Thesis (M.Sc. Agriculture (Animal Production)) -- University of Limpopo, 2020 / The smallholder dairy sector in South Africa is characterized by a low input production system and poor animal productivity. Research has been carried out to benchmark cow productivity on smallholder dairy herds; however, there is a paucity of information on the current status of breeding practices and the genetic consititution of cattle used in this production system. This information is vital for the development of sound and sustainable breeding programs for SHD production, which can have an enormous positive impact on food security and rural livelihoods. Thus, the aim of this study was to evaluate the levels of genetic diversity and population structure in South African smallholder dairy (SHD) herds using single nucleotide polymorphism (SNP) markers. A total of 192 animals from SHD dairy herds were genotyped using the GeneSeek® Genomic Profiler (GGP) 150K-BeadChip. Four specialized dairy breeds included the Ayrshire(n = 200), Holstein(n = 231), Jersey (n = 224) and Nguni (n = 209) were used as the reference populations. The mean MAF values ranged from 0.30 Ayshire (AYR), Jersey (JER), and Nguni (NGI) to 0.31 Holstein (HOL) and SHD between the populations. There were slight differences in the levels of genetic diversity ranged between 0.39 (JER and NGI) to 0.40 (AYR, HOL, and SHD). A moderate level of inbreeding (0.02) was observed in the SHD population, which results in high genetic diversity among this herds. Principal Component Analysis (PCA) revealed four homogeneous clusters comprising of AYR, HOL, JER, NGI, and a heterogeneous cluster of the SHD. The heterogeneity observed in the SHD population indicates widespread crossbreeding. The model-based cluster analysis corresponded with the PCA and pointed out the predominance of HOL, JER, with marginal gene flow from the AYR and NGI. These results have provided a useful insight into the genetic structure and prevailing breeding practices on South African SHD herds. / National Research Foundation (NRF), Agricultural Research Council (ARC) and University of Limpopo (UL)