The aim of the present study was to determine if the breeding potential of 25-month old, extensively kept, Bonsmara beef bulls can be predicted from production data, spermatozoal characteristics and/or blood hormone concentrations at that age. A further objective of the study was to determine if any of the above mentioned criteria could be associated with the libido of these bulls. Forty-one Bonsmara bulls were included in an on-farm performance test (Phase D1 growth test) for a period of 180 days. At an average age of 24.7 months, blood sampling took place (before and after GnRH treatment) and the bulls were subjected to a libido test, after which further blood samples were collected. Blood sample were analysed for cortisol and testosterone concentrations. The bulls were also subjected to an Overall Breeding Soundness Evaluation. This procedure involves an evaluation of the physical genitalia of the bulls, a measurement of scrotal circumference and semen evaluation. For purposes of statistical analyses the bulls were categorised into independent breeding potential categories according to the scores they obtained for the measured reproductive traits. The categories included scrotal circumference, spermatozoal morphology and motility and the overall breeding soundness category. A statistical analysis of the data was done by using the general linear models (GLM) procedure of the Statistical Analyses System (SAS version 8.2 BMDP). The production and growth measurements of the Bonsmara bulls did not differ between any of the high and low fertility categories and can not be used to predict the breeding potential of young bulls. The correlation between pre-weaning growth rate and the percentage morphologically normal spermatozoa was positive (r = 0.33; P<0.1), suggesting that relatively high growth rates before weaning may have a positive effect on potential fertility under normal extensive feeding conditions. By contrast, numeric differences in growth after weaning suggest that a high growth rate after weaning may have a negative effect on potential fertility. The results showed that the overall breeding soundness categories tended to be influenced by the pre-weaning growth rate (r = 0.24; P>0.1) and body lengths (r = 0.18; P>0.1) of bulls. Sampling time had a statistically significant effect on blood cortisol and testosterone concentrations for all of the breeding potential categories. Testosterone concentration increased significantly (P<0.001) after GnRH treatment. High plasma cortisol concentrations were associated with low plasma testosterone concentrations. High testosterone concentrations were associated with less spermatozoal morphological defects (r = - 0.21; P>0.1). The testosterone concentrations before GnRH treatment was higher for bulls with exceptional fertility (P<0.05), while testosterone concentration after GnRH treatment tended to be higher (P<0.1) for the bulls with acceptable fertility. This observation may be explained by the negative feedback system that operates between LH and testosterone secretion. The percentage spermatozoal defects were influenced to a greater extent by morphological abnormalities leading to reduced motility of the sperm than by any other abnormalities. From the results it seems that the semen morphology category is a better indicator of semen quality than the SC and semen motility categories. Overall breeding soundness classifications of bulls were largely influenced by spermatozoal motility (P<0.001) and to a lesser extend by spermatozoal morphology and SC. None of the reproductive and production measurements showed a correlation with libido scores, implying that optimal bull reproductive evaluation should include the assessment of both breeding soundness and libido. / Dissertation (MSc(Agric))--University of Pretoria, 2009. / Animal and Wildlife Sciences / unrestricted