Approximately half of Australiars beef breeding herds are located in northern Australia with reproductive performance being identified as having a significant impact on the profitability of these businesses. Although many of the risk factors affecting the reproductive performance of beef cattle have been previously identified, the relative contribution of major risk factors to key measures of performance, after partitioning the effects of other confounding variables, is lacking, particularly for extensive tropical rangeland systems typical of northern Australia. The overall objective of the research described in this thesis was to, i) describe the reproductive performance of commercial beef breeding cattle in northern Australia and, ii) determine, and quantify the effect of the major risk factors associated with three key measures of reproductive performance (lactating cows becoming pregnant within 4 months of calving, cows failing to become pregnant within an approximate 12 month reproductive cycle, and foetal/calf loss between confirmed pregnancy and weaning). It was hypothesised that the major risk factors determining the probability of lactating cows becoming pregnant within four months of calving are similar to those determining the probability of foetal/calf loss between confirmed pregnancy and weaning. To address these objectives, a prospective population-based epidemiological study was conducted between 2007 and 2011 involving 78 commercial beef properties located across northern Australia. Attainable and typical levels of reproductive performance for cow-age cohorts were described using a dataset containing 114,154 annual production cycles of performance data that represented approximately 78,000 heifers and cows managed in 142 breeding groups. Properties were assigned to one of four broad country types following a subjective assessment of the production potential of the grazing land and cross-referencing with pasture and vegetation descriptions reported by the herd owners/managers. The typical overall reproductive performance of cows in the Northern Forest (57 live calves at muster per 100 surviving mature cows), was considerably lower than that for the other country types (71-83 live calves at muster per 100 surviving mature cows). This was due to typically much lower percentage of cows becoming pregnant within 4 months of calving, higher foetal and calf loss and higher estimated mortality of pregnant cows in the Northern Forest compared to the other genotypes. These findings are consistent with the generally poorer quantity and quality of pasture, harsher environment and higher disease risk of cattle managed in the Northern Forest compared to the other country types. Country type and predicted calving period were top-order determinants of pregnancy within 4 months of calving while lactating. The effect of cow-age class cohort was dependent on country type with the expected occurrence of lactating cows pregnant within 4 months of calving estimated to be 23.1-53.8, 38.1-61.6, 35.4-56.5 and 38.5-59.4 percentage points lower for first-lactation, second-lactation, mature and aged cows, respectively in the Northern Forest, compared to that in other country types. The expected occurrence of cows predicted to calve in July-September which became pregnant within 4 months of calving while lactating was 49 percentage points lower than those predicted to calve in December-January (15% vrs 64%, respectively). The outcome of the previous production cycle (including the current period of calving) and the average digestibility of dry season pasture in the current production cycle were ranked as the most important risk factors for non-pregnancy. Cows which grazed pasture during the dry season with an average dry matter digestibility of l55% were estimated to have 10.2 percentage points higher occurrence of non-pregnancy than those which grazed pasture with a dry matter digestibility of g55%. Cows predicted to calve between October and November were expected to have 15.6% fewer occurrences of non-pregnancy, compared to cows expected to calve between February and March. Hip height of cows, risk of phosphorus deficiency and heat stress were determined as high-ranking contributors of foetal/calf loss in the study population. Foetal/calf loss was estimated to be 3.7 percentage points higher in tall cows than short cows. The association between temperature-humidity index (THI) and foetal/calf loss was dependent on country type, with the largest effect expected in the Northern Downs with 6.7 higher expected occurrence of foetal/calf loss estimated when the THI exceed 79 for at least two weeks during the expected month of calving. The associated effect of the risk of phosphorus deficiency was moderated by body condition score measured at the pregnancy diagnosis muster and country type. An 8 percent point higher expected occurrence of foetal/calf loss was estimated where risk of phosphorus deficiency was categorised as high in heifers and cows in poor body condition (BCS l2.5). While there was evidence at a management group-level that infectious diseases had a significant impact on reproductive performance, the dominating effect of more universal risk factors such as nutritional, management and environmental factors were highlighted in this study. In groups which had evidence of widespread infection with bovine viral diarrhoea virus (BVDV) at the time of the pregnancy diagnosis muster (g80% seropositive with evidence of recent infection) there were 23 percentage points fewer expected occurrences of lactating cows pregnant within 4 months of calving, compared to that in groups with a low BVDV seroprevalence (l20%). In groups where there was widespread evidence of Campylobacter fetus venerealis (Cfv) infection at the time of the pregnancy diagnosis muster (g30% of vaginal mucus ELISA test results positive), the percentage foetal/calf loss was 7 percentage points higher than where estimated prevalence of infection was lower (l30%). The research presented in this thesis describes the first ever population based study of the performance of commercial beef cattle breeding herds in northern Australia and the factors affecting performance. Generally the major risk factors identified were different for each measure of reproductive performance, but overall all were non-infectious factors, primarily nutritional and herd management, and environmental factors. However, when outbreaks of infection with BVDV or Cfv did occur in breeding groups this did result in significantly lower reproductive performance. The findings from this research clearly identify what factors beef cattle producers and their advisors need to focus on when investigating poorer than expected reproductive performance, and provide the basis for future research to investigate the impact of specific management interventions to control these factors.