Acute audiogenic seizures are a model of generalized tonic-clonic seizures, induced by high intensity acoustic stimulation in genetically susceptible rodents. The neural substrate are sensory motor brainstem nuclei. Recruitment of forebrain structures takes places upon repetition of acoustically evoked seizures. The term audiogenic kindling means forebrain kindling evoked by repeated brainstem seizures and has been described in several strains of genetically epilepsy-prone rats. Thus, the present work was conducted in order to test the hypothesis that audiogenic kindling recruits the forebrain, which may be behaviorally evaluated and associated with morphological changes as well. The behavioral sequences observed during the development of audiogenic kindling were assessed by neuroethological methods (cluster analysis), with the ETHOMATIC program. Seizure severity indexes (brainstem and limbic seizures) and latencies of wild running and tonic-clonic seizures were measured to quantify seizure evolution. Densitometric analysis of Neo-Timm staining was used for assessing morphological changes associated with audiogenic kindling. In group I, II resistant (R) and 16 susceptible (S) animals were stimulated (120 dB) 21 times, and allowed a 10 day recovery period prior to retesting. In group II, 22 R and 20 S were stimulated 60 times, and allowed a 2 month recovery period prior to retesting. Repetition of the acoustic stimulation in group I and group II susceptible animals led to a progressive and statistically significant attenuation of the behaviors associated with brainstem seizures and a concomitant increased expression of the behaviors associated with limbic seizures. After either a 10 day (group I) or 2 month (group II) recovery period, acoustic stimulation preferentially evoked brainstem-associated behaviors and seizures rather than limbic ones in the audiogenic susceptible animals, although in some animals overlapped brainstem and limbic seizures were detected. Latencies for the wild running and tonic seizures after acoustic stimulation significantly increased during audiogenic kindling for both group I and group II susceptible animals. The quantitative ethological evaluation in both group I and group II, illustrated by flowcharts, showed the evolution of the kindling installation by the presence of limbic seizure clusters, competing in time with the original tonic-clonic clusters. Expression of limbic seizures by group I animals, after acoustic stimulation, was not associated with changes in the mossy fiber Neo-Timm staining pattern of these animals. In group II however, Neo-Timm staining revealed mossy fiber sprouting in the ventral hippocampus (but not in the dorsal), and a significant change in the optical density of amygdaloid nuclei and perirhinal cortex in susceptible animals as compared to resistant ones. In conclusion, audiogenic kindling effectively recruits forebrain structures, responsible for the appearance of limbic seizures. It is possible that the paradigm used in group I was subthreshold for the development of clear-cut synaptic reorganization in the hippocampal mossy fiber system, since the behavioral patterns reverted ten days after the last seizure induction. In group II, however, an increased number of evoked seizures and a more prolonged time after the last chronic seizure showed structural re-arrangements in amygdala, perirhinal cortex and hippocampus, associated with permanence in terms of behavioral data (lack of regression of limbic seizures to control values).