The onset of cracking during drying of alumina suspensions cast onto a substrate has been studied experimentally. It is shown that the capillary pressure would impose a compressive stress on the particle array and cannot itself cause cracking. Based on experimental observations, it is proposed that the driving force for cracking arises from a misfit strain that occurs when repulsive layers between the particles collapse completely, after the particles have adhered to the substrate. This predicts that cracking should occur while the space between the particles is still filled with liquid, which is consistent with observations using laser speckle interferometry.