In this study, xanthan gum-facilitated ethyl cellulose microsponges were prepared by the double emulsification technique and subsequently dispersed in a carbopol gel base for controlled delivery of diclofenac sodium to the skin. Scanning electron microscopy revealed the porous, spherical nature of the microsponges. Increase in the drug/ polymer ratio (0.4:1, 0.6:1, 0.8:1, m/m) increased their yield (79.1-88.5 %), drug entrapment efficiency (50.0-64.1 %), and mean particle diameter (181-255 mm). Compared to the microsponges with high drug/polymer ratio (0.8:1, m/m), the flux of entrapped drug through excised rat skin decreased by 19.9 % and 17.0 %, respectively, for the microsponges prepared at low and intermediate drug/polymer ratios. When an equivalent amount of pure drug (not entrapped into microsponges) was dispersed into the gel base and the flux was compared, the microsponges (drug/polymer ratio 0.8:1, m/m) were found to reduce the flux by 33.3 %. Whether the drug was dispersed either in un-entrapped or entrapped form into the gel base, the drug permeation through rat skin followed Higuchi's diffusion kinetic model. The microsponges prepared at the lowest drug/polymer ratio exhibited a comparatively slower drug permeation profile and were hence considered most suitable for controlled drug delivery application. FTIR spectroscopy and DSC analyses indicated the chemically stable, amorphous nature of the drug in these microsponges. The gel containing these optimized microsponges was comparable to that of a commercial gel formulation and did not show serious dermal reactions. Hence, the microsponge system obtained at the lowest drug/polymer ratio could be useful for controlled release of diclofenac sodium to the skin.