We sampled ca 2500 specimens of Philaenus spumarius (Hemiptera: Aphrophoridae) throughout Corsica without a priori knowledge on the presence of symptoms on plants. We screened 448 specimens for the presence of Xylella fastidiosa (Xf) using qPCR and a custom nested PCR. qPCR appeared versatile and under-estimated the prevalence of Xf. Nested PCR showed that Xf was present in all populations. Molecular results were validated by prediction on the distribution of Xf made from tests conducted on plants, which shows the pertinence of using vectors in risk assessment studies. Xf was detected in tenerals and adults. Thus, P. spumarius could acquire Xf from its host plant, mostly Cistus monspeliensis in Corsica, which may act as reservoir for the next season. This contrasts with other observations and suggests that management strategies may have to be adapted on a case-by-case basis. At least two genetic entities and several variants of Xf not yet identified on plants were present in the insects, which suggests ancient introductions of Xf and a probable underestimation of the current diversity of the strains present in Corsica. Interestingly 6% of the specimens carried two subspecies of Xf. Studies are required to better characterize the strains present in Corsica and to determine how the disease was introduced, spread and why no sign of a potential epidemic was detected earlier. This study shows that, when sensitive enough methods are implemented, spittlebugs (and more specifically P. spumarius for which species distribution modelling shows it could be a good sentinel for Europe) can be used to predict and better assess the exact distribution of Xf. Furthermore, Xf multiply only in their foregut and does not become circulative, which facilitates its detection.

11We sampled ca 2500 specimens of Philaenus spumarius throughout Corsica without a priori 12 on the presence of symptoms on plants. We screened 448 specimens for the presence of 13Xylella fastidiosa (Xf) using qPCR and a custom nested PCR. qPCR appeared versatile and 14 under-estimated the prevalence of Xf. Nested PCR showed that Xf was present in all 15 populations. Molecular results were validated by prediction on the distribution of Xf made 16 from tests conducted on plants, which shows the pertinence of using vectors in risk 17 assessment studies. Xf was detected in tenerals and adults. Thus, P. spumarius could acquire 18Xf from its host plant, mostly Cistus monspeliensis in Corsica, which may act as reservoir for 19 the next season. This contrasts with other observations and suggests that management 20 strategies may have to be adapted on a case-by-case basis. At least two genetic entities and 21 several variants of Xf not yet identified on plants were present in the insects, which suggests 22 ancient introductions of Xf and a probable underestimation of the current diversity of the 23 strains present in Corsica. Interestingly 6% of the specimens carried two subspecies. Studies 24 are wanted to better characterize the strains present in Corsica and know how the disease was 25 introduced, spread and why no sign of a potential epidemic was detected earlier. This study 26shows that, when sensitive enough methods are implemented, insects can be used to predict 27 and better assess the exact distribution of Xf. Insects are indeed easy to collect, Xf multiply 28 only in their foregut and does not become circulative, which facilitates its detection. 29 30