1. The effects of acute and chronic labyrinthectomies on Fos-defined neuronal activity induced by rotation were determined with the use of quantitative image analysis procedures. Unilateral sodium arsanilate labyrinthectomies (UL) were performed either 24 h (acute) or 2 wk (chronic) before exposure to a 90 min, 2-G centripetal acceleration along the interaural axis that stimulated the intact otolith organs. The results obtained from both acute and chronic UL animals subjected to centripetal acceleration were compared with data obtained from nonrotated UL animals and fully intact, normal animals exposed to centripetal acceleration. Such comparisons allowed the definition of functional projections from the otolith organs of one labyrinth to vestibular related and inferior olivary brain stem nuclei in the rat. 2. The effect of the labyrinthectomy on nonrotated animals was first assessed. After acute UL, asymmetric Fos expression was present in the medial and inferior vestibular nuclei, the prepositus hypoglossi (bilaterally), the ipsilateral (with respect to the side of UL) dorsolateral periaqueductal gray, and the contralateral inferior olivary beta subnucleus, as previously described (Kaufman et al., 1992b). Except for minimal labeling in the contralateral prepositus hypoglossi and the dorsolateral periaqueductal gray, the Fos labeling that was present in the brain stem of acute UL animals was absent in chronic UL animals. Thus Fos neuronal activity appears to define a pattern of brain stem activation associated with the initial events that underlie vestibular compensation. 3. In acute UL rats, which were rotated, the contralateral beta subnucleus of the inferior olive had greater labeling (compared with nonrotated UL animals) when the lesion was away from the axis of rotation. In contrast, the ipsilateral beta subnucleus labeled when the lesion was towards the axis of rotation. Fos expression was observed bilaterally in the prepositus hypoglossi when the lesioned side was oriented toward the axis of rotation but was observed only in the contralateral prepositus nucleus when the lesioned side was oriented away from the axis of rotation. Finally, the dorsomedial cell column of the inferior olive (DMCC) was heavily labeled when the lesioned side was oriented towards the axis of rotation but was unlabeled when the lesioned side was oriented away from the axis of rotation. In acute UL nonrotated animals the DMCC was only lightly labeled. All other brain stem nuclear labeling was similar between the acute UL rotated and nonrotated animals. 4. In chronic UL animals which were rotated with the lesioned side oriented towards the axis of rotation, Fos labeling occurred primarily in the following nuclei: the ipsilateral beta and bilateral dorsomedial cell column of the inferior olive; the medial, inferior, and y vestibular nuclei and prepositus hypoglossi predominantly on the contralateral side; and the dorsolateral periaqueductal gray and Darkschewitsch nucleus with greater labeling on the ipsilateral side. There were also differences in the rostrocaudal pattern of labeling in the prepositus hypoglossi and the medial vestibular nucleus. 5. In chronic UL animals that were rotated with the lesioned side away from the axis of rotation, Fos labeling only occurred in the contralateral prepositus hypoglossi and in the dorsolateral periaqueductal gray (stronger ipsilaterally). 6. A comparison of the pattern of labeling in the prepositus hypoglossi and olivary subnuclei in acute and chronic rotated versus nonrotated UL animals suggests that these nuclei support different roles during vestibular compensation and adaptation to hypergravity. Thus the contralateral beta olivary subnucleus is activated by the initial compensation process (acute UL), which appears to increase the sensitivity of this nucleus to hypergravity. The ipsilateral beta subnucleus, on the other hand, is only activated by hypergravity in a specific direction (both acute and chronic UL). 7. In addition to counting the number of Fos labeled neurons, we also quantified Fos expression by using digital image analysis to measure the average intensity of Fos immunolabeling per cell in a given nucleus. Intensity variation among individual neurons in any particular nucleus was approximately ±15% of the mean value within the maximum possible range of values. These differences might be correlated with varying functional roles among the cells within a particular nucleus.