Figures S1 - S5 from Predator specific responses and emergent multi-predator effects on oviposition site choice in gray treefrogs, Hyla chrysoscelis

Figure S1. Photo of one block (5)(top) and a tank (bottom).; Figure S2. a) Cumulative number of eggs for the 7 treatments and controls across the entire experiment. Control values were adjusted for the additional replicates. b). Cumulative number of eggs for the 5 avoided treatments.; Figure S3. Regression of total number of H. chrysoscelis eggs per patch on total fish mass per patch. This is typical of oviposition and colonization responses to fish, which are typically threshold responses to species identity rather than linear biomass-dependent responses. Masses were taken at the conclusion of the experiment on 8 December, after 209 days of growth, whereas ~95% of oviposition was completed by 8 August, by which time fish had had only 87 days of growth. Biomass would have been much less variable among treatments during the period of oviposition. Though body sizes diverged by the end of the experiment, initial sizes were matched among species, but N. crysoleucus and No. phaeus have larger terminal body sizes and F. chrysotus was much closer to its maximum size at the outset.; Figure S4. Regression of total number of H. chrysoscelis eggs per patch on total number of insects per patch. Symbols/colors as in Figure S1, cyan triangles = controls. There was no relationship between the two variables, and also no correlation within the fishless controls (r = 0.422, p = 0.224, n=10), which had the highest mean insect abundances and covered the greatest range of insect abundances. None of the major subgroups of insects showed any relationship with the number of eggs per patch, with the exception of hemipterans, which had a weak, marginally non-significant positive relationship (r = 0.25, p = 0.098, n=45) with the number of eggs, driven largely by shared preference for control and No. phaeus treatments: all beetles (r = 0.024, p = 0.8746, n=45), hydrophilid beetles (r = 0.168, p = 0.27, n=45), dytiscid beetles (r = 0.23, p = 0.1278, n=45).; Figure S5. Untransformed treatment means (± 1 SE) and p-values for one tailed Dunnett's procedure (with Dunnett-Hsu correction) comparing each treatment to the control for abundance of all insects, illustrating absolute and relative abundance of the 3 dominant taxonomic groups. N=total abundance. Treatment codes: Fundulus chrysotus (FC), Notemigonus crysoleucas (NC), Noturus phaeus (NP), F. chrysotus x N. phaeus (FCxNP), N. crysoleucas x F. chrysotus (NCxFC), N. crysoleucas x N. phaeus NCxNP, X3 (all three predators). (from Resetarits et al. In press).