**2. Materials and Methods**

Planulae larvae of *C. xamachana and C. frondosa* were collected from multiple female medusa from Florida Bay, Key Largo, Florida. The developing planulae were maintained in 100 μg/mL antibiotic solution of neomycin and streptomycin in filtered (0.8 μm) seawater; they were removed after hatching from the egg and placed in 35 ppt artificial seawater (ASW, Instant Ocean made in deionized (DI) water) containing no antibiotics. The oxybenzone solutions were prepared using pure oxybenzone (Benzophenone-3) powder solubilized in 5 μL of dimethyl sulfoxide (DMSO) solution and then diluted in ASW water to form stock and experimental solutions. Each concentration of oxybenzone used in experiments contained 5 μL/L DMSO. Oxybenzone has a very limited solubility in water, and even less so in seawater (some oxybenzone will come out of DMSO and float on the meniscus or adhere to the side walls of the container). Because of this, the concentration (molarity) in the dosing dishes was not known. Therefore, this experiment did not follow a validated methodology, but it does not detract from the observed dose-response toxicological behavior.

Swimming speed of planulae larvae was measured for 12 planulae in each concentration of oxybenzone. Four replicates (of 10–20 planulae per replicate) were used in the mortality experiments, and six replicates (of 10–20 planulae per replicate) were used in the settlement/metamorphosis experiments for each concentration of oxybenzone (228.0, 22.8, 2.28, 0.228, 0.0228, or 0.0 μg/L). All the experiments were conducted at 26–27 ◦C under indoor ambient light (20 μEsec<sup>−</sup>2h−1).

The larvae usually swam smoothly in a pattern following the circumference of the well. Using a stopwatch, the amount of time each larva took to swim the circumference, or for the treated larvae half the circumference, of the circular well was recorded. Larvae that did not swim around at least part of the circumference were not recorded. The number of larvae were counted in each well at the start of the mortality and settlement/metamorphosis experiments (0 h), and 1, 2, and 3 days after the start. Typical numbers ranged from 10–20 larvae per well. The difference between the number of larvae at the beginning of the experiment was compared the number at 1, 2, or 3 days to determine the number that died. Planulae usually disappear shortly after dying, with the single layer of cells in the epidermis and gastrodermis rapidly falling apart. The proportion of larvae swimming vs. those that had completed metamorphosis was monitored 1, 2, and 3 days after placing them in 100 μg/mL of the settlement inducer peptide Z-Gly-Pro-Gly-Gly-Pro-Ala-OH [16].

Proportions were arcsine square root transformed before statistical analysis. Data were tested for normality (Shapiro–Wilk test) and equal variance. The one-way ANOVA was used in parameteric tests with equal variance. When data did not meet the assumption of normality and homogeneity, a Kruskal-Wallis one-way ANOVA was used. A Tukey post-hoc test (95% confidence interval) followed. Differences were considered significant at *p* ≤ 0.05.
