We have used a one-dimensional chemical diffusive model of the ionosphere, in conjunction with the measured Voyager ultraviolet spectrometer (UVS) upper atmospheric temperature and composition structure, to analyze the Voyager measurements of Saturn's upper ionospheric electron densities. Electron density measurements are available from the analysis of the radio science (RSS) experiments. In addition, if interpreted as an atmospheric phenomenon the Saturn electrostatic discharges (SED), detected by the Voyager planetary radio astronomy instrument, the measurements also yield important information on electron densities at the peak. These latter data suggest a strong day-to-night variation in the peak electron densities with a maximum of ~5¿105 cm-3 near noon and a minimum of ~103 cm-3 near midnight. The analysis of RSS data yields a peak of ~104 cm-3 in the electron density profiles near the local terminators. As with simpler ionospheric models, we find serious difficulties in interpreting these electron density measurements because time constants in the model for the chemical sinks in the ionosphere, of mostly H+ ions, are >106 s much longer than a Saturnian day, suggesting little diurnal variation. We have investigated the effects of H2 vibrational excitation (characterized by a single vibrational temperature) and a gaseous H2O influx from Saturn's rings, as enhancers of plasma recombination, since they result in the conversion of slowly recombining H+ ions to more rapidly recombining molecular ions. This does introduce a larger diurnal variation in the peak electron densities, but not as strong as that deduced from the analysis of SED measurements. Furthermore, it also reduces the magnitude of the electron densities at the peak well below that inferred from the SED data. Based on the model results and extrapolation of calculations by Kim and Fox <1991> from Jupiter, we suggest that SEDs may not be associated with atmospheric storm systems. In our attempt to interpret the RSS measurements of electron density profile at sunrise, we have invoked an upward field-aligned drift, H2 vibrational excitation and an influx of H2O to explain the results and, in particular, maintain plasma peak at the measured altitude. ¿ American Geophysical Union 1996