ACTA PALaEONTOLOGICA ROMANIAE V. 7 (2011), P. 335-342 335 REASSESSMENT OF THE SPATIAL EXTENT OF THE MIDDLE MEMBER, DENSUŞ-CIULA FORMATION (MAASTRICHTIAN), HAŢEG BASIN, ROMANIA ŞTEFAN VASILE 1 , ZOLTÁN CSIKI 1 , DAN GRIGORESCU 1 Abstract. The fossil-bearing Maastrichtian continental deposits of the Haţeg Basin have been assigned to different lithostratigraphic units and subunits based on their lithology and fossil content; among these, the vertebrate fossil- and volcanoclast-bearing beds from the north–western part of the basin were referred to the Middle Member of the Densuş- Ciula Formation. Continuing fieldwork in this area allows a reconsideration of the spatial extent of the previously- separated units. Based on their lithology and fossil remains, deposits cropping out near General Berthelot locality can be also included into the Middle Member of the Densuş-Ciula Formation. Accordingly, the eastern boundary of this subunit has to be moved eastwards than it was previously considered. Keywords: Haţeg Basin; Maastrichtian; Densuş-Ciula Formation; eggshells; microvertebrates. 1 Laboratory of Paleontology, Faculty of Geology and Geophysics, University of Bucharest; 1 N. Bălcescu Ave., 010041, Bucharest, Romania; e-mail: yokozuna_uz@yahoo.com INTRODUCTION The vertebrate-bearing continental deposits of Haţeg Basin, well-known for the remains of dwarf dinosaurs they yielded, cover the north–western and central–eastern parts of the basin. While in the central–eastern part the deposits occur extremely patchy, along several of the main river courses (see Codrea et al., 2009 for a review), at the north–western margin these deposits are distributed as an elongated belt extending eastward from Ciula Mică and Densuş localities, towards the town of Haţeg. Nevertheless, the outcrops here are also discontinuous, due to the extensive cover of soil and vegetation in this hilly area. Due to this setting, as well as to the absence of traceable marker-beds in the dominantly siliciclastic sequence, dating of the deposits and establishment of a well-defined lithostratigraphic framework had proven to be difficult. One of the main controversial points concerns the spatial extent of the Maastrichtian deposits and the position of the Cretaceous/Tertiary boundary (K-TB), as it was often assumed (see below) that continental deposition in this area was continuous into the Paleogene. Lacking other means of dating, the presence and, respectively, absence of clearly-identifiable Cretaceous fossils was used as the main argument to draw the possible position of the K-TB. This limit is considered to mark the boundary between two subunits of the lithostratigraphic unit that groups the continental deposits from the north-western part of the Haţeg Basin. Recent discovery of fossil remains, including dinosaurian eggshells, suggests that the area covered by Maastrichtian continental beds extends probably farther eastward along the northern border of the basin than previously considered. LITHOSTRATIGRAPHY OF THE MAASTRICHTIAN CONTINENTAL DEPOSITS – A HISTORICAL OVERVIEW Nopcsa (1905) was the first to describe the Cretaceous deposits of the Haţeg Basin in detail, trying to fit the dinosaur-bearing deposits into the general geological framework. Nopcsa outlined the lithological features of the fossiliferous deposits, which he considered to be Danian in age, the Danian representing at that time the last stage of the Cretaceous. Nopcsa considered most of the coarser-grained sediments (sandstones and conglomerates with thin silty mudstone intercalations) cropping out at Sânpetru, Sântămăria-Orlea, Nălaţ-Vad, Haţeg, Crăguiş, Fărcădinul de Jos (today General Berthelot) and Fărcădinul de Sus (today Fărcădin) to represent the same type of freshwater deposits, generically-named “the Sânpetru Sandstone”. In the area between Densuş, Ciula Mică, Răchitova and Stei, Nopcsa described tuff- rich deposits, associated with porphyric rock intercalations, considering them to be a local facies variation of the Sânpetru Sandstone. Nopcsa (1905) also noted the constant increase in the size and the quantity of the volcanogenic clasts proceeding from Sânpetru towards Densuş, a good hint to where the source supplying this volcanogenic material was located. Laufer (1925) recognized the same distinction between the tuffaceous and the fluvio-lacustrine facies of the “Danian” deposits, while also mentioning leaf impressions from Densuş area. However, he raised some doubts about the Cretaceous age of the conglomerates cropping out around Crăguiş, considering them to represent a “problematic Paleogene”, based on the lack of fossil remains in this area. To the contrary, Mamulea (1953) assigned the continental deposits lying east of a Ciula Mică–Densuş line to the Paleogene–Lower Miocene (Aquitanian), mainly based on a perceived colour difference (i.e., dominantly gray deposits in the Răchitova-Ştei-Densuş- Ciula perimeter, as opposed to dominantly red deposits east of it) and the purported presence of in situ dinosaur bones in this area. Although not expressed straightforward, the bones reported previously by Nopcsa (e.g., 1905) from the Vălioara area were considered as being reworked from the underlying tuffaceous deposits of the “Danian”. This line of reasoning was followed subsequently by other workers as well (e.g., Iliescu et al., 1972, unpublished report); accordingly, the Geological Map of Romania also shows these deposits as being of Paleogene age (Codarcea et al., 1968).