Let M be a commutative monoid (resp. group) and f a mapping of X into M. There exists one and only one homomorphism of N(X) (resp. Z(X)) into M such that g(δx)=f(x) for all x∈X. If M is written additively, then g(α)=∑x∈Xα(x)⋅f(x) for all α in N(X) (resp. Z(X)).
Let g be a homomorphism of N(X) (resp. Z(X)) into M such that g(δx)=f(x) for all x∈X. For all α in N(X) (resp. Z(X)), it follows from (12) that
g(α)=x∈X∑α(x)⋅g(δx)=x∈X∑α(x)⋅f(x), whence the uniqueness of
g.
For all α in N(X) (resp. Z(X)) we write g(α)=∑x∈Xα(x)⋅f(x). Then obviously g(0)=0; for α,β in N(X) (resp. Z(X)),
g(α+β)=x∈X∑(α(x)+β(x))⋅f(x)=x∈X∑[α(x)⋅f(x)+β(x)⋅f(x)]=x∈X∑α(x)⋅f(x)+x∈X∑β(x)⋅f(x)=g(α)+g(β) and hence
g is a homomorphism of
N(X) (resp.
Z(X)) into M. Also, for
y in
X,
g(δy)=x∈X∑δy(x)⋅f(x); now
δy(x)⋅f(x)=0 for
x=y and
δy(y)⋅f(y)=f(y), whence
g(δy)=f(y).
Let u:X→Y be a mapping. By Proposition 10 there exists one and only one homomorphism of Z(X) into Z(Y) which maps δx to δu(x) for all x∈X. It is denoted by Z(u); it is immediately seen that it maps α∈Z(X) to the element β∈Z(Y) defined by
β(y)=x∈u−1(y)∑α(x). As in the case of magmas (no. 1), the formula
Z(v∘u)=Z(v)∘Z(y) is established for every mapping
v:Y→Z; it is also shown that
Z(u) is injective (resp. surjective, bijective) if
u is.
Let S be a subset of X; if i is the injection of S into X, the mapping f=Z(i) is an isomorphism of Z(S) onto the subgroup H of Z(X) generated by the elements δs for s∈S. By (15),
(f(α))(x)={α(x)0if x∈Sif x∈X−S
and therefore H is the set of elements of Z(X) of support contained in S. Henceforth Z(S) will be identified with H by means of f.
Formula (15) shows that the restriction of Z(u) to N(X) induces a homomorphism N(u) of N(X) into N(Y). Then N(v∘u)=N(v)∘N(u) for every mapping v:Y→Z; further, N(u) is injective (resp. surjective, bijective) if u is. If S is a subset of X,
N(S)=Z(S)∩N(X).
Bourbaki puts the proof directly after the statement, so what is above is the statement and its proof together.
Nothing.