The effects of kindling of the right anterior medial amygdala of Wistar rats was studied. Kindling lastingly increased anxiety (decreased open-arm exploration) in the elevated plus-maze 1 week after the last kindled seizure, replicating previous findings. Changes in anxiety were independent of changes in exploration or activity in either the plus-maze or hole board. A new finding is the dependence on baseline behavior of kindling induced behavioral changes. Using a novel-retest paradigm, it was possible to retest rats in the plus-maze without changes in their open arm explorations. This permitted pretesting rats to determine their baseline levels of plus-maze anxiety. Controls proved to be stable in their plus-maze behavior over a retest interval of 3 weeks. Rats below the median level of Test 1 open-arm exploration were unaffected by kindling. Those above the Test 1 median level of open-arm exploration showed reduced exploration following kindling. Kindling did not affect closed-arm entries in the plus-maze in this analysis. However, it was discovered that rats with arm entries below a critical level on Test 1 showed an increase in closed-arm entries following kindling. These findings point out how baseline behaviors can interact with kindling to influence behavioral outcome. Risk assessment was unchanged by kindling in this study, unlike previous reports. Subtle changes in focus location within the medial amygdala may have altered the effects of kindling on risk assessment. The electrodes in this study were in a slightly but significantly different location in the medial amygdala than in previous studies. As in previous studies, risk assessment was measured as frequency and duration of stretch attend postures toward the open arm of the plus maze when the hind quarters were in the closed arms. Risk assessment was taken as a ratio of time spent in the closed arms of the maze. This study, along with others reviewed elsewhere, suggest that a complex set of factors contributes to the effects of kindling on behavior. The fact that previous studies have not taken them into account perhaps explains inconsistencies in the reported behavioral effects of kindling on behavior in rodents.