This study examines ASTM Standard E1 191-90, “Standard Guide for Conducting Life-cycle Toxicity Tests with Saltwater Mysids”, 1990, using Mysidopsis bahia, by comparing several test designs to assess growth, reproduction, and survival. The primary objective was to determine the most labor efficient and statistically powerful test design for the measurement of statistically detectable effects on biologically sensitive endpoints. Five different test designs were evaluated varying compartment size, number of organisms per compartment, and sex ratio. Results showed that while paired organisms in the ASTM design had the highest rate of reproduction among designs tested, no individual design had greater statistical power to detect differences in reproductive effects. Reproduction was not statistically different between organisms paired in the ASTM design and those with randomized sex ratios using larger test compartments. These treatments had numerically higher reproductive success and lower within tank replicate variance than treatments using smaller compartments where organisms were randomized, or had a specific sex ratio. In this study, survival and growth were not statistically different among designs tested. Within tank replicate variability can be reduced by using many exposure compartments with pairs, or few compartments with many organisms in each. While this improves variance within replicate chambers, it does not strengthen the power of detection among treatments in the test. An increase in the number of true replicates (exposure chambers) to eight will have the effect of reducing the percent detectable difference by a factor of two. The results indicate that, of the five test designs compared, test design two, with randomized sex ratios using larger test compartments, employed the same number of test organisms in fewer compartments, produced good survival and growth, yielded reproduction comparable with the ASTM method, and required less animal handling and monitoring time. We recommend adoption of this test design with the incorporation of an increase in the number of true replicates to eight.