The study aimed at classifying Nigerian indigenous breeds of sheep on the basis of their head conformation using multivariate analyses. Twelve cephalic traits of 1,200 sheep belonging to four breeds (Balami, Yankasa, Uda and WAD) were measured. The traits were subjected to principal component (PC) analysis to reduce data dimensionality. A discriminant canonical analysis was also applied to differentiate between the breeds. The cephalic index revealed that the sheep breeds are brachycephalic (short-headed) except Uda which was mesocephalic (medium-headed). The first PC reflected variables related to length and diameter amplitudes and explained 37.5 percent of the total variance. The second PC gave major relevance to skull, face and neck lengths of the sheep and contributed to 21.30 percent of the total variance, while the third PC loaded highly for skull length, head depth and neck length and contributed 17.35 percent of the total variance. The stepwise discriminant analysis revealed that skull width, head width, head length and head depth were the most discriminating variables to separate the four breeds of sheep.
