Abstract The present paper is aimed at a structural and semantic analysis of dynamic and stative verbs in a linguistic reflection of spatial relations. The investigation is based on the data of dynamic verbs referring to motion/movement- the most widespread and salient kind of motion -and of stative verbs referring to immobility/location. Specifically, I will use dynamic verbs that refer to both motion/movement proper (motion per se) and to transfer (to take): ul-a ‘to go,’ purin-ua ‘to fly,’ nčur-ua ‘to swim,’ rula-a ‘to run,’ and xox-ua ‘to crawl.’ Parallel stative verbs will be presented as forms referring to standard location/state: -rina ‘to be, to stand,’ -xvena ‘to sit,’ -’n’ǯira ‘to lie,’ - ʒvena ‘to be laid,’ and dgvena /dguma ‘to be laid’ (for inanimate objects), etc. I will also use data from morpho-syntactic analyses of dynamic and stative forms. Megrelian dynamic and stative verb roots convey types of motion/immobility, however, they are essentially neutral in terms of any locative reference. This kind of information is rendered by adverbial derivational formants included in the verb stem – preverbs which are taken on by verb roots referring to motion/movement to derive new, semantically distinct, verb stems. Owing to this, it is necessary to discuss in detail the structure and usage of the preverbs as a basic means for the encoding of spatial relations within a neutral root.With respect to the issue in question, the following was established: a) types of directions and orientation of motion encoded in the aforementioned verb forms; b) peculiarities of spatial location of motion/immobility, and; c) semantic features of dynamic and stative forms.