The use of sport-specific training is crucial to enhance performance in professional road cycling, where the competition between the top riders is incredibly close. To achieve specific physiological adaptations, training programmes are developed to stress the physiology in a sport specific manner. In the sport of endurance cycling, this means targeting and overloading the working muscles and the physiology used during a race. One training technique that has found been found to induce specific kinematic and kinetic adaptations in other sports is functional resistance training (FRT). FRT involves applying a load to the body while performing sport-specific movements to induce overload and performance in those movements (Macadam et al., 2016). However, to be able to use FRT in practice requires an understanding of the physiological effects of FRT in the endurance sport of cycling, which previously has not been widely researched. Furthermore, no study has researched the use of FRT in a simulated ‘real-world’ uphill cycling environment. The objective of this thesis was to extend the current body of knowledge on the use of functional resistance training using limb loading (LL) in cycling. This thesis had two aims, answered in two studies. Firstly, to determine the acute physiological effects of LL in cycling. Secondly, to determine how LL alters physiological responses to cycling at various cycling gradients, and do responses differ when compared to adding load to the bicycle (Bload). In Study One participants performed 5 submaximal exercise bouts for 5 minutes at first ventilatory threshold (VT1) under different LL conditions (1/3 of total added load on calf; 2/3 of total added load on thigh, at 0, 2%, 4%, 6% and 8% body weight (BW)) on a stationary cycling ergometer. Physiological measures of oxygen consumption (VO2), heart rate (HR) and blood lactate (BLa) were recorded for each loading condition throughout the submaximal bouts. Pedal force measurements (PFM), and perceptual measures of rating of perceived exertion (RPE), “Comfort” and “Pain” were also recorded. LL was found to have trivial or unclear effects on physiological measures. Cycling efficiency decreased and VO2 increased with a negative linear relationship (r = -0.97 ± 0.05), despite only trivial effect sizes established for the relationship between added load and VO2. This was despite the exercise being perceptually harder with every 1% added BW (r = 0.94 ± 0.09)) (RPE: 2% = small (effect size (ES) ± 90% confidence limit (CL)) (0.24 ± 0.25); 4% = trivial; 6% and 8% = moderate (6% = 0.67 ± 0.28; 8% = 0.85 ± 0.38)), more “uncomfortable” (r = 0.89 ± 0.17)) (“Comfort”: 2% and 4% = unclear; 6% = moderate (0.82 ± 0.64); 8% = large (1.31 ± 0.90)), and more “painful” (r = 0.89 ± 0.17)) (“Pain”: 2% = trivial; 4% = unclear; 6% = small (0.57 ± 0.46); 8% = moderate (0.80 ± 0.62) compared to baseline. Consequently, it was deemed that LL did not have any physiological effect on submaximal cycling. Despite no physiological benefit found from using LL in submaximal cycling, if LL were to be used in practice, 2% and 4% BW would be the most appropriate due to their limited impact on “Comfort”, “Pain”, RPE and efficiency. In Study Two, participants completed three separate testing sessions each consisting of 4 x 5-minute exercise bouts at different gradients (2%, 4% 6% and 8%), under different loading conditions (no added load, LL and Bload). Physiological measures of VO2, HR and BLa were recorded throughout for each loading condition, alongside perceptual measures of RPE, “Comfort” and “Pain”. HR had a small effect (ES ± 90%CL: -0.33 ± 0.49) using Bload at 8% gradient but unclear for LL. The remaining HR responses at different gradients for LL and Bload were trivial (LL: 6%; Bload: 4%) or unclear (LL: 2%, and 4%; Bload: 2% and 6%). VO2 had a small effect at 2% (0.28 ± 0.36), 4% (-0.27 ± 0.38) and 6% (0.24 ± 0.22). Whereas, Bload only induced a small (0.26 ± 0.32) effect at 6% gradient, with unclear effects at 2% and 4% gradients. Trivial effects were seen at 8% gradients for both LL and Bload. Bload did not induce any effects in BLa at any of the gradients. Perceptually, LL induced a moderate (0.60 ± 0.55) effect at 4% gradient for RPE, whereas Bload only induced a small (0.53 ± 0.47) response at 4% gradient. At 2% gradient the RPE effect for Bload was trivial, whereas for LL the effect was unclear. The effects for both LL and Bload at 6% and 8% gradients were unclear. As a result, LL was found to induce greater metabolic overload at 2% and 6% gradients and to induce greater sport-specific overload than Bload. Collectively the studies in this thesis demonstrate that application of LL results in trivial effects on physiological overload during cycling. However, at specific gradients some form of physiological overload was observed (VO2 and BLa at 2% and 6% gradients), which is worthy of further consideration. This thesis has extended the limited current body of knowledge on the physiological and perceptual effects of LL, and will help to inform practitioners, coaches and athletes of its use in cycling. Additionally, it will help to guide and target future research on sport-specific overload in endurance cycling.