1) For each number q such that 0⩽q<1, the sequence (qn) ("geometric progression of ratio q") is summable in R, since
sn=1−q1−qn+1⩽1−q1 the sum of this sequence is
limn→∞sn=1−q1.
2) Let a and b be two numbers such that 0⩽a<1 and 0⩽b<1; then the family (ambn)(m,n)∈N×N is summable in R. For each finite subset of N×N is contained in a subset of the form [0,p]×[0,p], and we have
m=0∑pn=0∑pambn=(m=0∑pam)(n=0∑pbn)=1−a1−ap+1⋅1−b1−bp+1⩽(1−a)(1−b)1.
3) For each integer p>1, the sequence (n−p) (n>0) is summable, since
s2n+1−s2n=k=1∑2n(2n+k)−p<2n⋅(2n)−p and therefore, adding these inequalities together,
s2n<1−21−p1.
4) The sequence (1/n) (n>0) is not summable in R, since
s2n+1−s2n=k=1∑2n2n+k1>2n+12n=21 and therefore, adding these inequalities together,
s2n>n/2 so that the criterion of Theorem 1 is not satisfied.
5) Let (In) be a sequence of mutually disjoint non-empty open intervals, all contained in an interval of finite length l. The sum of the lengths of a finite number of intervals of this family is ≤l (§1, no. 5) and therefore the family of lengths of the In is summable in R, and its sum is ≤l.
Bourbaki puts the proof directly after the statement, so what is above is the statement and its proof together.
Nothing.