The severe lethal chondrodystrophies in man result in a common clinical syndrome including shortening of the face, mandible, and limbs. Studies of three lethal chondrodystrophic mutants in mice, viz., chondrodysplasia (cho), cartilage matrix deficiency (cmd), and disproportionate micromelia (Dmm), which share this syndrome, were performed with the aim of identifying histochemical, immunofluorescence, or ultrastructural differences which might exist among these hereditary cartilage disorders. We examined limb cartilage epiphyses from day 18 normal and mutant fetuses and observed repeatable, mostly qualitative differences. All observations were made relative to the normal control. Histochemical staining of matrix proteoglycan was moderately decreased in cho and Dmm cartilage and markedly decreased in cmd when compared to the normal control. Staining of matrix collagen was irregular in distribution in cho, increased in cmd, and decreased in Dmm. Immunofluorescence of proteoglycan was increased in the matrix of cho and Dmm and decreased in cmd. Immunofluorescence of type II collagen was heterogeneous and moderately decreased in the matrix of cho, increased in cmd, and markedly decreased in Dmm. Immunofluorescence of link protein in cho was localized in the cellular-pericellular region as in the normal and appeared increased in the matrix of cmd and Dmm. Immunofluorescence of chondronectin was localized in the cellular-pericellular region and appeared normal in all three mutants. Major differences in cellular and matrix ultrastructure were observed among the mutants, including a decreased frequency of small-diameter collagen fibrils in cho and Dmm, increased density of collagen fibrils in cmd, and dilated RER in Dmm. These observations demonstrate that distinct structural and possibly molecular differences exist among the chondrodystrophies. In the case of cmd, the differences correlated with a previously reported molecular defect, viz., absence of core protein of cartilage specific proteoglycan in the cartilage of this mutant. It is anticipated that the methods used in the present study can be applied to humans in case classification and in identifying potential mouse-human correlates.