An experimental design is described to test the hypothesis that ice segregation near the top of permafrost and in the lower part of the active layer can brecciate frost-susceptible bedrock. Seasonal temperature cycles in the active layer and the top of permafrost were simulated to a first approximation within a block of chalk measuring 0Á30 m  0Á31 m wide and 0Á33 m high and insulated around the sides to minimize lateral heat transfer. The block, moistened first by capillary rise, was initially frozen from the surface downwards, simulating permafrost aggradation. Thereafter, the upper half of the block was cycled above and below 0 C (simulating seasonal freezing and thawing of the active layer) while the lower half remained below 0 C (simulating permafrost). During thaw cycles water was supplied to the surface and base of the simulated active layer. Temperature, unfrozen water content, surface frost heave and porewater pressure within the chalk were monitored during the experiment.Preliminary data from three freeze±thaw cycles indicate that: (1) both downward and upward freezing occurred during freeze-back of the simulated active layer; (2) 78 per cent of the surface frost heave (total = 4Á8 mm) took place during the thaw periods, when the active layer was approaching or at its maximum depth; and (3) negative porewater pressure was recorded in some instances during freeze-back, suggesting the occurrence of cryogenic suction. The thaw-related frost heave suggests that ice segregation within the chalk occurred beneath the descending thawing front, in the still frozen part of the simulated active layer and the uppermost permafrost, and marked the commencement of rock brecciation.

Physical modelling has been developed in order to simulate the effects of periglacial erosion processes on the degradation of slopes and scarps. Data from 41 experimental freeze-thaw cycles are presented. They attest to the efficiency of periglacial processes that control both erosion and changes in scarp morphology: (i) cryoexpulsion leads to an increase of scarp surface roughness and modifies significantly the internal structure of the active layer; (ii) combined effects of frost creep and gelifluction lead to slow and gradual downslope displacements of the active layer (0·3 cm/cycle); (iii) debris flows are associated with the most significant changes in scarp morphology and are responsible for the highest rate of scarp erosion; (iv) quantification of the erosion rate gives values close to 1 cm 3 cm − − − − −2 for 41 freeze-thaw cycles. These experimental results are consistent with field data acquired along the La Hague fault scarp (Normandy, France) where an erosion rate of 4·6 ± ± ± ± ± 1 m 3 m − − − − −2 per glacial stage has been computed from the volume of natural slope deposits stored during the Weichselian glacial stage. These results show that moist periglacial erosion processes could lead to an underestimation of Plio-Quaternary deformation in the mid-latitudes. CopyrightEvidence of erosion and decay of a scarp can be found along the la Hague Fault Zone ( Figure 1A). In this metamorphic basement area, detailed analyses of the scarp morphology and slope deposits have allowed identification of periglacial processes (Elhaï, 1963;Watson and Watson, 1970;Lautridou, 1985;Font, 2002).Cryoclastic processes that occurred during periglacial periods have contributed to the fracturing of metamorphic rocks near the surface, leading to a widespread cover of cryoclasts.Downslope sliding of cryoclasts is indicated by footslope deposits several tens of metres thick (Elhaï, 1963;Watson and Watson, 1970). Slope deposits are roughly stratified and made of coarse angular metamorphic and sedimentary rocks, which come from the Palaeozoic units exposed in the slope above (Watson and Watson, 1970). The slope deposits lie on a raised beach and show distinctive geometrical and lithological characteristics: (i) along the scarp, variations of the lithology follow variations of the bedrock on the back slope; (ii) the stratification is characterized by Fault scarp degradation 1733 Figure 6. (A) Displacement profiles of lateral tile columns indicating the continuous deformation of the active layer during thawing. One can note that higher values of displacement are measured in the lower part of the scarp. (B) Cumulated displacements of the top tiles from central columns. (C) Quantification of slope deposits translated during the continuous deformation of the active layer (lateral downstream tile column).Figure 8. Progressive changes in scarp morphology: H, heaving; P, packing; He, headward erosion; Ab, ablation due to debris flows.Figure 9. Correlation between slope variations and volume changes on the basal steeper slope: 1, progre...

(Llandoverian, Canada) were studied using image analysis. After digitalization of the objects, shape parameters were calculated. The boundary of each fossil was then traced by a vector centred at the centroid for Fast Fourier Transform (FFT). Results of the two methods were used as variables in a hierarchical cluster analysis in order to group the samples. These results show that Chitinozoa can be significantly classified in terms of taxa using independent shape parameters obtained by image analysis.

La confrontation entre les données paléoenvironnementales et les résultats d'analyse tracéologique d'artefacts en silex provenant de sites paléolithiques a amené à soupçonner le rôle de processus liés au gel du sol dans l'altération de la surface des silex ayant été soumis à des ambiances périglaciaires. Afin de vérifier cette hypothèse et d'examiner les modalités des phénomènes post-dépositionnels incriminés, des expériences de laboratoire ont été menées entre 2004 et 2007 au Centre de Géomorphologie du CNRS à Caen, où des modèles contenant des artefacts en silex ont été soumis à des cycles gel/dégel. Les premiers résultats, qui demandent à être confirmés et précisés par une étude tracéologique, montrent des modifications microscopiques de l'état de surface des silex enfouis dans le limon.