Intervals between sampling are a major concern in studies of behaviour. Objective choices compromise accurate definition, collection of enough data and avoidance of time dependence between samples. This paper tests whether significant differences occur in the proportion of activities of Sotalia guianensis when the protocols of observation using different intervals between registers are varied. The study was carried out at Formosa Bay, south coast of Rio Grande do Norte over a period of 40 days (251 hours). The results show that the use of different intervals of data registering result in non-significant differences in the quantification of behaviour. However, a tendency to over-estimate forage and under-estimate travel at 2 minutes relative to 8 minutes intervals was detected. This warns of the need to define a proper interval between behavioural sampling and addresses the question of the adoption of a common protocol for fast and continuous assessment of different estuarine dolphin populations throughout the Brazilian coast.
