In Africa, there are two well-known subspecies of the desert locust; the Sahara subspecies, Schistocerca gregaria gregaria (Forskål, 1775), and the southern African subspecies, Schistocerca gregaria flaviventris (Burmeister). The northern locust, S. g. gregaria, extends across the Sahara desert to western Asia. The invasion area of the southern subspecies, S. g. flaviventris, extends to Namibia, Angola, South Africa, Ascension Islands and Botswana. Four different locust populations were selected, three populations for the northern S. g. gregaria (f, e, r) from different localities and one population for the southern S. g. flaviventris (s). Three characters were selected, the length of the forewing (E), the length of the posterior femur (F), and the widest part of the head capsule (C). Two ratios (E/F, F/C) were also calculated and compared for the different populations. The univariate analysis technique using the ANOVA test was performed for the four locust populations. The means of the different measured characters and the two ratios indicated significant differences between the southern population and the other three northern populations. The mean values of E, F, C, and E/F ratio were lower for the southern population compared to the three northern populations. Multivariate analysis of variance indicated that the four locust populations were highly significantly different (P < 0.0001). Standardised canonical coefficients for the three characters and the two ratios were given. The Mahalanobis distance (D2) between the centroids of the different populations for the three measured characters and the two ratios were significantly different (P < 0.0001). The distances recorded between the southern and the three northern populations were higher than the distances between the three northern populations. The results indicated that the southern males and females are more morphologically closer to each other than to either sex of the northern locusts. Also, the two sexes of the northern populations are more or less similar in their morphological features than the southern locust population. We may suggest that the southern population may categorise an evolutionary dichotomy that has arisen on the African continent and may separate the two locust populations into two distinct species.