Theoretical values of mean statistical time lag have been estimated for overvoltages of up to 5% for a uniform-field discharge gap of 2 cm length in nitrogen at 400 torr and 20°C, with 2 mCi of cobalt 60 located in the cathode. The results are in reasonable agreement with measured values, and indicate that in order to render the mean statistical time lag negligible in comparison with the formative time lag under these conditions it would be necessary to increase the strength of the radioactive source to about 30 mCi. When the source is located in the anode it is not possible to estimate reliable theoretical values because of the difficulty of making reasonable simplifying assumptions regarding the distribution of initial ionization close to the cathode surface: the measured values do not differ by a factor of more than two from those obtained with the source in the cathode.