In August 2006, bluetongue virus (BTV) was detected in the Netherlands, Belgium, western Germany, Luxembourg and northern France for the first time. Consequently, a longitudinal entomological study was conducted in the affected region of northern France (Ardennes) throughout the autumn of 2006. Data on the spatio-temporal distribution of Culicoides (Diptera: Ceratopogonidae) associated with livestock were collected and an attempt was made to identify the vector(s) involved in BTV transmission by means of virus detection in wild-caught biting midges. Weekly sampling using standardized Onderstepoort-type blacklight traps were performed simultaneously both outdoors and indoors in one BTV-free and three BTV-affected farms between September and December 2006. Culicoides were sorted according to farm, location (outdoors vs. indoors), time point (in weeks), species and physiological stage. BTV detection was conducted by RT-PCR on monospecific pools of non-bloodfed parous female Culicoides. The principal results showed: (i) the absence of the Mediterranean vector, C. imicola, (ii) the relatively low abundance of C. dewulfi and C. pulicaris, (iii) the widespread occurrence and abundance of C. obsoletus/C. scoticus with longevity and behaviour compatible with BTV transmission, and (iv) all Culicoides pools tested for BTV were negative. In France, the very low levels of BTV-8 circulation were probably due to the limited introduction of the virus from affected neighbouring countries, and not due to the absence of local vector populations. A key finding has been the substantiation, for the first time, that Culicoides, and particularly the potential vectors C. obsoletus/C. scoticus and C. dewulfi, can be active at night inside livestock buildings and not only outside, as originally believed. The endophagic tendencies of members of the Obsoletus group are discussed in light of the prolonged period of BTV transmission during the autumn of 2006 and the risk of BTV overwintering and resurgence in the spring of 2007. Overall, there is an urgent need to improve our knowledge on the ecology of local Culicoides species before any clear, effective and reliable recommendations can be provided to the veterinary authorities in terms of prevention and control.