In this paper, a method which is different from that used by Leontief in testing the validity of the conclusions of the Heckscher-Ohlin theorem is recommended. We note that the empirical significance of the theorem can be established only if both its assumed
