The narrative art depicted on various sema stones (Buddhist boundary markers) from northeast Thailand dating from the seventh to eleventh centuries CE and generally considered to be part of the Dvāravatī art style or culture1 have long been admired for both the quality of their carvings and the skilful rendering of their subject matters. 2 The subjects consist primarily of jātaka tales and scenes from the life of the Buddha. Similarly, a group of semas from the Kalyāṇī Sīmā in Thaton, lower Myanmar, also depict narrative scenes from jātakas and have therefore invited comparison from a number of scholars. Piriya Krairiksh (1974: 63) in particular stresses this point, arguing that a group of Mons may have fled northeast Thailand as the Khmers began to take over their territories. 3 The natural place for them to seek refuge, he argues, was Thaton, being as it was, one of the main centres of “Mon Buddhism.” Furthermore, he states that as there does not seem to be any “Burmese” forerunners to the Thaton semas, it is plausible that the Mons of northeast Thailand brought this tradition with them. This hypothesis, however, has a number of problems. Upon close investigation of the semas from both northeast Thailand and Thaton, it becomes apparent that while the content of these narrative scenes are similar and at times identical, the style, composition and morphology of the semas from both locations differ considerably. What follows is a comparison between the sema stones of lower Myanmar and northeast Thailand from the seventh to eleventh centuries. The origins and use of semas shall first be briefly discussed, followed by an overview of the earliest evidence for these objects in both Myanmar and Thailand. Excavations at Vesāli in Rakhine (Arakan) state in the west of Myanmar have led to the discovery of a number of semas from circa fifth to seventh centuries CE, while the so-called “Buddhist megaliths”(Luce 1985: 130-131, pls 12, 13, 15) at the site of Śrīkṣetra, while not clearly identifiable as sema, do appear to have fulfilled a somewhat similar function in defining ritual space and provide evidence for a well-established stone carving tradition. The narrative art on the semas from northeast Thailand, in particular from the sites of Mueang Fa Daet and Ban Nong Chat, 4 shall then be compared with that from Thaton. In doing so, it will be shown that whilst there are many similarities in terms of function and narrative content, there are also various differences in terms of how the narrative art is depicted, the form of the semas themselves and certain stylistic renderings. Overall, it will be shown that the “Mon refugee” hypothesis is untenable. This raises a further question, that is, whether or not the sema tradition of northeast Thailand can be seen as a model or forerunner for those in lower Myanmar. The evidence discussed