salut bon la fonction h=f-g est bien décroissante sur [a;b],donc pour deux elements de [a;b] le plus grand aura la plus petite image par h,or tout element x de [a;b] est superieur à a(exepté le a bien sur)donc pour tout x de ]a;b] on aura:h(x)<h(a) or on a h(a)=f(a)-g(a)=0 d'où on obtient h(x)<0 c'e...