Variations in the Townsend first ionization coefficient for gases

The distributions of sizes of electron avalanches in methane gas subjected to uniform electric fields no longer satisfy the simple exponential Furry distribution when the ratio E/p 0 exceeds 50 v cm -1 torr -1 . The reasons for this deviation were explained previously by Legler, who suggested that the probability of ionization was a function of the distance travelled by electrons from their previous ionizing collision. An alternative explanation is offered which suggests that the effect is due to the first Townsend ionization coefficient α decreasing monotonically with the avalanche size, so giving a Polya distribution. Such a decrease in α is consistent with the process by which the energy distribution of electrons emitted from the cathode gradually reaches a stationary state in the gas. An analysis based on the Polya distribution also gives results in good agreement with the hitherto apparently unrelated investigations of other workers.