The use of a seaweed habitat by red deer Cervus elaphus L. on the Isle of Rum, Scotland, was examined in detail. New information is provided on diet selection, timing of seaweed use relative to tides, inter-individual differences in seaweed use, and sex differences in site use within the seaweed habitat (‘site segregation’). Interestingly, seaweed use by adult males and females was closely correlated to that of their mothers. This implies that deer ‘learn’ early in life to include seaweed into their diet. Formerly, it has been suggested that male inferiority in indirect competitive ability relative to females causes site segregation in dimorphic ungulates (‘indirect competition hypothesis’). The observed pattern of site segregation within the seaweed habitat was used to test the hypothesis, which predicts that males should be found at sites where they can achieve higher intake rates, but where forage quality is lower than at female sites. With respect to seaweed use, the hypothesis further predicts that segregation should be lower within the seaweed habitat than within terrestrial vegetation communities, and that males should time seaweed use earlier (relative to the tide) than females. This is because seaweed availability is more subject to tidal rhythm than to indirect competition in comparison to terrestrial habitats. Males and females used different bays, and within bays they used different fractions of seaweed. However, male-preferred sites did not yield higher intake rates and were not of lower forage quality than sites preferred by females. Moreover, segregation was not lower within the seaweed habitat than within terrestrial vegetation communities, and males did not time their seaweed use earlier relative to the tide than did females. The indirect competition hypothesis could not explain the observed pattern of site segregation. Other factors, such as sex differences in sheltering or anti-predator behaviour, or social harassment, could be responsible instead.