Note 478 — Chapter IV · THE CENTURIES AND THE CLASSES
Reading edition
Estimates have been made by Müller, in Philol. xxxiv (1876). 127; Delbrück, Gesch. d. Kriegsk. i. 224; Beloch, Bevölk. d. griech.-röm. Welt, 42 f.; Smith, Röm. Timokr. 67. In the United States the ratio is more than four to one; Special Reports: Suppl. Analysis and Derivative Tables, Twelfth Census of the United States, 1900, Washington, 1906. p. 170 f. The estimate given in the text is based upon the “Deutsche Sterbetafel” for men, in E. Czuber, Warscheinlichkeitsrechnung (Leipzig, 1903), p. 572, 574. The ratio is almost exactly three.