A series of sustained release formulations for the potent anti-inflammatory drug indomethacin were prepared by dispersing indomethacin in polysaccharide matrixes to form small microspheres. The matrixes comprised insoluble polycationic salts of polysaccharides such as, polygalacturonic acid, alginate, or carboxymethyl cellulose salts or Gantrezan (a copolymer of methyl vinyl ether maleic anhydride). The influence of the type of cation used to prepare the polymer salts and the type of polymer on the release rate of indomethacin from the microspheres was tested in vitro, and some formulations were examined in vivo in dogs. The pH of the medium was found to have a considerable influence on the in vitro release rate, the rate at pH 7.2 being much faster than that at pH 6.2 or 1.5. In vivo tests in mongrel dogs showed a big difference in plasma levels among the dogs, while the highest plasma levels were obtained after administration of the nonformulated product. Statistical treatment of the pharmacokinetic data for nonformulated and the formulated products showed no differences in T max , t 1 2 , kel, AUC and MRT. Cma obtained from the Ca-polygalacturonic acid salt formulation and the commercial product Indocin S.R. (Merck) were statistically significantly lower than the values of the non-formulated product. Since it is likely that this reduction would result in a lessening of side effects, we propose that our indomethacin formulation based on Ca-polygalacturonic acid salt would be suitable for further testing in human volunteers.