AbstractCalculations have been made, on the basis of the ionic balance equations derived by Phillips (1967), of the electrical conductivity of clouds in the presence, and in the absence, of secondary ion production resulting from corona.In weakly electrified clouds ‐ no corona ‐ the immobilization of ions on cloud droplets resulting from diffusional capture results in a reduced ionic concentration and conductivity. The normalized conductivity λ the ratio of the in‐cloud value to the clear‐air value at the same altitude Z was calculated for three different cloud types (cumulus congestus, strato‐cumulus and fog) as a function of Z, liquid water content L, droplet charge Q and field strength E. λ was found to be very sensitive to variations in L and E, but only slightly affected by changes in Z, Q and the manner in which the charge is distributed over the size‐spectrum. For example, when E = O λ possessed the values 0.034 and 0.17 in a cumulus congestus cloud for L = 0.5 gm/m3 and 0.1 gm/m3 respectively, whereas when E = 90 kV/m λ was 1.5 times; 10−3 and 7.6 times; 10−3 respectively for the same two values of L.The experimental measurements of corona onset fields and currents from ice hydrometeors, Ec and ic respectively, made by Griffiths and Latham (1974) were used to calculate the normalized conductivity within strongly electrified clouds. Corona from ice provides a copious supply of positive and negative ions which causes λ to increase from between 10−3 and 10−4 to greater than 1 when Ec is achieved. The value of λ once corona is initiated depends on L and the concentration of corona sites, J. In this situation altitude Z plays an important role because Ec decreases quite rapidly with decreasing pressure. For example, with a typical distribution of temperature and pressure within a thundercloud, corona will be initiated from hailstones at Ec = 720 kV/m if Z = 2.5 km (0° C) and 540 kV/m at 5.5 km (‐18° C); the corresponding values of Ec for snowflakes are 540 and 390 kV/m respectively. At 2.5 km is slightly higher than and lower than 10 respectively, as L is increased from 1 to 5 gm/m3 with J = 1 m−3.It is concluded that the introduction of an ionic leakage current into calculations of field growth within thunderclouds is unnecessary, and that the distribution of electric field is crucial in determining whether corona, once initiated, will trigger lightning or merely increase the ionic conductivity.