“…Following Nakamura et al (2003) and Egan et al (2017), trophic-guild statistical analyses were limited to species for which at least five individual stomachs contained identifiable prey. Fish diets are known to shift throughout ontogeny (Chipps & Garvey, 2007;Gray et al, 2016Gray et al, , 2017Werner & Gilliam, 1984), making it appropriate to divide fish species into size groupings prior to guild classification if such shifts indeed exist (Buchheister & Latour, 2015;Elliott et al, 2007;Egan et al, 2017;Garrison & Link, 2000;Scharf et al, 2000;Specziár & Rezsu, 2009). Therefore, prior to the guild analysis, we employed a distance-based linear modelling procedure (DISTLM) to screen species for significant relationships amongst fish L T and the prey response data, which included up to 25 prey groups per species (Table 1).…”