Estimating numbers of whales present in an area from recorded call rates could be a useful conservation tool. We recorded southern right whale Eubalaena australis vocalisations and presence in Walker Bay on the south coast of South Africa. In all, 45 sessions with synchronous acoustic and visual data were analysed to determine call rates directly (overall call rate, OCR) and in relation to the number of whales sighted (call rate per whale, CPW) and number of groups sighted (call rate per group, CPG).The OCRs were examined in the presence of varying numbers of whales, using a loglinear model to investigate the dependence of the call rate on whale density. The number of whales present exerted a strong quadratic effect on the OCR, which peaked at around 15 whales and decreased to a low rate as whale presence approached a maximum, for all calls combined and for four of their constituent 13 call types: a quadratic trend was present to varying degrees among the remaining call types. Both quadratic and linear trends were absent when OCR was assessed against number of groups present, possibly because group size increased with increasing density of whales. A linear regression on the CPW and CPG suggested that there was a negative inverse relationship with the number of whales and groups present respectively. These findings are important in that, while they preclude the estimation of absolute numbers from call rates, they imply that under the conditions prevailing in Walker Bay, southern right whales were aware of the presence, arrival and departure of other animals in their vicinity and adjusted their vocal behaviour accordingly.Keywords: acoustics, call rates, Eubalaena australis, log-linear model, southern right whale, vocalisations, relative abundance IntroductionThe ability to locate and estimate the density and movement patterns of whales present in an area is an important prerequisite for many aspects of their 2 conservation, but such factors are often difficult to ascertain. Whales spend much of their time submerged, and may cover long distances daily. Visual sightings are severely hampered by rough weather conditions, and are almost impossible at night.Communication among individual cetaceans in the contexts of reproductive display, prey detection, predator avoidance and navigation, is primarily acoustic (Clark 1982, MacLennan and Simmonds 1992, Parks 2003a. Therefore, using biological sounds to detect whales (Mellinger et al. 2011) is suited both to the underwater medium and the properties of the calls. While this method can only detect vocalising individuals, acoustic detection of whales has several advantages over visual detection, as has been demonstrated for fin whales Balaenoptera physalus (McDonald and Fox 1999). It can be applied in remote situations and for extended periods of time (Clark et al. 2010, Mellinger et al. 2011, in situations where boat and aerial surveys would be impractical or expensive. Additionally, remote acoustic detection can cover a greater range than visual detection, can occur under...