Treatments included a no-virus control and a virus exposure of 103 plaque-forming units mL21. Both treatments were replicated 20 times for a total of 40 experimental units per trial. We inoculated the water with 29.5 mL of Eagle��s Minimal Essential Media for the no-virus control tubs and 29.5 mL of MEM containing the virus for the virus tubs. The resulting virus concentration was 103 PFUs mL21, which is within the range of doses used inother studies and ecologically relevant. Given that some species in our study developed rapidly, we used a 3-day exposure in an attempt to target the intended developmental stage rather than a subsequent stage. After three days, individuals were removed from the containers, rinsed with sterile water, and placed into a new container with 500-mL of fresh aged tap water. For the remainder of the experiment, water was changed every three days to maintain water quality. After each water change, individuals in the larval and metamorph experiments were fed ground TetraMinH at a daily rate of 8% body mass. Prior to the water change, we weighed a group of 10 non-experimental individuals housed under identical conditions to calculate food rations based on the average mass. Individuals in the embryo and hatchling experiments were fed if they reached stage 25 prior to the end of the experiment, which is when yolk reserves are exhausted and jaw development is complete in most species. After the initial exposure and water change, platforms were placed in the metamorph experimental units to allow individuals to crawl out of the water to complete metamorphosis. Once individuals in the metamorph stage experiments began tail resorption, feedings were terminated and water depth was slowly reduced until a minimal amount of water remained to provide moisture for the individual and TetraMinH was no longer added. Following tail resorption, individuals were fed 10 seed weevils every three days. The experimental units were monitored three times daily for mortality. Dead larvae and metamorphs were necropsied using sterilized forceps and scissors. Because the kidneys and liver are known sites of ranavirus infection, we removed sections of these organs from each individual, placed the pooled sample in a 1.5-mL microcentrifuge tube, and froze at 280uC for molecular testing. Dead embryos and hatchlings were rinsed with sterile water and frozen at 280uC, because their small size prevented consistent necropsies. After 14 days, all live individuals were euthanized in benzocaine hydrochloride and the identical necropsy procedures followed. We set 14 days as the experiment duration because previous research has shown this is sufficient duration to observe disease from ranavirus amino acid infection with a 3-day water bath exposure. The response variables for each experiment included final mortality and infection prevalence calculated from binary data. Differences in final mortality and infection prevalence were tested among species and developmental stages using logistic regression analysis. We did not include the control treatment in the analysis because control mortality was low resulting in low or zero counts for prevalence estimates of several developmental stages, which could have biased the logistic regression results. Instead, median control mortality among developmental stages was provided for each species. If the Wald��s chi-square test associated with the logistic regression analysis was significant, we used binomial tests that were Bonferroni corrected to test for pairwise differences between proportions.