Table 3. ANOVA using CT_max as dependent variable, predator cues and Roundup as categorical factors (including the interaction of these factors) and, mesocosm nested within the interaction of predator cues and Roundup, for Hyla versicolor.

	SS	d.f.	MS	F	p
Predator	1.993	1	1.993	14.9	<0.001
Roundup	0.006	1	0.006	0.04	0.834
Predator*Roundup	0.009	1	0.009	0.06	0.801
Mesocosm (Predator*Roundup)	1.329	12	0.111	0.83	0.622
Error	5.350	40	0.134

Univariate tests of significance for CT_max. In this model, we used Sigma-restricted parameterization and Type III sum of squares.