[en] The isotopic composition of the Callovo-Oxfordian argilites of Bure site (Eastern Paris basin) has been used to determine the nature of the pore water and of the water-rock interactions. A specific degassing apparatus has been implemented for the extraction and purification of gases from argilite core samples (CO₂, water vapor, methane, ethane and propane). ¹⁸O and ¹³C contents have been determined by mass spectroscopy (IRMS and GC-C-IRMS). Results indicate the marine inorganic origin of the CO₂ and the deep extra-formation origin of the desorbed hydrocarbons. They confirm the meteoric origin of the interstitial water and the very limited bacterial activity. The main advantage of this method is that no pore water extraction is needed, which protects the clay-water system from any irreversible disturbance. (J.S.)

[en] Shells extracted from loess deposits at Achenheim (Rhine Graben) have been investigated for AMS ¹⁴C and for U-Th dating by the isochron method on single shells. The two chronometers provide results that are not in agreement : the obtained ¹⁴C date is 38 ka B.P. In contrast, U-Th results provide much older ages. The ²³⁸U/²³²Th - ²³⁰Th/²³²Th isochron diagram gives an age of 69 kyr, whereas a slightly younger age of 64 kyr is derived from the other, ²³⁴U/²³²Th - ²³⁰Th/²³²Th, diagram. ²³⁴U/²³⁸U ratios have not been found to vary from a shell to another: the average value is 1.199, slightly higher than radioactive equilibrium. For both isochrones, the Y-axis intercept is slightly negative, which indicates that there is no evidence for a significant detrital component. Thorium 232 that is present in the shells is probably incorporated in the material. The negative intercepts suggest that the two ages have been probably 'aged' 'resulting from a slight mobilization of uranium. By constraining the Y-axis intercepts at 0, ages that are in good agreement from one to another, can be calculated at 60.9 kyr and 60.4 kyr, respectively. In both cases, the statistical quality of the alignment is only slightly deteriorated. Since U-Th ages are in agreement with stratigraphic data, in contrast to the ¹⁴C date, one may believe that they date the end of early diagenesis, with uranium trapping. In contrast, the AMS ¹⁴C date is either thought to reflect analytical limitations of AMS ¹⁴C dating for such type of material, or to date a younger geologic event. In this latter case, a possible hypothesis is that this event has triggered the slight opening of the ²³⁸U - ²³⁴U system. In any case, this study demonstrates once again that surface formations often tell such complex stories that only one single dating method cannot usually he sufficient to describe precisely these histories. Instead, informations derived from different independent chronometers have to be compared. (authors)