Evolution is based on genetic variability and subsequent phenotypic selection. Mechanisms that modulate the rate of mutation according to environmental cues, and thus control the balance between genetic stability and flexibility, might provide a distinct evolutionary advantage 1-4 . Stress-induced mutations stimulated by unfavorable environments, and possible mechanisms for their induction, have been described for several organisms 2-4 , but research in this area has mainly focused on microorganisms. We have analyzed the influence of adverse environmental conditions on the genetic stability of the higher plant Arabidopsis thaliana. Here we show that a biotic stress factorattack by the oomycete pathogen Peronospora parasiticacan stimulate somatic recombination in Arabidopsis. The same effect was observed when plant pathogen-defense mechanisms were activated by the chemicals 2,6-dichloroisonicotinic acid (INA) or benzothiadiazole (BTH), or by a mutation (cim3). Together with previous studies of recombination induced by abiotic factors, these findings suggest that increased somatic recombination is a general stress response in plants. The increased genetic flexibility might facilitate evolutionary adaptation of plant populations to stressful environments.In plants, somatic recombination events are an important cause of genetic variability 5 , and, as plants lack a predetermined germ line, may also affect genetic composition of the progeny 6 . A variety of genotoxic factors, such as DNA-damaging chemicals, γ-irradiation and UV-B irradiation 7-10 , stimulate somatic recombination, an important pathway for the repair of DNA lesions in plants 5 . In addition, abiotic stress factors like heat and increased salinity seem to increase somatic recombination 7,8 . It is not known, however, whether biotic stress affects recombination and whether increased recombination might therefore be a general stress response in plants. We examined whether exposure to pathogens, a major biotic stress factor for plants, affects plant genome stability. To measure somatic recombination in whole plants, we used Arabidopsis lines carrying reporter transgenes that allowed visual detection of recombination events (Fig. 1) [8][9][10][11][12] . By testing populations of genetically identical plants grown under different conditions, this quantitative assay allowed us to determine the average recombination frequency per plant at the reporter gene locus and thus measure the influence of pathogen stress on genome stability.We sprayed seedlings of Arabidopsis recombination reporter line 651 (ref. 8) with a spore suspension of P. parasitica isolate EMWA 13 , which resulted in localized necrotic lesions at the points of attempted penetration but no further proliferation of the pathogen, as detected by microscopic observation (data not shown). This suggests the activation of plant defense mechanisms [14][15][16] . There was no macroscopic difference in appearance or growth of infected or uninfected plants. The infected plants showed a significant (P<0.0...