Serum corticosterone concentration 24 hours after implantation of corticosterone pellets Mice from one of the experiments noted in Figures 1 and 2 were bled prior to removal of the thymus, and serum corticosterone concentrations were determined by radioimmunoassay. In panel A, values shown are means ± SE (n = 5 mice/group), and values significantly different from the naive control are indicated by * (p < 0.05) or ** (p < 0.01). The treatments are described in the legend for Figure 1. In panel B, non-linear regression was used to determine a best-fit line through points from 0 (ADX-N) to 5 mg corticosterone pellets (open squares). The extrapolated pellet size required to produce the same corticosterone concentration shown for Naive mice in panel A is indicated by an open circle. Dotted lines indicate the 95% confidence interval for the regression line. The R-squared value for this relationship was 0.987.