Proposition 3
Tag 0BOO, permanent. Cite it as [Bourbaki, Tag 0BOO].
General Topology, chapter III, § 7. Inverse limits of topological groups and rings
Suppose that is Hausdorff and that the are compact and satisfy (AP).
a) *Let be a closed subgroup of ; then, for each , the subgroup of is closed, and the isomorphism of onto gives by restriction an isomorphism of onto . If also is normal in , then is normal in for each , and by passing to the quotients, induces an isomorphism of onto .
b) Conversely, for each let be a closed subgroup of , such that whenever . Then there is a unique closed subgroup of such that for each ; and if in addition is normal in for each , then is normal in .
a) Since is compact, is closed in ( 4, no. 1, Proposition 1, Corollary 1) and therefore is closed in . Since identifies the topological groups and , and since may be identified with a (topological) subgroup of , identifies the subgroup of with , and to prove the first assertion it is enough to remark that by Lemma 1. On the other hand, if is normal, then for each the mapping induced by is a surjective strict morphism ( 2, no. 8, Remark 3), whose kernel is the compact normal subgroup of , the canonical image of the compact subgroup of . Since the subgroups of satisfy condition (AP) (§ 2, no. 8, Proposition 24), and since is Hausdorff, the last assertion of a) is a consequence of Proposition 2.
b) Let be the restriction of to (). Then is an inverse system of topological groups, whose inverse limit can be identified with the subgroup of . By hypothesis, is surjective and its kernel is the compact subgroup of ; consequently (Chapter I, § 9, no. 6, Corollary 1 to Proposition 8) we have for all . If is another closed subgroup of such that for all , then , whence (Lemma 1) . Finally, the last assertion of b) follows from the formula , since the are now normal subgroups of .
Bourbaki puts the proof directly after the statement, so what is above is the statement and its proof together.