Invasive cane toads (Bufo marinus) cause mass mortality of freshwater crocodiles (Crocodylus johnstoni) in tropical Australia Mike Letnic * , Jonathan K. Webb, Richard Shine School of Biological Sciences, Building A08, University of Sydney, NSW 2006, Australia ARTICLE INFO Article history: Received 13 March 2008 Received in revised form 11 April 2008 Accepted 16 April 2008 Available online 18 June 2008 Keywords: Bufonidae Crocodylidae Ecological impact Invasive species ABSTRACT Invasive species are frequently blamed for faunal declines, but there is little direct evidence about the pathways, magnitude and size-selectivity of mortality induced by invaders. Top predators are of particular interest in this context, because their removal can generate sub- stantial cascades of secondary effects on community composition. Cane toads (Bufo mari- nus) are large South American anurans currently spreading rapidly through tropical Australia. Native predators that attempt to consume these highly toxic toads may die as a result. During surveys of the Victoria River in the semi-arid tropical region of the North- ern Territory, we documented massive mortality of freshwater crocodiles (Crocodylus john- stoni) at the toad invasion front. Dead crocodiles spanned a wide size range (0.6–2.1 m long) but with significant biases; intermediate-sized animals (0.6–1.5 m long) were more likely to be found dead. Population densities of crocodiles plummeted by as much as 77% following toad invasion, and population size-structures changed. The negative impacts of toads on crocodiles appear to be greater in these hot semi-arid landscapes than in cooler, higher rainfall areas where crocodiles have access to a wider prey base, and the toads are less prone to desiccation and can rehydrate in small, scattered water bodies rather than in the main river. Hence, the impact of cane toad invasion on this top predator may increase with increasing aridity. Ó 2008 Elsevier Ltd. All rights reserved. 1. Introduction Human activities have dramatically increased the rate at which animal species are translocated around the globe, and many of these taxa have thrived in their new homes (Mooney and Drake 1986). If these invasive taxa substantially modify native ecosystems, the consequent threat to biodiver- sity may rank second only to direct habitat destruction (Vito- usek et al., 1997; Clavero and Garcia-Berthou, 2005). Despite this broad consensus, for most ecological systems there is lit- tle detailed information on the mechanisms of impact of invasive taxa, the magnitude of their effects, and the ways in which those effects target specific size classes within pop- ulations of native taxa (Mack et al., 2000; Baxter et al., 2004). Most studies reporting the impacts of biological invasions are conducted post-invasion and typically evaluate the im- pacts of invasive species by manipulating their abundance or access to the species or ecosystem of interest. Few quanti- tative studies have reported the ecological impacts of a bio- logical invasion using data collected before and after the invasion, hence the true extent of invasive species’ impacts can be difficult to deduce (Parker et al., 1999; Sanders et al., 2003). One reason for this is that selective pressure exerted by the invasive species is likely to change the genome of 0006-3207/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.biocon.2008.04.031 * Corresponding author. Tel.: +61 2 9036 7833; fax: +61 2 9351 4119. E-mail address: mletnic@usyd.edu.au (M. Letnic). BIOLOGICAL CONSERVATION 141 (2008) 1773 – 1782 available at www.sciencedirect.com journal homepage: www.elsevier.com/locate/biocon