The aim of the present investigation is the analysis of the influence of liquid-fuel compressibility on the simulation of sprays produced by high-pressure injection systems. Two different equations have been introduced in the KIVA3V code to calculate liquid-phase density. The first one determines fuel density by using a second-order function of drop temperature and pressure, while the second one also takes into account the quantity of air dissolved in the fuel. Breakup, vaporization, and collision models as well as the energy, momentum, and air-spray mass exchange equations were modified so that each droplet would have a different density, according to its position and evolution. A comparison between experimental and numerical data for sprays injected in a constant-volume vessel at ambient temperature and pressure has been carried out to test the practical capability of the modified KIVA3V subroutines. The predicted and measured results of penetration versus time and drop size distribution showed good agreement. An in-depth study of the influence of gas temperature on the droplet vaporization rate has been performed for a single droplet and for sprays injected in a high-temperature, medium-pressure, constant-volume chamber. The effect of fuel density variability on vaporizing noncombusting sprays has been investigated for both models. The air dissolved in the fuel was found to affect liquid-phase density only at low ambient pressure. Finally, the experimental data measured on a small-bore diesel engine have been used to verify the provisional capabilities of constant- and variable-density models. NO and soot predictions have shown to be dependent on the model used for liquid-phase density.