The objective of this study was to determine the effects of partially substituting cottonseed cake with graded levels of baobab (Adansonia digitata L.) seed cake (BSC) on milk yield and quality in Guernsey cows. Sixteen cows in mid-lactation and in their third parity were allocated to diets containing 0% (control), 5%, 10%, and 15% BSC in a completely randomized design. Each cow was given a daily feed ration of 6 kg and a basal diet of soya bean stover ad libitum. There were no differences in daily feed intake (P > 0.05), but basal intake differed among all treatment groups with cows on the control diet having the highest intake (30 ± 0.34 kg/day). Mean daily milk yield differed (P < 0.05) among all treatment groups. However, the control had higher milk yield of 12.1 ± 0.73 kg/day, and the 15% BSC had the least yield of 7.46 ± 0.73 kg/day. Cows on the control diet had higher milk butterfat content (6.12%; P < 0.05) than those on the BSC-based diets. Protein content differed (P < 0.05) across all treatment groups with cows on 15% BSC producing the highest protein content (3.43%) while the control had the least (2.6%). The concentration of milk total solids for cows fed on 15% BSC was higher (P < 0.05) than that from cows on other diets. Lactose content was not affected by the diets (P > 0.05). These results indicate that BSC can substitute soya bean cake in dairy diets, but milk production and butterfat content are compromised.