.463 and -.401 respectively; P <0.003 in both cases). Thus even in a cohort of individuals performing similar amounts of exercise, where the effects of inactivity are negligible only 21% (VO2max) and 16% (power) of the variability is associated with age. Fig 1 B and D show the same data, but now depicted where each variable is categorised into bandwidths of equivalent function1. Here it can be clearly seen that a wide range of ages (covering up to 2 decades) is present in the majority of bandwidths. Depiction of the data in this way highlights the spread of data in a manner which suggests functionality and age are not linear processes. These findings emphasise that even in highly active ageing populations with similar activity patterns, that the relationship between physiological function and age as depicted in cross-sectional studies is not clear and unlikely to be linear. This observation is exemplified when the data are viewed in domains of equivalent function irrespective of age. These data bring in to question the commonly held belief that physiological function linearly declines with age. Longitudinal studies using appropriate models of human ageing are needed to clarify the relationship between age and physiological function.