in two-digit manufacturing industries. These tests were based on the theoretical proposition that the South, being relatively labor-abundant by comparison with the non-South, should possess a comparative advantage in producing relatively labor-intensive commodities. Since estimates of aggregate regional capital stocks were not (and are not) available in the United States, the empirical criterion used for defining the South as the relatively labor-abundant region was that the average wages of production and nonproduction workers in the South are, respectively, 78 and 87 percent of the non-South average. Estle contends, however, that the South is relatively capital-abundant by comparison with the non-South and that the theoretical basis for our tests is therefore incorrect. He reaches this conclusion by demonstrating that in the manufacturing sector and in nine of the twenty two-digit major industry groups the South manifested higher gross capital-labor ratios in 1957 than the United States (and, a fortiori, than the non-South). My purpose here is to suggest that, although the South indeed displayed higher capital-labor ratios than the non-South in selected two-digit industries, the South must nonetheless be regarded as the overall laborabundant region. I suggest and test in the following section a simple hypothesis designed to reconcile the observed facts of the South's being macroeconomically labor-abundant, while displaying comparatively higher capital intensity in certain industries. The essence of this hypothesis