Thin films of Boron Nitride (BN) and Boron Carbide (BC) possess properties that make them attractive for various applications. Epitaxially grown BN exhibits potential for optoelectronic devices, as piezoelectric materials, and graphene technology. Epitaxial BC is a semiconductor that could allow bandgap tuning and has potential applications in thermoelectric and optoelectronic devices. Both BN and BC material systems, generally deposited using chemical vapour deposition (CVD), are limited by the lack of control in depositing epitaxial films. In my thesis work, I have studied the evolution of various crystal phases of BN and BC and the factors that affect them during their CVD processes. I deposited and compared the growth of BN on Al2O3 (0001), (11 2 over bar 0), (1 1 over bar 02) and (10 1 over bar 0) substrates and used two organoboranes as boron precursors. Only Al2O3(11 2 over bar 0) and Al2O3 (0001) rendered crystalline films while the BN growth on the remaining substrates was X-ray amorphous. Furthermore, the less investigated Al2O3(11 2 over bar 0) had better crystalline quality versus the commonly used Al2O3 (0001). To further understand this, I studied crystalline BN thin films on an atomic scale and with a time evolution approach, uncovering the influence of carbon on hexagonal BN (h-BN). I showed that h-BN nucleates on both substrates but then either polytype transforms to rhombohedral-BN (r-BN) in stages, turns to less ordered turbostratic-BN or is terminated. An increase in local carbon content is the cause of these changes in epitaxial BN films during CVD. From the time evolution, we studied the effect of Al2O3 modification on h-BN nucleation during CVD. The interaction between boron and carbon during BN growth motivated studies also on the BxC materials. BxC was deposited using CVD at different temperatures on 4H-SiC(0001) (Si-face) and 4H-SiC(000 1 over bar) (C-face) substrates. Epitaxial rhombohedral-B4C (r-B4C) grew at 1300 °C on the C-face while the films deposited on the Si-face were polycrystalline. Comparing the initial nucleation layers on both 4H-SiC substrates on an atomic scale we showed that no interface phenomena are affecting epitaxial r-B4C growth conditions. We suggest that the difference in surface energy on the two substrate surfaces is the most plausible reason for the differences in epitaxial r-B4C growth conditions. In this thesis work, I identify the challenges and propose alternative routes to synthesise epitaxial BN and B4C materials using CVD. This fundamental materials science work enhances the understanding of growing these material systems epitaxially and in doing so furthers their development.