Taphonomic aspects of the paleoherpetofauna of the Santa Maria and Caturrita Formations (Triassic of southern Brazil) are for the first time specifically dealt with. A total of 1096 specimens, from articulated skeletons to reworked and isolated bone fragments were studied. The four major reptilian groups of the paleoherpetofauna (cynodonts, dicynodonts thecodonts and rhynchosaurs) from three distinct local faunas. Preservation of bones and their spatial distribution were analysed, allowing the establishment of four taphonomic classes: (I) relatively articulated skeletons; (II) articulated bones; (III) isolated well-preserved bones; (IV) isolated fragmented bones. The numerical sequence of classes reflects the increasing disorganization of the skeletal material; the higher the class number, the more complex is the taphonomic story to beinvestigated. Class I is composed of transported and rapidly buried carcasses, under the influence of catastrophic events; on the other extreme, fragmented pieces from Class IV represent elements exposed for a long time to the influence of weathering, trampling and scavenger activity. The study of selective transport ( Voorhies's Groups) of bone elements from dicynodonts and rhynchosaurs (the most representative groups of the Triassic herpetofauna) revealed that hydraulic selection by moving water was not important as a Factor of disarticulation and scattering, although it did occur at some sites. The computation of the fossil recovery rate showed that, in a general way, the preservation is low and that there is a bias favouring skulls (over-represented in the collection) and vertebrae (under-represented). The analysis of size classes of the Triassic reptilian remains (taken as analogues to age classes) showed a distribution very similar to that exhibited by living communities, indicating that small (= young), intermediate and large (= old) individuals were buried together in proportions approximating the distribution pattern in living populations. This is taken as an evidence of catastrophic rather than attritional death. By bringing together all the stratigraphic, paleontological and taphonomic data, the reconstruction of the taphonomic history was possible. This model postulates a multi-episodic history of catastrophic death and rapid burial of complete carcasses caused by periodic floods, which alternated with periods of exposure of bone elements and little or no burial, indicating that large amounts of water in the depositional system alternated with periods of low water level or even dryness.