Sixty specimens of Caranx hippos (Linnaeus, 1766) and fifty-five specimens of Caranx latus Agassiz, 1831 collected from the coastal zone of the State of Rio de Janeiro (21-23oS, 41-45oW), Brazil, from October 1998 to October 1999, were necropsied to study their metazoan parasites. All specimens of C. hippos were parasitized, and the majority of specimens of C latus (96.4%) were parasitized by one or more metazoan species. Nineteen species of parasites were collected in C. hippos: 5 digeneans, 5 monogeneans, 2 cestodes, 5 nematodes, and 2 copepods. Seventeen species of parasites were collected in C. latus: 6 digeneans, 2 monogeneans, 3 cestodes, 4 nematodes, and 2 copepods. The endoparasites (digeneans, cestodes, and nematodes) were the majoriry, 76.4% and 63.3%, of the total number of parasite specimens collected in C. hippos and C. latus, respectively. The monogeneans Allopyragraphorus hippos (Hargis, 1956) and Cemocotyle carangis (MacCallum, 1913) were the most dominant species with the highest parasitic prevalence in the parasite community of C. hippos and C. latus, respectively. The metazoan parasites of the two host species showed the typical overdispersed pattern of distribution. Bucephalus varicus Manter, 1940. A. hippos, Protomicrocotyle mirabilis (MacCallum, 1918), Cucullanus pulcherrimus Barreto, 1918, and Lemanthropus giganteus KrΦyer, 1863 had a positive correlation only between the host's total length and abundance andlor prevalence in C. hippos. Bucephalus varicus, Tergestia pectinata (Linton, 1905), C. carangis, and Pseudoterranovo sp. had a positive correlation between the host's total length and abundance and/or prevalence in C. latus. In C. hippos, the copepod Caligus robustus Bassett-Smith, 1898 had the highest values of prevalence and abundance in the female hosts. No parasite species showed influence of the host's sex on their prevalence and abundance in C. latus. The mean diversity and the parasite species richness of the parasite infracommunities of C. hippos and C. latus were not significantly different. Only the parasite species diversity of C. hippos was correlated with the host's total length; in both host species the parasite diversity did not showed differences in relation to the sex of the host. Only one pair of ectoparasite species, A. hippos - P. mirabilis, showed significant positive co-occurrence and covariation in the parasite infracommunities of C. hippos. Two endoparasite species, B. varicus - Parahemiurus merus (Linton, 1910), showed negative co-occurrence and positive covariation; and the pair B. varicus - Pseudoterranova sp. had positive co-occurrence and covariation in the infracommunities of C. latus. The values of qualitative and quantitative simila rity coefficients between the parasite communities of C. hippos and C. latus were 55.5 and 30.4, respectively. The parasite communities of C. hippos and C. latus were defined as closest to isolationist type because there are few evidences of interspecific associations or covariations. Additional parasitological studies on other species of carangid fishes from the South American Atlantic Ocean are needed to evaluate the structure of carangid parasite communities in the Neotropical region.