Photoelastic studies for different spanto-depth ratios were given by Frocht [17] The altered tress distributions for short beams have been investigated by Timoshenko [16 , p. 66] Caswell [18] Seewald [16, p. 99] and for short cylinders by Milligan [19]. Nadai [13, p. 164] evaluated the corrections in strength required 'when plastic flow occurs in the transverse specimens. The statistical theory relating specimen size to strength was discussed by Weibull [20] Griffith [21] and Salmassy [9, 10] Barriage [22] discussed this theory and gave data indicating causes of heterogeneity of specimens.