Summary
Molecular techniques have recently been used to identify prey in the scats of pinnipeds. This new approach has developed in an attempt to overcome some of the inadequacies of traditional diet assessment methods. This is the first study to experimentally compare molecular and traditional dietary analyses in captive seals fed a known diet. Currently, the most widely used method of diet sampling in seals is identification of fish otoliths, squid mouth-parts and crustacean exoskeleton remains in scats. However, prey species are often poorly represented by these hard parts in scats. The diet of fur seals (Arctocephalus spp.) is an ongoing focus of Southern Hemisphere marine ecosystem research. We therefore conducted a feeding experiment on captive Arctocephalus seals (N = 4) to compare detection of prey remains in scats using hard part and molecular analyses. Animals were fed Nototodarus sp. (13% of daily intake by mass) for 9 days followed by Sillago spp. and Arripis georgianus (35% each) for 9 days. This regime was then repeated. The remainder of the diet was made up of fish genetically unrelated to these prey at the level of family. Otoliths and squid beaks were recovered from scats and identified morphologically where possible. Prey DNA was detected using a PCR based technique. Sillaginidae and Arripidae DNA were detected in more scats (78% and 48% respectively) than were otoliths (55% and 24% respectively). Even though both fish species tested have large, robust otoliths which are more likely to be present in seal scats than otoliths of many other fish taxa, they were more consistently represented in scats by their DNA remains. Squid DNA was detected in 48% of scats, while squid beaks were present in only 9% of scats. Although not yet quantitative, this molecular technique shows promise for improving assessment of pinniped diets, particularly where prey have fragile otoliths or are soft bodied.