Because erodibility quantifies the easiness of the soil to be eroded, erodibility is a key concept to predict soil erosion. Nevertheless, erodibility does not have a unique description (Bryan, 2000). In this paper, we review recent developments in the assessment of erodibility for interrill areas using structural stability. Based on this, future research directions are underlined. Structural stability as measured by Le Bissonnais (1996) is a convenient method to assess soil erodibility: it is fast, reproducible and low-cost. It assesses the sensitivity of soil aggregates to three basic disaggregation mechanisms (slaking, differential swelling and mechanical breakdown). Le Bissonnais et al. (2007) showed a high correlation (90%) between structural stability and soil loss for Mediterranean vineyards. It confirms structural stability as a good estimator of soil erodibility. Moreover, Leguédois & Le Bissonnais (2004) and Legout et al. (2005) showed that aggregate size distributions of structural stability tests matched the size distributions of eroded aggregates under rainfall simulations. It means stability tests could be used not only to estimate the erodibility but also the size fractions that are available to erosion processes. We propose a pedotransfer function predicting structural stability, based on a database including structural stability and results of usual soil analyses (texture, carbon content, etc.) for French soils. This allows for an assessment of soil erodibility based on soil properties. Annabi (2005) and Cosentino et al. (2006) showed that structural stability of a given soil could vary within a year, but the factors controlling such changes remain to be specified. Darboux & Le Bissonnais (2007) showed that surface crusts did not have the same stability than the plough layer. From these recent works, erodibility in interrill areas appears as a dynamic parameter. There is no unique value of erodibility but a range of values depending on the crusting stage, the climate and the biological activity. Future research effort should look at the causes of short-term changes of structural stability. This will lead to a better prediction of interrill erodibility.