A one year study (February, 1985 - March, 1986) was conducted to investigate changes in community structure within a 484 m2 area of an intertidal mud flat (more than 30% silt +clay) in the Gulf of Nicoya, Costa Rica (10° N, 85° W). Core samples (core area: 17.7 cm2) for macrofauna were collected monthly to a depth of 15 cm into the sediment, and preserved in 10% buffered formalin in sea water stained with Rose Bengal. Organisms retained on a 500 micron mesh sieve were considered as macrofauna. A total of 79 species was collected, with an approximate density of 14,798 individuals per m2. This density is relatively high and very similar to that found the previous year. Diversity (H') ranged from 1.61 to 3.12 and evenness (J) varied from 0.46 to 0.85. Polychaete worms dominated the community in terms of the number of individuals and species, followed by microcrustaceans (the ostracod, Cyprideis pacifica and an undescribed cumacean), molluscs, and miscellaneous groups. The numerical dominance by the polychaetes was mainly due to the spionids, Polydora citrona and Paraprionospio pinnata and the capitellid, Mediomastus californiensis, which together represented 41% of the fauna. These results are in contrast with those found the previous year, when C. pacifica and the cumacean represented 43.4% of the fauna, and the polyehaetes, M. californiensis, P. pinnata and, Lumbrineris tetraura accounted for 19.2% of the total. Most of the remaining species, however, were collected during both surveys and at very similar abundances. Thus, the observed abundance fluctuations of the dominant species are considered as natural oscillations of the community. Cluster analysis divided the biological data set into groups that corresponded in time with the observed dry and rainy seasons typical of the Gulf of Nicoya region. Four abundance patterns (a,b,c,d) were found among the top 25 important species: species with peaks of abundance coinciding with the dry (a) or rain y (b) seasons, and species increasing (e) or decreasing (d) in abundance during the sampling period. Multiple discriminant analysis on the environmental variables assigned relatively high scores to the organic matter, silt + clay, and very fine sand contents of the sediments. The abundances of some species showed significant association with one or several of the environmental variables. The importance of these factors in community structure needs further experimental work. Caging experiments (5 mm mesh size cages) resulted in non-significant changes in total abundance and species number inside cages after a period of three months. Cluster analysis, however, revealed differences between caged and uncaged sets of samples. C. pacifica and juveniles of an unidentified bivalve mollusc were more abundant inside cages, and the cumacean was more abundant outside. These results indicate that the role of macropredators (birds, crabs, fish) in community structure might be relatively unimportant. This benthic community is considered as representative of relatively unpolluted conditions since published information on trace metal concentrations and raw sewage discharges indicate that these appear characteristic of nonindustrialized estuaries. The main likely source of pollution (chemical compounds derived from agricultural practices) into the Gulf of Nicoya, however, needs to be quantified.