Excavation counts (absolute density), burrow counts and surface active individuals visual counts in a Uca tangeri population were compared. Population density estimated from burrow counts was not signi¢cantly di¡erent from population density estimated by excavation, but the population density estimated with visual counts was signi¢cantly lower than the absolute density. Ovigerous females should be the main cause of these di¡erences because they frequently remain underground, but other groups of crabs are also contributing to this di¡erence. Burrow counting is an accurate and quick non-invasive method for the census of U. tangeri. Other population measures (e.g. sex ratio) are not obtainable by burrow counting and only further understanding of the activity patterns of this species along the tide cycle, moon phase and reproductive season will allow visual counting of surface active individuals to be used as quick, e⁄cient and non-invasive method for crab density and population structure estimation.Excavation is regarded as the most accurate method to estimate the density of ¢ddler crab populations but it is intrusive and labour intensive, compromising the equilibrium of mud£at and salt marsh habitats. Therefore, alternative methods have been used, such as burrow counting (e.g. Ens et al., 1993;Mouton & Felder, 1996) and visual counts of surface-active individuals (e.g. Zucker, 1978; Wolfrath, 1993;Oliveira et al., 2000).Uca tangeri is the only ¢ddler crab to occur in Europe and is found on almost all mud£ats and salt marshes at Ria Formosa. Fieldwork took place from May to July 1999 in two periods between full moon and waning moon, at Cacela Velha and Fuzeta (Ria Formosa Natural Park, Algarve, Portugal). The air temperature ranged from 21.08C to 33.58C, higher than the minimum temperature needed for U. tangeri surface activity (188C), (Wolfrath, 1993). Each day, ¢ve square plots of 1m 2 were placed on the ground and the counting of surface-active individuals was conducted on the same day, after the set-up. The observer would remain motionless for 15 min at a distance of 3.5 m from the squares, back towards the sun, before starting observations. A total of six counts were done with a 5 min interval between them, always during ebb tide. At each count, the number of males and females per square were recorded, using binoculars. Individuals di⁄cult to sex at that distance were counted in a third category. In subsequent analysis, the average of the six observations was used when referring to visual counts of surface-active individuals. After the end of the observations, the number of burrows in each square was counted, and their diameter was measured using Vernier callipers ( AE0.01mm). This procedure was done for six days, completing a total of 30 observed squares. The burrows were then excavated individually throughout their depth until the occupant was found or the burrow was found to be empty. The rest of each plot was also excavated to a depth of approximately 50 cm and all the crabs found in the soil were collecte...