The cerebellar cortex is known to be central to motor learning. However, the question of whether it is a motor memory store or just a crucial element of the circuitry involved in encoding such memory has yet to be resolved. It is also not clear how the deep nuclei, the other major cerebellar component, function in motor learning. For the last quarter of a century much of the research into these problems has used eyeblink conditioning as a model to investigate these problems. Most recently, in an attempt to identify the site or sites of memory encoding and storage, pharmacological intervention during this classical conditioning procedure has been used to reversibly inactivate restricted elements of the cerebellum and associated brain structures. The delivery of such inactivating infusions, and the means by which their spread is assessed, are becoming increasingly accurate. Reversible inactivation has enabled the development of a progressively more detailed picture of the circuitry essential for eyeblink conditioning. Such studies have not, however, conclusively revealed the site, time-course or mechanism of memory storage. It has recently become clear that crucial elements of the conditioned eyeblink circuitry involve a dynamically connected cerebellar/brainstem loop. This feature of eyeblink circuitry means that reversible inactivation of one component may have consequences for activity in the other components. Conclusive identification of a locus of memory storage has therefore been impossible. This thesis details the results of a series of experiments that address this problem by limiting experimental manipulations to the processes commonly known as consolidation, and thereby the mechanism by which memories are stored long-term, without interfering with the initial process of memory encoding.