Notes: Abstract. Cloning/sequencing and fragment analysis of ribosomal RNA genes (rDNA) are becoming increasingly common methods for the identification of microbial taxa. Sequences of these genes provide many additional taxonomic characters for species that otherwise have few distinctive morphological features, or that require involved microscopy or laboratory culture and testing. These same approaches are now being applied with great success in ecological studies of natural communities of microorganisms. Extensive information on the composition of natural microbial assemblages is being amassed at a rapid pace through genetic analyses of environmental samples and comparison of the resulting genetic information with well-established (and rapidly growing) public databases. We examined microbial eukaryote diversity in a natural seawater sample from the coastal western North Atlantic Ocean using two molecular biological approaches: the cloning and sequencing of rRNA genes and by fragment analysis of these genes using the terminal restriction fragment length polymorphism (T-RFLP) method. A simple experiment was carried out to examine changes in the overall eukaryote (largely protistan) diversity and species composition (phylotype diversity) of a natural microbial assemblage when a seawater sample is placed in a container and incubated at ambient light and temperature for 72 h. Containment of the natural seawater sample resulted in relatively minor changes in the overall eukaryote diversity (species richness) obtained by either molecular method at three time points (time-zero, time-24 h, time-72 h). However, substantial changes in the dominance of particular eukaryote phylotypes took place between the three sampling times. Only 18% of the total number of phylotypes observed in the study were observed at all three time points, while 65% (108 of 165) phylotypes were observed only at a single time point (54 unique phylotypes initially, 37 more unique phylotypes at 24 h, and 17 more at 72 h). The results of this study indicate that a high diversity of protistan taxa existed in the original seawater sample at very low abundance, and thus were not observed in the initial characterization of community structure. Containment resulted in significant shifts in the dominance of these taxa, enabling the presence of previously unobserved phylotypes to be documented after 24 or 72 h of incubation.