The theory for double potential step chronoamperometry at spherical electrodes (HMDE and DME) previously presented (J. Electroanal. Chem., 341 (1992) 15) has been revised to explain the large discrepancies between theoretical predictions and experimental results for amalgam forming systems when the duration of the second pulse is very long. This has been accomplished by : (i) including second-order (and even third-order in the case of an HMDE) spherical corrections to the equations previously derived; (ii) incorporating terms of higher-order to the contributions included in these spherical corrections. The expressions obtained show a much better agreement with experimental results, and also some unexpected consequences predicted by the former theory at long pulse times (namely, that the current response for the second step could eventually become cathodic even if the potential for the second pulse is stepped at large positive values) are avoided.