The greenbug,Schizaphis graminum(Rondani), is an important pest of cereal crops, but can be managed through use of resistant crop varieties. Research concerning the nature of resistance and the development of resistant varieties requires careful experimentation, which is often conducted in plant growth chambers. However, significant variation in tests conducted with plants in growth chambers has been reported. Therefore, the objective of this research was to determine suitable experimental designs and optimum sample sizes for greenbug antibiosis tests in plant growth chambers. In our study, a greenbug antibiosis test was conducted in a growth chamber, utilizing a Latin square design to detect possible row-to-row or column-to-column variation. No significant row-to-row or column-to-column variation was found. Thus, simpler designs can be used in our chamber. However, uniformity trials should be conducted in each chamber to determine direction of possible gradients before developing appropriate statistical designs and conducting actual tests. To aid the researcher, a coefficient of variation (CV) table for different combinations of replications and subsamples was developed. In addition, the theoretical number of required replications for a given power of the test at significance levels of 0.01 and 0.05 and detectable percent differences of 5, 10, and 20% was also developed. The CV table and power of the test curves may be used to determine the number of replications required in future tests in chambers of this nature.