The effects of ionic conductivity and buffer concentration of electrolytes used for in vitro measurement of the charge-injection limits of activated iridium oxide (AIROF) neural stimulation electrodes have been investigated. Charge-injection limits of AIROF microelectrodes were measured in saline with a range of phosphate buffer concentrations from [PO(4)(3-)] = 0 to [PO(4)(3-)] = 103 mM and ionic conductivities from 2-28 mS cm(-1). The charge-injection limits were insensitive to the buffer concentration, but varied significantly with ionic conductivity. Using 0.4 ms cathodal current pulses at 50 Hz, the charge-injection limit increased from 0.5 mC cm(-2) to 2.1 mC cm(-2) as the conductivity was increased from 2 mS cm(-1) to 28 mS cm(-1). An explanation is proposed in which the observed dependence on ionic conductivity arises from non-uniform reduction and oxidation within the porous AIROF and from uncorrected iR-drops that result in an overestimation of the redox potential during pulsing. Conversely, slow-sweep-rate cyclic voltammograms (CVs) were sensitive to buffer concentration with the potentials of the primary Ir(3+)/Ir(4+) reduction and oxidation reactions shifting approximately 300 mV as the buffer concentration decreased from [PO(4)(3-)] = 103 mM to [PO(4)(3-)] = 0 mM. The CV response was insensitive to ionic conductivity. A comparison of in vitro AIROF charge-injection limits in commonly employed electrolyte models of extracellular fluid revealed a significant dependence on the electrolyte, with more than a factor of 4 difference under some pulsing conditions, emphasizing the need to select an electrolyte model that closely matches the conductivity and ionic composition of the in vivo environment.