Bourbaki

Proposition 4

Tag 0CCE, permanent. Cite it as [Bourbaki, Tag 0CCE].

General Topology, chapter X, § 2. Equicontinuous sets

Let X,YX, Y be two uniform spaces and let HH be a set of uniformly continuous mappings of XX into YY. Let X^,Y^\hat{X}, \hat{Y} be the Hausdorff completions of X,YX, Y respectively, and let H~\tilde{H} denote the set of mappings u^:X^→Y^\hat{u} : \hat{X} \to \hat{Y} as uu runs through HH (Chapter II, § 3, no. 7, Proposition 15). Then HH is uniformly equicontinuous if and only if H~\tilde{H} is uniformly equicontinuous.

We recall that the diagram

X→uYi↓j↓X^→u^Y^ \begin{array}{ccc} X & \xrightarrow{u} & Y \\ i \downarrow & & j \downarrow \\ \hat{X} & \xrightarrow{\hat{u}} & \hat{Y} \end{array}

is commutative, where ii and jj are the canonical mappings; moreover, the uniformity of XX (resp. YY) is the inverse image under ii (resp. jj) of that of X^\hat{X} (resp. Y^\hat{Y}). Hence HH is uniformly equicontinuous if and only if its image under the mapping u→j∘uu \to j \circ u is uniformly equicontinuous (Proposition 3), and we may already restrict ourselves to the case where YY is Hausdorff and complete; moreover, if H~\tilde{H} is uniformly equicontinuous then so is HH, because it is the image of H~\tilde{H} under the mapping u^→u^∘i\hat{u} \to \hat{u} \circ i; thus it remains to prove the converse when Y=Y^Y = \hat{Y}. Let VV be a closed entourage of YY; by hypothesis there is an entourage UU of XX such that the relations (x,x′)∈U(x, x') \in U and u∈Hu \in H imply that (u(x),u(x′))∈V(u(x), u(x')) \in V. Now, if U′U' is the image of UU under i×ii \times i, the closure U′‾\overline{U'} of U′U' in X^×X^\hat{X} \times \hat{X} is an entourage of X^\hat{X} (Chapter II, § 3, no. 7, Proposition 12); the hypothesis implies that, whenever (z,z′)∈U′(z, z') \in U' and u∈Hu \in H, we have (u^(z),u^(z′))∈V(\hat{u}(z), \hat{u}(z')) \in V. Since VV is closed and u^\hat{u} is continuous, we have also (u^(t),u^(t′))∈V(\hat{u}(t), \hat{u}(t')) \in V for all (t,t′)∈U′‾(t, t') \in \overline{U'} and all u∈Hu \in H; this completes the proof.

Bourbaki puts the proof directly after the statement, so what is above is the statement and its proof together.

Cited by

Cites

Transcribed from the printed volume. Not translated.