Then there exists one and only one homomorphism u1:Ker(f)→Ker(g), and one and only one homomorphism u2:Coker(f)→Coker(g), such that the diagrams
Ker(f)↓iMu1uKer(g)↓jN(8)
and
M′p↓Coker (f)u′u2N′q↓Coker (g)
are commutative, i and j denoting the canonical injections, p and q the canonical surjections.
Indeed, if x∈Ker (f), we have f(xˉ)=0 and g(u(x))=u′(f(x))=0, hence u(x)∈Ker (g), and the existence and uniqueness of u1 are then immediate. Similarly, we have
u′(f(M))=g(u(M))⊂g(N),
hence u′ gives by passing to the quotients a homomorphism
u2:Coker (f)→Coker (g),
which is the only homomorphism for which (9) is commutative.
Let us now start from a commutative diagram (4) of A-modules; by virtue of Lemma 1, there corresponds to it a commutative diagram
Ker (f)i↓Mf↓M′p↓Coker (f)u1uu′u2Ker (g)j↓Ng↓N′q↓Coker (g)v1vv′v2Ker (h)k↓Ph↓P′r↓Coker (h)
where i,j,k are the canonical injections, p,q,r the canonical surjections, u1,u2 (resp. v1,v2) the homomorphisms deduced from u,u′ (resp. v,v′) by Lemma 1.
Bourbaki puts the proof directly after the statement, so what is above is the statement and its proof together.