Experimental results obtained on the transmission of a planar detonation wave from a cylindrical tube into an unconfined volume are reported. Experiments are performed using a 9.1 m long tube connected to an 88.9 cm inner diameter receiver vessel. Both the detonation tube and receiver vessel can be heated. Tests are performed using 10,20, and 27.3 cm tubes. The minimum, or critical, hydrogen concentration in air that results in detonation transmission into the receiver vessel is measured at an initial pressure of 1 atm and initial temperatures of 300,500, and 650 K. The value of the ratio of the tube diameter, d, and the critical mixture detonation cell size, lambda(c), is found to be between 18 and 24. The data show no correlation between the value of d/lambda(c) and the initial mixture temperature and there is no observable effect of temperature on the regularity of the soot foil line spacing. The uncertainty in the average cell size measurement of +/-25% is not sufficient to explain the significant departure from the classical empirical correlation d/lambda(c) = 13. It is proposed that the d/lambda(c) = 13 is neither a unique nor adequate correlation for describing detonation diffraction in mixtures considered to have a regular or irregular detonation cellular structure.