Galimberti,F. & Sanvito,S. (2001) Modelling females haul out in southern elephant seals (Mirounga leonina) Aquatic Mammals 27(2): 92-104. Abstract In Pinnipedia species is difficult to achieve good estimates of population size by direct counts, because a part of the population is at sea at any time. In southern elephant seals (Mirounga leonina) the estimation of population size is carried out starting with the number of females hauled out during the breeding season, by applying a correction factor calculated from life tables. Different models were proposed in the literature to estimate the total number of females hauled out. In this paper, we consider the model proposed by Rothery & McCann (1987) for the South Georgia population, applying it to a five years data set for the population of Sea Lion Island (Falkland Islands). We test the assumptions of the model, finding them reasonable. We fit the model to our data set, obtaining an excellent fit in all cases, better than or equal to other models proposed in the literature. The precision of the estimation depended mostly on the length of presence on land of females, which is a constant of the model. A 1 day variation in length produced a 4% variation in the estimated total number of females. When a good model of the haul out process is already available for the population, even a single count close to the peak of the season is enough to estimate total females in the ± 2% range. When such a model is to be estimated from data, at least 8 counts are needed to have a good estimate. The model was not only a good description of the haul out process at population level, but its application to a set of daily counts of single harems demonstrated its usefulness also at sub-population level.