An area inside the Paulista Peripherical Depression (Sao Paulo, Brasil) contains a sandy Cenozoic deposit overlying a silty Permian deposit of the Iratí formation (Pi). This sequence is abruptly interrupted by a diabase sill overlain by a red clay which extends westward towards the Piracicaba River. These materials are separated by two stonelines made up of rounded quartzite pebbles. Two hypothesis were considered to explain the local landscape evolution: a local "in situ" evolution and a Quaternary tecto-morphoclimatic evolution. The sandy and red clay deposits could be considered as "in situ" formations derived directly from the local parent material, a sandy Triassic-Jurassic deposit of the Piramboia formation (TrJP) and the diabase, respectively. On the other hand, the local Paleo-Mesozoic stratigraphy shows that, if this hypothesis is considered, a whole stratigraphical layer corresponding to a late Permian clayey deposit of the Corumbatai formation (Pc) would be missing. This is possible in areas were erosive discordances are identified, which is not the case of this study area. Climatic changes during the Quaternary, leading to alternating periods of erosion and pedogenesis, were considered as the best hypothesis to explain this area's landscape evolution. In this way, two depositioal events were considered. The first event was the deposition of a fine sandy sediment in the form of stepped alveoluses during a first erosional period. Excavations of the Paulista Peripherical Depression exposed a sill barrier that stopped these sediments that were transported from higher landscape positions. The second depositional event, leading to deposition of the red clay, was probably due to an incision of the first event's pediplane followed by a second pedimentation process during a later erosional period. The origin of the stonelines was related to these depositional events due to the fact that they separate materials that present litohological discontinuities and the high degree of pebble roundness would result from the high friction these materials suffered during transport rather than "in situ" geochemical formation processes.

These stonelines could be interpreted as been formed by geochemical processes and have an authoctonous origin. The first hypothesis confers an authoctonous origin to both stonelines where the rounded pebbles could be formed by geochemical processes. This is possible but the very high degree.

At least two post-Permian depositional events were identified as marked by stonelines; one of rounded pebbles separating the fine sandy Cenozoic layer from the basal silty Permian base deposit and a second stoneline separating the diabase and part of the Irati formation saprolites from the red clay.

Lithological discontinuities in four profiles on each of the parent materials were identified through the granulometry of sand fractions, elemental analyses of Zr and Ti, semi-quantitative determinations of quartz and micromorphological observations. The first post-Permian depositional event is indicated by an abrupt lateral transition between the diabase and the sandy sediments. Significant differences in the granulometry of sands, the quantities of Zr and Ti, the Ti/Zr and Zr/quartz ratios and the degree of roundness of the quartz grains were observed between the materials above and below the second stoneline, indicating its allochthonous origin and the presence of a second depositional event.